Shortly before my last research trip I learned that, in celebration of EDF's 50th anniversary, limited portions of the utility's official archives would finally open to researchers. Upon arriving in Paris, I wasted no time calling for an appointment. A young woman was assigned to my case. I described my project and the documents I wanted to see.

She listened quietly for a while, making notes and ominous "tut-tut" sounds. After I finished talking, she explained that she would need several weeks to go through the inventories and figure out whether she had what I needed. It was out of the question for me to examine the inventories myself, and of course I must understand that she had many other things to do. She was pessimistic about the outcome of her search, though she did recall seeing something that might have been a training manual for power plant operators. That had potential, I replied; did she remember the date of the manual? Not exactly, she said, but it was definitely a historical document: it came from the late 1980s. Quelling a mounting (and familiar) feeling of frustration, I noted patiently that this could only be of limited interest since (as I had already explained) my study ended in 1969. 1969! -- she exclaimed -- but there was no nuclear program before 1970! How could one write about something that didn't exist?

Unable to restrain myself, I burst out laughing. Not surprisingly, this did nothing to improve the woman's good will. She was extremely dubious of my assurances that yes, EDF really had been involved in producing nuclear power since the mid 1950s. Only the guerre des filières -- the battle of the systems -- rang a bell for her. Perhaps, she conceded, the current "nuclear park" did have a predecessor. She certainly knew nothing about it.

I eventually realized that this woman had been hired very recently and (like most of the archive staff, as it turned out) was not a professional archivist. In fairness, her ignorance of EDF's pre-1970 nuclear activities turned out to be highly atypical for people working within the institutions of the nuclear program. But the incident stayed with me as a vivid demonstration of the power of a popular technological-determinist narrative to erase not just the process of technological choice, but history itself. For this young woman, the gas-graphite program might as well not have existed.

Indeed, the guerre des filières -- the subject of this chapter and the endpoint of my own narrative -- has become the origin story of France's current nuclear program. The standard version of the story which circulates in French industrial circles goes something like this. In the beginning, there were two nuclear systems: one centered around gas-graphite reactors, the other around light water reactors. In the late 1960s France had to choose between these two systems. A political battle ensued. The nationalist CEA wanted to pursue the gas-graphite system for irrational political reasons, while the eternally reasonable EDF wanted to switch to the light water system for rational economic reasons. Engineering corps rivalry aggravated the quarrel. De Gaulle supported the CEA's position for several related reasons. He trusted that institution more because of his involvement in its foundation and its role in creating the force de frappe. Top CEA officials therefore had the general's ear and could advance their cause directly to him. As a result, he became mistakenly and unfortunately convinced that only the gas-graphite system was compatible with national independence. Once de Gaulle stepped down, his better informed successor, Georges Pompidou, could make the obviously correct choice. France could finally abandon the inferior gas-graphite system and embark upon building the superior light water reactors under license (to American companies; these days, this part is underplayed). Thus emerged EDF's current nuclear program. Later, the light water system became francisé ("Frenchified"), thereby providing the much-coveted national energy independence as well as a source for national pride. In his closing speech for the utility's 50th anniversary celebration, EDF president Edmond Alphandéry proudly noted that the utility's success was "recognized throughout the world."[1]

Like many origin stories, this one has served to erase a number of events and circumstances that are crucial to understanding not only the process and outcome of the guerre des filières, but also the meaning of both. A few scholars have begun the work of unraveling this history. In particular, Robert Frost and Georges-Henri Soutou have shown that the positions for and against the gas-graphite system did not divide neatly along institutional lines. The light water system had serious supporters within the CEA, especially in the upper echelons of the institution, while many middle-ranking EDF engineers supported the gas-graphite system. But other aspects of the process remain unclear. How were the comparisons between the two systems carried out? How did the light water system emerge as the "apolitical" solution? Furthermore, the meaning of the guerre des filières for engineers, workers, and technicians at both institutions has never been the subject of historical analysis. When unionized CEA employees learned that the gas-graphite system would be terminated, they conducted nation-wide strikes for over a month; today, only the strike leaders remember that these protests centered around programmatic issues.[2] Meanwhile at EDF, cleaning up the Saint-Laurent reactor after an accident that had caused a partial meltdown became a means for employees to enact the meanings of the guerre des filières and its consequences. Ironically, then, even though this is the most studied episode in the history of the gas-graphite program, it is also the least well understood. This chapter offers a new, more inclusive framework for understanding the conflict.

The chapter also concludes this book. It takes up most, though not all, of the threads explored thus far and weaves them together in a series of connected episodes. We will see how advocating the light water system gave EDF managers a way to enact their reformulation of the utility's technopolitical regime. We will see how the relationship between technology and politics formed a strategic centerpiece for both light water and gas-graphite advocates. We will look at how engineers, managers, and union leaders responded to media representations of the guerre des filières. We will see how technological, economic, and political uncertainty enabled engineers, managers, and union leaders to propose different technological scenarios and corresponding conceptions of France and its future. And we will see how cleaning up the Saint-Laurent reactor offered workers and engineers a means of responding to the technological and cultural threat posed by the termination of the gas-graphite program.


The idea that France might pursue other reactor technologies did not appear out of the blue. The research and military branches of the CEA had been investigating other designs for some time. These included a small light water reactor intended to power a submarine as well as prototypes for a heavy water reactor, a high-temperature reactor, and a fast-neutron, or breeder, reactor. EDF researchers and engineers participated in designing and building the "classical" part of the CEA's heavy water prototype starting in 1961. Breeder reactors also seemed promising, and some utility engineers closely followed the development of the CEA's experimental reactor Rapsodie, later becoming involved in developing the industrial-scale breeder Phénix. Finally, EDF's nuclear division had sought involvement in reactor projects not tied to the CEA. In the late 1950s the United States concluded an agreement with the newly formed Euratom intended to favor the import of American reactor designs into Europe. Somewhat reluctantly, the French government allowed EDF to cooperate with Belgium in the construction of the first such reactor in 1960.[3] Despite these efforts, however, technopolitical support for the gas-graphite system's dominance held fast through the mid 1960s.[4]

What happened in 1966 to shake this consensus? Most scholars agree that the trigger for the guerre des filières came from outside France. In 1965, American reactor manufacturers embarked on an aggressive marketing campaign based on extremely optimistic capital cost estimates. Persuaded by these attractive numbers, American utilities jumped on the nuclear bandwagon, ordering 49 reactors in 1966 and 1967, destined to produce nearly 40,000 MW of electricity. Not until the mid-1970s did utilities discover that actual costs were over twice these estimates.[5] In 1966, however, the American estimates presented a serious temptation to French program leaders.

Of course, these figures did not, by themselves, mandate a change in policy. Had harmony reigned in the nuclear program, it is unlikely that the guerre des filières would have taken place at all. As we saw in Part I, however, relations between CEA and EDF engineers and scientists had worsened as their collaboration deepened. Conflicts had also emerged within each institution. The possibilities raised by the American figures served to deepen and rearrange existing fault lines.

The first official intimation that EDF leaders seriously wanted to consider replacing the gas-graphite system came in March 1966, in a letter from André Decelle, EDF's director-general, to Robert Hirsch, the CEA's administrator-general.[6] Affirming that gas-graphite reactors were competitive with conventional power plants, Decelle stated that they were nonetheless "significantly more expensive" than light water reactors. While a system based on natural uranium fuel had the advantage of promoting French autarchy in this domain, in the long term this advantage might not continue to be worth the price difference.[7] Noncommittally, Hirsch agreed that the matter required further study. In May the two leaders created a committee to undertake a comparison of the various reactor systems operating in Europe and America. The committee was jointly headed by Jules Horowitz, the CEA's Directeur des Piles, and Jean Cabanius, EDF's Directeur de l'Equipement.[8] We shall return to this committee in the next section.

Much as Decelle and Hirsch may have wished these investigations to keep a low profile, this quickly proved impossible. The press learned about the committee and its purpose, and a wave of articles appeared asserting, with varying degrees of force, that the current nuclear program would soon be abandoned. These articles flustered EDF managers and worried labor union leaders. At the following meeting of the board of directors, CGT representative Claude Tourgeron demanded an explanation. Was the gas-graphite system being abandoned? Decelle assured him that EDF would continue to pursue the system. The media's wording -- including terms like "abandon" and "rupture" -- was excessive; it was important not to be shaken by journalistic language. The two establishments had simply decided to study both breeder and light water reactors. Hirsch (the CEA's representative on the board) backed Decelle, declaring firmly that "this information... is more spectacular than it is well-founded, since of course the question of abandoning the natural uranium, gas-graphite system has never been raised anywhere."[9]

Clearly the directors hoped that the publicity problem would simply disappear. But a series of technical mishaps at EDF3 simultaneously aggravated the utility's public relations problem and intensified ongoing debates within the establishment over EDF's proper relationship with private industry.

EDF engineers and managers had had high hopes that EDF3, the latest Chinon reactor, could compete economically with conventional power sources. Construction delays had somewhat dampened these hopes, but in late 1966 all appeared ready. As engineers began to power up the reactor, however, more problems arose. Two of these were serious: the heat exchangers developed numerous leaks, giving engineers good reason to fear that the entire exchanger ensemble might suffer systemic flaws and require replacement, and the system which detected problems with the uranium fuel rods failed to function adequately.[10]

Both Nicholas Vichney at Le Monde and staff writers at Le Canard Enchaîné were quick to publicize these problems. Vichney blamed both the builders and EDF. He suggested that the problems arose in part because French technology could not meet the high standards demanded by nuclear installations and in part because of EDF's industrial contracting practices.[11] EDF had tried to build something too complicated too fast, and the technical abilities of its personnel could not rise to the occasion. As for the CEA, Vichney wrote, it should stop trying to distance itself from EDF's difficulties and make more serious efforts to resolve the problems, which compromised the very future of gas-graphite technologies. Predictably, the Canard interpreted events more bluntly: "In short, our home-grown nuclear equipment doesn't hold up." The rag also gleefully noted de Gaulle's displeasure at the incidents: "Heads will roll, citizens!"[12]

Such accusations of incompetence further disquieted EDF's board. In an effort to express sympathy and mitigate these harsh judgments, Hirsch made a special point of praising EDF teams for their dynamism in recent nuclear development.[13] Management representatives noted indignantly that the press had neglected to mention a number of important points: the complexity of the reactor, the difficult trial periods undergone by all new technologies, and the troubles experienced by foreign reactors.[14] Union representatives were particularly upset at the accusations leveled against EDF's technical know-how. Tourgeron insisted that the personnel was experiencing a deep "malaise," which he attributed to the "harm done to the prestige of [our] Establishment."[15] The personnel urgently needed reassurance, he said; the press had to be met with a counter-attack.[16] CEA and EDF management, however, preferred not to engage in an active battle with the press, and instead issued a brief statement that simply affirmed the intimate collaboration between the two institutions.[17]

As time went on, media attacks on the nuclear program -- particularly from these two newspapers -- continued to escalate. Relations between Vichney and the CEA had become openly hostile. Suggestions that French nuclear engineers lacked technical expertise angered both unionized and non-unionized employees at the CEA.[18] In April 1968, Robert Hirsch took Le Monde's editor to task for lack of patriotism. The paper's recent report of an EDF2 incident had not been objective, he said, demonstrating Vichney's "customary lack of good will in the face of the difficulties inherent in the development of nuclear energy." Hirsch continued, "such publicity over an incident that even the article characterized as minor can only complicate the task of French industry, currently in charge of the construction of the same kind of reactor abroad," and contrasted the unhelpful attitude of Le Monde with the discretion of the German press during similar incidents.[19] Vichney, however, declared that he was the one acting in the public interest. He argued that recent technical incidents in France's reactors were "`serious' to the extent that they cast doubt both on the competence of French industry and on the competence of engineers in the different institutions called on to build these different plants. It is therefore a matter of national interest."[20] Vichney thus evoked journalism's claim to speak in the public interest to challenge the nuclear program's claims that it acted inherently in the national interest.

On the surface, such media attacks appeared to close ranks. EDF and the CEA affirmed their solidarity, and EDF's unions defended the utility as an institution. Officially, technical problems were just a normal part of technological innovation. Internally, however, these technical problems, translated by the media into a failure of French technology, spurred renewed battles over the utility's technopolitical regime. As in the past (see Chapters 2 and 3), the issue centered on the role of private industry in nuclear development. Repairing EDF3, for example, would be costly and time-consuming; who should bear the responsibility? Two solutions seemed possible. In the first, EDF's Direction de l'Equipement could accept the contractors' suggestion simply to repair the existing leaks and restart the reactor to find out whether the problems were systemic. If the problems did turn out to be systemic, however, the plant would suffer frequent shutdowns, which in turn would have heavy "psychological" consequences.[21] The threat of more negative press, in other words, militated against this course. The second solution involved having contractors rebuild all the potentially problematic components from scratch. But the builders would undoubtedly object, leading to "a lawsuit whose outcome could only be uncertain and far away."[22] This too, then, entailed technopolitical costs.

EDF management proposed a third solution, which it hoped would reduce technopolitical costs. This scheme called for the complete replacement of EDF3's heat exchangers. The contractors would pay for the faulty exchanger components (an estimated 13 million francs). EDF would pay for the rest of the new exchangers, and also pay the cost of dismantling, rebuilding, and improving the exchangers (estimated at 29 million francs). This solution would allow EDF3 to restart more rapidly with better equipment and would avoid a long and difficult quarrel with the manufacturers.

But several directors objected that this solution was too easy on private industry. Tourgeron protested that the builders had had plenty of opportunities to test the exchanger prototypes and that the problem stemmed quite simply from manufacturing defects. Management's solution would "reward mediocrity."[23]

Though most board members would have liked contractors to take greater financial responsibility, they ended up voting for management's solution. The government's representative to the board, Charles Chevrier, took the opportunity to make a little speech, reported in the meeting's minutes:

EDF3's difficulties thus had heterogeneous origins (including media scrutiny) and demanded a hybrid solution. This was not simply a matter of fixing a few leaks. Instead, a whole host of issues clamored for attention: the reactor's capacity factor, the relationship between EDF and private industry, institutional prestige, the components of the heat exchangers, France's competition with England, and the public image of the nuclear program.

Tensions over EDF3's leaky heat exchangers were also tensions over the changes that EDF's new managers wanted to make in the institution's technopolitical regime. Management's willingness to foot most of the bill was not just about saving time and repairing EDF's public image. Managers also sought to forge a new relationship between EDF and private industry, and ultimately a new definition of "public service" for EDF. These efforts worried the labor unions, who wanted to preserve a regime in which EDF would remain the technological and political leader of French industry. The CFDT, for example, worried that EDF employees would play a decreasing role in nuclear development, while industry's role would get larger. It also worried about internal changes, claiming that more and more decisions were being taken by top management, leaving rank-and-file employees with less and less responsibility. EDF's original mission as the model for a new society was being betrayed. "EDF's ambiguous and weak attitude towards the Manufacturers in applying contracts" exemplified these problems.[25] The CFDT did not oppose management's desire to help industry ready itself for international competition. But these efforts could not come at the expense of salaried employees or of "nationalized firms which represent a social and economic attainment which we all value."[26]

EDF3's leaky heat exchangers were thus enrolled in a growing struggle over the utility's technopolitical regime. Would nationalization continue to mean the contractual and technological subordination of private industry? Or would it come to represent a new, more ambiguous position for the utility, in which EDF would continue to lead national energy policymaking by supporting rather than dominating private industry? As we shall see, the guerre des filières would provide a new terrain for this struggle. All the issues raised in the course of negotiating the repair of EDF3 -- industrial relations, reactor capacity factors, public relations, EDF's technopolitical regime, France's international position -- would be played out in the debates over the future of the nuclear program and its relationship to the future of France.


In late January 1967, the Horowitz-Cabanius committee, charged with comparing available nuclear systems, presented the results of its work. It immediately became obvious that the effort to smooth over engineering quarrels between the two institutions had failed. Though Horowitz and Cabanius agreed on some basic numbers -- such as the capital costs for various reactors -- they strongly differed on the technopolitical meaning of these numbers, so much so that they produced separate reports. The points of contention formed a dense weave composed of:

The introduction of each report revealed the position of its author. EDF's Cabanius stated that his goal was simply to compare the cost of the kilowatt-hour produced by different reactor systems. Politics were not his affair: "Political considerations, in particular those relating to the acquisition or manufacturing...of enriched uranium are up to the public authorities and will not be raised in this study. The rapporteur has strictly limited himself to the industrial, technical, and economic side [of the issue]."[27] Horowitz, in contrast, stated his goals as follows: "reveal the lessons to be learned from the current program," "appraise once more the possibilities of the gas-graphite system," and examine the future orientation of the French program.[28] And while Cabanius sought to distance himself from the "politics" of system choice, Horowitz aimed to address "political" issues (especially the problem of fuel supply) directly.

Cabanius began his report with a short chapter on the state of French industry, clearly reflecting EDF management's desire to change the utility's technopolitical regime by redefining public service. EDF, wrote Cabanius, played the dual role of customer and supplier to French industry. As a supplier, it had to offer its clients, the manufacturers, the cheapest possible electricity in order to help French industry compete with foreign companies. As a customer, EDF had to help French companies reorganize themselves into large consortia capable of taking on the massive investments required by reactor manufacturing. Encouraging these consortia to work under a foreign license would further help French industry because the dynamism and success of the licensers would provide important financial and technical support for the licensees.[29] Ostensibly, then, EDF's first priority should be to strengthen French industry.

Horowitz's first priority was also to strengthen France, but this, he argued, was best done by strengthening the gas-graphite system. His first chapter argued that the system had already exceeded expectations in several ways: the price of natural uranium fuel had dropped faster than anticipated, the fuel rods had proved themselves technically reliable, and the reactor cores were performing well. As for the costs of the system, Horowitz argued that these would have been much lower without the many problems that plagued the construction and start-up of Chinon's reactors. Here EDF was to blame, not the system itself. Marcoule proved that French reactors were fully capable of attaining a high capacity factor. Had EDF been more technically competent and shown more consistency in choosing its contractors, the technical and economic performance of the gas-graphite system would have been far better.[30]

Of course, Cabanius disagreed, though he did not engage with Horowitz directly -- each report was written as though the other one did not exist. For him, the blame lay not on EDF but on the process of developing a new technological system, and, to some degree, on the technology itself. The complexity of the gas-graphite system posed particularly delicate construction problems. Yes, Marcoule had performed well, but its reactors were smaller and less complex than EDF's, and the difficulties of building gas-graphite reactors increased dramatically with the scale of the reactor. "The natural uranium-gas-graphite system," wrote Cabanius, "therefore contains a source of incidents which could have serious consequences not so much for the safety of people as for the length and frequency of stoppages and therefore for the capacity factor of a series of plants which are integrated into an energy system."[31] The gas-graphite system was, therefore, inherently flawed.

The light water system, however, was not. Cabanius described this system with considerable enthusiasm, referring to light water reactors in the United States, Germany, and Italy. Capital costs were low, and with so many reactors on order, manufacturing could be standardized (leading to even lower costs and greater reliability than the gas-graphite system). Best of all, General Electric was prepared to guarantee the price for reactor fuel as a function of the number of kilowatt-hours produced. "This formula leaves a powerful manufacturer [i.e., GE] with the costs of technological uncertainties. Surely its acceptance is based on tremendous confidence in the technological quality of the supplies. This trend will probably be irreversible."[32] GE's confidence justified France's confidence. The spread of light water reactors was becoming inexorable. The seeds of technological determinism were taking root.

Horowitz, however, admitted no such determinism. Yes, American utilities had ordered a remarkable number of light water megawatts in the last two years. What was truly amazing was that this enthusiasm was based on the actual performance of just two 200 MW reactors. The performance data, therefore, was not statistically significant. True, the American program would probably succeed, thanks to its technical rigor and the vast resources of its manufacturers. But this did not mean that the same program would succeed in France. And too much data was missing to make an accurate prediction: "The catalogue [listings] for `nuclear boilers' not give a breakdown of the price per [boiler] component; they do not, therefore, enable one to make a detailed techno-economic analysis. And in any event, as the promoters themselves admit, these prices do not correspond to the cost of a few isolated plants; it is only by anticipating the effect of [building] a whole series [of reactors] that nuclear power has been able to penetrate the market in the United States."[33] While Cabanius treated the American figures as reasonably accurate characteristics of the technology, Horowitz treated them as rough estimates based on the context. In Cabanius's analysis, the technology, abstracted from its context, was the most important variable. What mattered was that light water reactors worked and were cheap. In Horowitz's analysis, the functioning and cost of technologies could not be separated from their contexts. What was good for the United States was not necessarily good for France.

Handling uncertainty

Despite his skepticism about the validity of the American figures, Horowitz used these numbers in his calculations (they were, after all, the only ones available). The two men therefore gave similar numbers for the cost of various kilowatt-hours. They gave identical figures for the cost of the conventional kilowatt-hour (3.95 centimes to 3.35 centimes, depending on the plant's capacity factor) and the cost of the pressurized water kilowatt-hour (2.67 centimes).[34] They differed only on the cost of the gas-graphite kilowatt-hour: Cabanius priced it at 3.14 centimes, while Horowitz priced it at 3.04 centimes. The difference occurred because Horowitz felt optimistic that the CEA's new fuel rods would perform yield lower costs, while Cabanius refused to rely on this experimental data. Horowitz thus emphasized the paucity of data on actual operating light water reactors while expressing great confidence in the CEA's equally unconfirmed estimates. Cabanius took the opposite approach, expressing confidence in the American estimates and skepticism in the CEA's.[35]

The two men also handled the uncertainty in the data differently. Cabanius offered quantitative methods. Admitting that the parameters used to calculate costs were subject to change, he constructed a table describing the "sensitivity of the cost of the kilowatt-hour." (see Figure 9.1) This table showed how the cost of the two nuclear kilowatt-hours would change with variations in parameters such as fuel burn-up rate, capacity factor, or capital costs. Plotting uncertainty offered a sense of control, suggesting that because it was quantifiable, it was manageable. Naming and describing uncertainty, in other words, eliminated the need for qualitative assessment.

Figure 9.1 here

For Horowitz, however, uncertainty opened the door to qualitative judgment. In the absence of hard "facts," political acumen had to guide the choice process. This came through most clearly in his discussion of enriched uranium supplies. Cabanius had dismissed this topic in a single short paragraph: while light water plants would initially rely on foreign supplies, eventually France or Europe would build enrichment plants. Further discussion fell into the realm of "politics," and therefore beyond Cabanius' self-defined mandate. Horowitz also saw this as a political issue, but felt on the contrary that it fell squarely within his mandate. Where Cabanius had written of foreign "supplies," Horowitz wrote of foreign "dependence." Unspecified "political reasons" would make a European enrichment plant impossible, and France could never afford one on its own. Furthermore, the enriched uranium produced in France or Europe would cost considerably more than American uranium, ultimately negating the cost advantage of the light water system.

Horowitz also found other reasons not to plunge headfirst into the light water system. Foremost among these was the need to capitalize on the time, money, and knowledge invested in France's existing technologies. In addition, France's plutonium needs would increase as its breeder reactor program took off; buying plutonium abroad would increase foreign dependency, providing a compelling reason to continue with plutonium-producing gas-graphite reactors.

Nonetheless, Horowitz did conclude that France should probably acquire some experience with light water reactors. But France could not afford to pursue both types of light water reactors, and the choice between them had to take political considerations into account. "General technological and economic arguments do not provide enough information to choose between pressurized and boiling water reactors; this [choice] must therefore follow from considerations proper to the French context, and in particular from the consequences in either case of the recourse to American licenses. We must ensure that, pending a decision, French industry does not get involved in... costly relationships that may later prove useless or premature."[36] Further, whatever the choice, it should not come at the expense of the gas-graphite program: "The pursuit of the program adopted by the Fifth Plan is...fully justified. Indeed, this should ensure -- in addition to a decrease in the cost of the kilowatt-hour -- an appreciable savings in currency, a secure fuel supply, the capitalization of the significant investments made by the CEA, Industry, and EDF, and the fully independent production of the plutonium necessary for civilian and military programs."[37] The gas-graphite program was thus a main source not only of France's political security, but also of its financial security.

Cabanius's conclusion conceded some of these advantages to the gas-graphite system and agreed that current plans to build two gas-graphite reactors at Fessenheim should continue. But he concentrated his efforts on subverting the connection between gas-graphite reactors and national glory by transferring French national interest onto the light water system. This system would "allow our manufacturers, grouped into consortia, to assert their technological value, acquire references that will carry great weight both for export purposes and for agreements with other European manufacturers, and participate in the great industrial confrontation of the next decade."[38] Light water reactors thus held the key to continued French technological prowess; pursuing this system would enable France to take its rightful place in the world's industrial hierarchy. The fact that this system operated under an American license was relevant only because it meant that the French would get technical support for their endeavors. What mattered in terms of national interest was that French companies would be manufacturing parts for the world's leading nuclear system.

Independence or interdependence?

Each man thus constructed his argument in terms of France's national interest. Horowitz's argument fit into the familiar CEA pattern: the gas-graphite system equaled French independence. Rather than deny this claim, Cabanius simply side-stepped it. His arguments formed part of the larger effort to reform EDF's regime. The national interest remained foremost, but its definition revolved around a vision of France in a set of interdependent international relations. French national interest now centered on the economy, which in turn depended on the growth and competitiveness of private industry, which in turn were measured in international terms. The corollary conception of EDF's regime reflected a subtle but significant shift: EDF no longer commanded private industry; instead, it helped to reshape industrial structures and competencies so that France could compete in world markets. In this regime, pursuing light water technologies under a foreign license made national sense. Horowitz, on his side, did not repudiate the goal of helping French companies compete internationally; indeed, this had been the goal of the CEA's "policy of champions." (see Part I). But for the CEA, industrial "champions" had always meant French companies building French technology. Abandoning this pursuit now was folly. In arguing for different systems, the two men argued for different versions of France. Choosing a system was also about shaping the national future. Technology and context were thus inseparably entwined for both men.

The differences in the two reports signaled divisions that would sharpen over time. Gas-graphite technology had succeeded as a technopolitical system. Its developers had consistently invested its technical features with political meaning, and promoted the resulting hybrid as the best choice for the nation. The success of this practice (particularly with de Gaulle) had made it extremely difficult to argue against the gas-graphite system, for to do so could appear fatally unpatriotic. Promoting the use of a foreign license, therefore, required a different strategy, one that involved rhetorically separating technology from politics. Cabanius claimed repeatedly that his responsibility did not include political analysis. For him, politics included anything that had to do with fuel supply. Resituating EDF with respect to private industry was not politics, but simple economic good sense. This position necessarily involved a reconstruction of EDF's role: the utility would move from the political to the economic epicenter of the nation. EDF's initial technopolitical regime had deliberately conflated politics, economics, and technology. In advocating a new regime, management sought a discursive separation of the first element from the latter two. Ironically, this strategy was itself a political quest to change the identity not only of EDF, but also of France.

Although in practical terms the Horowitz-Cabanius committee resolved nothing, some nuclear leaders attempted to extract a resolution from the reports, occasionally referring to them as "the" Horowitz-Cabanius report (singular), thereby erasing the divisions. Not surprisingly, the nature of this supposed "consensus" varied. For example, Pierre Ailleret (EDF) said that while the main arguments in favor of light-water system and against gas-graphite were political, engineers had to continue to search for technological improvements to the gas-graphite design.[39] Other EDF leaders emphasized that the numbers in both reports favored light water and that the utility simply had to bide its time until economic rationality could prevail. For the moment, the least politically problematic way to pursue light water technology was to participate in the Franco-Belgian and Franco-Swiss reactor projects. This would enable French industry to gain much-needed technological experience while escaping the "political vicissitudes of enriched uranium."[40] Clearly, light water reactors were economically desirable. But although EDF was "committed, as an industrial and commercial Establishment, to defend its point of view on the economic considerations relevant to its interests and those of French industry, [it] still had to bow to the political considerations that oppose this project."[41] Opposition to light water technology appeared as a politics separate from (and opposed to) technology and economics.

Those politics, incarnated in the will of General de Gaulle, were not to be trifled with. The general held fast in his commitment to French independence and glory, and his close advisors assured him that these were synonymous with the gas-graphite system. The historical record becomes murky here, especially because appropriate documentation remains inaccessible. In retrospect, the light-water victors argue that had de Gaulle truly understood the multiple dimensions of the problem -- namely its technological and economic aspects -- he might not have supported gas graphite technology so fervently.[42] But the general -- an enduring French icon who continues, by and large, to remain above serious criticism from all but the most ardent left-wing circles -- was "ill-advised." One man in particular usually emerges as the nefarious advisor: Maurice Schumann, the minister of scientific research and atomic questions in 1967-68. Light-water proponents claim that Schumann maneuvered de Gaulle into an intransigent position in favor of continued and exclusive gas-graphite development. One commented disparagingly: "Schumann wasn't an economist, but a typical homo politicus.... [He] did not reason in terms of international industry. He did not recognize the international situation."[43]

Of course, Schumann's version of the story is somewhat different. In a 1981 interview, he proudly admitted that he had defended the gas-graphite system. But, speaking 12 years after the launch of France's light-water program, he claimed more nuanced reasons for this defense. "After studying the file very attentively, I gave absolute priority to the breeder."[44] Breeder reactors would ensure French independence in the future; in the meantime, "the simple and pure abandonment of the French system was not justified; certainly not before the frenchifying [francisation] of the light water systems."[45] He portrayed his position not as a simple knee-jerk reaction against light-water technology, but as a reasoned argument that prioritized French independence and that saw in gas-graphite reactors the technological bridge to a future of breeders (to which we will return in the next section). His job was to ensure that de Gaulle understood the implications of each option. Proponents of the light water system "had advocates -- I was about to say `accomplices' -- inside the CEA, who invoked the dangers inherent in the French system as arguments against it. But the studies that I commissioned showed that foreign systems had at least as many accidents and delays."[46] Had the general remained in office , the former minister concluded, the gas-graphite program would have continued.

Whatever the case may be, all agree that de Gaulle had a formidable will. André Decelle, EDF's director-general, passionately desired a solution to what he construed as a frustrating impasse: economics dictated the pursuit of a light-water system, while politics dictated the opposite. He had tried to persuade various ministers to change de Gaulle's mind. But, he said in a later interview, Massé (EDF's president) would not back him -- not because he disagreed with Decelle's position, but because he disagreed with his strategy.[47] Deeply depressed and discouraged, Decelle resigned in September 1967, citing health and personal reasons.[48] Apparently, he and Massé had agreed not to mention Decelle's advocacy of light water in the resignation statement because doing so would only make it more difficult to pursue the American system later on.

But this precaution failed completely. The very day that Decelle announced his resignation to the board, newspapers proclaimed: "André Decelle, director-general of EDF, resigns. Partisan of the enriched uranium system, he disagreed with the government."[49] Massé, backed by Hirsch, denounced the headlines as "serious counter-truths." They objected to claims that Decelle and others had blindly "championed" particular technologies. Like others, Decelle had simply aimed at "determining, with maximum objectivity and in the spirit of science, where the interest of the country lies in this matter." Hirsch was "particularly shocked to see this effort transformed for public opinion into some kind of passionate conflict."[50]

Marcel Boiteux replaced Decelle at the end of 1967. But regardless of media proclamations, Decelle's resignation did not signal the end of the debate. If anything, it signaled the beginning of a new, more intense phase in the quarrel, in which it became increasingly clear that positions did not fall out neatly along institutional lines. EDF's top leaders still desired light-water technology. So did most private manufacturers of reactor components. And so did a handful of CEA top officials (including Robert Hirsch). But unanimity did not reign. In the CEA, Horowitz and many other engineers and scientists who had devoted their careers to gas-graphite technology continued to defend the French system. Some CEA employees thought that even if it should become necessary or expedient to build light water reactors, France should develop these itself based on the CEA's submarine prototype. In other words, reactors built in France should remain French, regardless of type.[51] French reactors had supporters in EDF as well. In addition to several dozen engineers who felt a personal and professional attachment to gas-graphite reactors, the labor unions opposed buying American reactors for the same reasons that they opposed relinquishing EDF control over contracts with private industry: buying an American license would rob work from EDF's Equipement staff and give it to private contractors. This would harm not only EDF careers, but also the utility's original technopolitical regime.[52]

All attempts at negotiation having failed, the impasse was referred to the PEON commission.


Founded in 1955, PEON (Commission pour la Production d'Electricité d'Origine Nucléaire) was a government-appointed commission composed of top EDF and CEA leaders, ministerial officials, and a few industrialists. Its ostensible purpose was to give the government nuclear advice. It was not, however, a decision-making body. At least until the late 1960s, programmatic decisions were negotiated within and between the CEA and EDF. PEON did little more than discuss and bless such agreements.[53]

The commission's role grew more subtle and complex during the guerre des filières. In the contentious climate fueled by technopolitical uncertainty, commission meetings provided a place for constructing notions of objective arbitration. PEON discussions and reports provided a stage on which members could play hybrid roles: while they were there to represent specific institutions, their membership in PEON symbolically separated them from their institution and gave them a larger constituency -- the nation. Objective arbitration emerged from this hybridity, since -- at least in principle -- any PEON conclusion or document represented a negotiation among various, otherwise competing interests for the greater good of the nation.

PEON members discursively constructed and maintained the commission's status as objective arbiter in their home institutions. Shortly after Decelle's resignation, for example, Massé told EDF's board of directors that for the moment, the program's future rested with the PEON commission.[54] In such statements, the fact that EDF was represented in PEON did not appear. Instead, PEON appeared completely separate from the utility. PEON's objectivity and legitimacy derived from this separation. The same policy conclusion would carry more weight -- both within and outside of EDF -- if reached by PEON. Meanwhile, the ministries -- which until the late 1960s had had very little substantive say in the direction of the nuclear program -- also appeared eager to promote PEON as the ultimate and objective source of resolution for the guerre des filières. The Ministry of Finance, for example, had considered the Horowitz-Cabanius discussions supremely unsatisfactory. It declared that only PEON (which included a Finance representative for the first time in 1967) could produce an objective and useful evaluation of the relative merits of the two systems.[55]

PEON therefore inherited the Horowitz-Cabanius mission. It was to investigate the ramifications of each reactor system and provide a rational, objective basis for short and medium term programmatic choices. Accordingly, in late 1967 its members produced reports on a variety of issues:[56]

The criteria for comparing different reactor types were heterogeneous:[57]

On the surface, these reports appeared to meet expectations for an objective and consensual conclusion, particularly in the domain of cost calculations. And PEON's cost calculations favored the light water system, though its numbers differed somewhat from those of Horowitz and Cabanius.[58] Further -- and this conclusion appeared especially objective, since PEON was thought to be fundamentally pro-nuclear -- the commission did not find nuclear power competitive with conventional sources (largely, it seemed, because the price of conventional fuels had dropped significantly).[59]

Once again, however, numbers did not tell the whole story. Many numbers were missing, uncertain, or incalculable. Comparing system costs raised the same problems: the data corresponded to very different economic contexts, and the prices offered by American and German companies did not necessarily reflect the real building costs. Commenting on the PEON discussions, Horowitz noted bitterly that "the variation in American prices, the sacrifices that AEG and Siemens admit having made recently in order to obtain their first large orders, and the difficulties encountered by Belgian industry in the Doël and Tihange affairs all illustrate well the distinction that must be made between the real cost of an undertaking and the price that must be conceded in order to succeed in certain markets."[60] Another CEA commentator reached similar conclusions:

EDF members were just as aware of the uncertainty in the numbers. Massé later acknowledged that the cost difference between the two systems was less than the error margin in the data used to calculate that difference.[62]

The numbers were uncertain, and calculations indicated that conventional power might prove a wiser course. But PEON did not recommend abandoning the entire nuclear endeavor. Instead, PEON members attempted to define and describe (and therefore shape), not the artifacts directly, but the context in which they would operate. For industrial leaders, this context was the Common Market. This context demanded the pursuit of nuclear technology regardless of the current price of fuel. Offering familiar arguments about competitiveness, the structure of French industry, and the increasing worldwide dominance of this technology, one letter added slyly: "To be convinced, just imagine the position of French industry in a Common Market in which, nuclear power having succeeded, German industry dominated this sector."[63] Since light water reactors dominated the world markets, the light water system was the only plausible choice for this context. Industrialists strengthened their construction of this context by stipulating that they would offer warranties for light water technology, but not for gas-graphite technology.[64] From EDF's perspective, manipulating the context in this way literally transformed the light water system into a more reliable technology than the gas-graphite system.

CEA representatives sought instead to limit the context to national frontiers. But here they met with stubborn resistance from the industrialists, who apparently refused to discuss matters in such terms:

When the issue of French economic independence arose separately from the issue of the Common Market, said another CEA member, "opposition came both from the industrialists, who refused to provide the smallest piece of data, and from the Planning Commission, which as always preferred multiple abstract schemes."[66]

In April 1968 these disagreements disappeared into PEON's formal report. The report concentrated on two elements: the outcome of the cost calculations once the uncertainties were factored out, and the need to base decisions on "objective" economic criteria rather than on political considerations. This second item reflected efforts to redefine the French context: "It is pointless to hope for total independence... The potential for economic independence can be defined as the capacity to maintain economic competitiveness over the long term and on the international front..."[67] The numbers and the context, in turn, pointed to a clear set of recommendations. 1) France should immediately build an American-style reactor. 2) Pending a re-evaluation in 1970, no new gas-graphite reactors should be ordered in the next two years. 3) the Canadian heavy-water design might deserve further consideration (apparently one of the industrialists proposed this recommendation as a way to ease the pain of moving to an American license, since the Canadians were known favorites of de Gaulle's[68]). One energy industry periodical joyfully proclaimed these conclusions the result of a "profound unanimity." And where did this unanimity come from? Quite simply, from the separation of technology from politics:

The important point was that political matters had not dominated the debate. This, in turn, provided the government with a clear basis for action.

The main achievement of PEON's 1968 report was therefore to legitimate two key strategies of light water supporters: 1) the separation of technology from politics and 2) the redefinition of the context of nuclear development as Common Market economics.


Nonetheless, turmoil continued to lurk underneath PEON's façade of consensus. De Gaulle continued to favor the French system. Within both EDF and the CEA employees remained split. Not everyone agreed that technology and politics should be separated, or that the right context for the nuclear program should be primarily economic in nature.

Yet a different source of consensus emerged in discussions outside PEON: the breeder reactor. As a technology that still existed primarily on paper -- only one prototype existed, the CEA's Rapsodie -- the breeder was still flexible enough to fulfill a broad spectrum of technopolitical scenarios. As we saw in the cases of Horowitz and Schumann, gas-graphite enthusiasts had already begun to endow breeders with the same power to carry France's technological glory (once) held by gas-graphite reactors. In the wake of the PEON report, gas-graphite proponents focused increasingly heavily on the breeder future. Light-water proponents, meanwhile, used that future to build a stronger constituency for the American solution.

For some engineers and labor leaders at EDF, the breeder future demanded further pursuit of the gas-graphite system. In July 1968, for example, engineer Claude Bienvenu, the project leader for Saint-Laurent 1, lambasted recent decisions that jeopardized the gas-graphite program. He was angry because an impasse over industrial contracting methods had stalled the launching of gas-graphite construction at Fessenheim. Worse, the companies in charge of construction were trying to re-invent everything: the pressure vessel, the heat exchangers, command and control systems, everything. "Saint Laurent will have been useless!" exclaimed Bienvenu. "The gas-graphite system, which had been ready to derive maximum profit from the experience accumulated and perhaps even to battle with some chance of success against the American system, will find itself blown away like a straw in the wind."[70] Breeders could return France to a more rational path. They also provided the best reason for maintaining the gas-graphite system, which could supply both the plutonium and the experience required by breeder development. Such a course would ultimately allow France to surpass the United States, which currently had no breeder experience.[71]

CGT representative Claude Tourgeron also saw a future of breeders. But his breeders differed from Bienvenu's; they pointed to a socialist future. Tourgeron juxtaposed his argument for breeders with an argument for the "formation of nationalized companies that would free this industry from the joint pressure of large capitalist monopolies and military management."[72] These nationalized companies would provide the basis for a true socialist democracy, which in turn could only lead to national economic growth. Breeder technology would take some time to mature, though, so France had to pursue an intermediate system in order to maintain its nuclear knowledge. Only a system based on natural uranium would allow France both to escape the clutches of American imperialism and to produce plutonium for the breeder future. As for cost calculations which disadvantaged the gas-graphite system, they resulted from nefarious capitalist practices. The Fessenheim estimates, for example, had been inflated by capitalist monopolies in their thirst for profit and their desire to tip the balance in favor of the American design. Thus gas-graphite success, breeders, and a socialist order were mutually dependent.[73] The technopolitical circle was complete.

Though their visions differed, Bienvenu and Tourgeron, along with Horowitz and Schumann earlier, all saw a future of breeders. This consensus was remarkable, since aside from their enthusiasm for gas-graphites and breeders the four men had little in common. Proponents of the American system seized on this consensus but created a different path to that future. For example, top EDF management sent a memo to the prime minister in February 1969, arguing that France should make every effort to research and develop breeders, "the system of the future." But the main road to that future went through the American system, which -- according to both EDF's Boiteux and the CEA's Hirsch -- was "without any doubt" the best bet for providing competitive and reliable electricity. Using an American license would allow France to recover from the disappointment of the gas-graphite experience and to "catch its breath while waiting for a new breakthrough -- that of the breeders -- to which it will devote all its research and development efforts."[74] Not even the CEA's experience in designing a light-water reactor for submarines would go to waste. Instead, it would help French teams "mix French intelligence with American experience to build a `Frenchified' reactor."[75] Thus they too transferred the burden of French grandeur over to the breeders. Further, the nebulous "Frenchifying" of American reactors would preserve French nuclear know-how (and, presumably, pride). In April 1969, Boiteux and Hirsch formalized this line of reasoning into a "plan of action." This plan essentially reframed the proposals and arguments of the PEON report by placing them in the logic of a breeder future.[76]

In presenting this plan to EDF's board, Hirsch and Boiteux emphasized that the plan kept the natural uranium option open out of prudence. They stressed that "the realization of the first light water reactor will take place in the framework of a general license in order to draw from the Americans a maximum of amount of knowledge about the chosen system, without going so far as a joint warranty by the licenser and the licensee."[77] Thus the French would derive maximum benefit out of the partnership while leaving the Americans responsibility for the technical warranties. At the meeting, Paul Delouvrier, Massé's successor to the EDF presidency, waxed enthusiastic about the plan. Although even light-water reactors were (temporarily) more expensive than conventional oil-fired plants, he affirmed that this was the price that France had to pay to keep up-to-date on matters nuclear. After all, "it is not without some sadness that one sees AEG and Siemens put a plant in Holland, given that the nuclear industry got a much later start in Germany than in France. It is definitely time for the country to get hold of itself in order not to be surpassed and dominated."[78] With that, he gave Boiteux and Hirsch his blessing, and that of the Conseil d'Administration, to present this plan to PEON. In the meantime, EDF managers had already begun to mobilize the utility's divisions in preparation for the first light-water reactor.[79] With Tourgeron absent from the board meeting, no one raised any objections to the plan. The following month, Hirsch and Boiteux presented their plan to a specially convened PEON subcommittee, which approved. This group produced yet another report, virtually identical to the plan of action, except that it added a discreet line announcing that the choice of reactor design for Fessenheim was being reconsidered.

Just as the CEA had capitalized on the ambiguity of the gas-graphite design in the mid-1950s to inch forward with the military atom, so now light water advocates capitalized on a variety of ambiguities to inch forward with plans to buy an American license. Each successive report tightened the case for light water, using a combination of three strategies. Foremost among these was the management of technological and economic uncertainty, either by quantifying and plotting potential data fluctuations or by pronouncing on the relevance and function of the different uncertainties. Another important strategy involved shaping the context in which future nuclear development would occur, notably by redefining the concept of "national independence." And the final strategy involved constructing a new logic for light-water development. Embedded in successive reports, these strategies created a discursive teleology of nuclear development. With each successive refinement, American light water reactors became more and more necessary for the future of France. As Simmonot argues in his analysis of the PEON literature, each successive report further instantiated a logic of technological determinism.[80]


The incremental way in which advocates of the American system developed their plans, along with the fact that these advocates occupied the top administrative positions in the CEA, EDF, and private industry, made opposition difficult to orchestrate. The carefully ambiguous construction of the plans also complicated matters. So far, for example, no one had actually propose terminating the gas-graphite program. And the stated goals -- to give France cheap energy and to make breeders the new symbols of French technological glory and independence -- were irreproachable. Finally, plans and reports notwithstanding, no one -- and specifically, not the government -- had made any actual public decisions.

Nonetheless, by June 1969 it was clear that EDF management and private industry, supported by top CEA officials, were poised to buy American. It was also clear that buying American would come at the expense of the French system. Furthermore, by then the one man seen as a guarantee against the purchase of a foreign license -- Charles de Gaulle -- had resigned from office. He had been replaced by Georges Pompidou, who clearly had sympathies for the American system.[81]

In an effort to obstruct a course of action that increasingly appeared inevitable, labor union members began to contest the economic analyses emitted by program leaders. Union members offered alternative figures, calculations, and interpretations. Such efforts began when CGT representative Claude Tourgeron registered an official protest at EDF's June board meeting.

Focusing on the uncertainty in the light water data, Tourgeron's protest contested the notion of worldwide light water dominance, resurrected the issue of national independence, and challenged the attempt to replace technopolitics with "pure" economics. He noted that light water reactor orders in the United States had dropped dramatically between 1967 and 1969. Tourgeron explained this drop-off by an increase in the capital costs of these plants (now up to 1000-1100 F/kW, the same range as Saint-Laurent 2). He also argued that American utilities had lowered their predictions for the capacity factor of light water reactors and were even building "rustic" thermal plants to take over when reactors had to go off-line. He reiterated familiar arguments about the threat posed to national independence by reliance on enriched uranium. He also added a new twist: American enriched uranium was inexpensive primarily because isotope separation plants "were paid off a long time ago by taxpayers" (presumably during World War II).[82] Neither France nor Europe could ever hope to approach American prices. Finally, he argued that the criteria for making decisions about the future of the French program could not be economic: prices fluctuated too much to provide a reliable foundation for decision-making. Both technical and political considerations militated in favor of more gas-graphite reactors to link the present with the breeder future.

Several board members denied Tourgeron's allegations. Hirsch explained the decline in American reactor orders by market cycles. Others denied the validity of Tourgeron's calculations not by addressing these directly, but by simply reiterating PEON's economic estimates. Finally Boiteux closed the discussion by insisting that a consensus existed on two matters. First, the future belonged to breeders and that France should do everything to preserve its lead in this domain. Second, the country had to engage in some kind of intermediate program to ensure that French industry acquired and mastered nuclear technology as a whole. The only two truly viable candidates for this intermediate program, he continued smoothly, were the gas-graphite system and the light-water system. His next statement completely ignored Tourgeron's estimates, asserting that "all the numbers cited in this discussion -- which are based on experimental results, [actual] developments, and recent requests for bids -- proved that the light water system was the most economically viable and the least capital intensive. This is the reason it was chosen."[83] Had a chioce been made? Was this last statement a slip of the tongue, or a reference to American decisions? It was unclear. But Boiteux hastened to add that for the moment, the natural uranium option had not been closed. He postponed that decision for another twelve to eighteen months pending further data. Other board members murmured their assent, and the issue was temporarily tabled.

Meanwhile, CEA union leaders had also begun objecting to the emerging plan, which they felt threatened both their jobs and the future of the French nation. Initially, their primary objection centered around the fact that these plans were drawn not by the government -- the ultimate representative of the people, however objectionable it might be -- but by institutional leaders. Contrary to media allegations, political rather than technical weakness had caused the current problems. The CEA's CFDT section wrote that "The difficulties faced by the [CEA] today do not come from technical failures, but from the government's lack of research policy and industrial policy."[84] The CEA's scientific and technical potential was being ignored. The government needed to establish a coherent research program, one based on "technological areas in which France is especially and dangerously under-developed."[85] The CEA's CFDT section advocated a new institution similar to that proposed by Tourgeron: a state-run financial institution that could create new companies or regroup existing companies. The state could thus manage private industry and give rational direction to the nation's research and industrial development. This would also prevent Westinghouse from taking over France's electromechanical industry -- which, after the publication of a Westinghouse report on the European electrical industry, appeared to be a very real threat.[86] Finally -- echoing the autogestion (self-management) demands of the May 1968 strikes, during which many CEA employees had become radicalized -- the CFDT asked that CEA white and blue collar workers be given more say in program management and decision-making.[87]

By October 1969, rumors had begun to circulate that the CEA's programs would be cut back and that layoffs would ensue. The CEA's five main unions joined forces to protest the layoffs, the introduction of American light-water reactors, the implied slurs on their technical competence, and the incoherence of national nuclear research policy. On October 10, 800 employees staged a demonstration at Marcoule. Meanwhile, at the Saclay research center, unions avidly defended the performance of the French nuclear program, arguing that it had been "submitted to systematic, ...unfounded criticism by part of the press, encouraged by the eloquent silence of CEA and EDF top management."[88] The real problem, they continued, "contrary to what is written daily in the press, has nothing to do with the high price of French nuclear plants, but instead [comes from] on the one hand, the dumping [?] prices practiced by oil companies... and on the other hand the current structure of the French electromechanical industry in general and the nuclear industry in particular."[89] The price of the gas-graphite kilowatt-hour was already 30-40% lower than the most optimistic estimates emitted by the Plan several years earlier. On that basis, the Plan -- which, however imperfectly, still represented popular will better than EDF and CEA leaders -- had called for 2500 MW worth of new reactors. Only 1300 MW of these had been built.

The unions demanded a coherent nuclear program whose main criteria of success would be continuity, independence, and the development of a national electromechanical industry. This policy "must first and foremost be translated into the development of the gas-graphite system."[90] It would be "stupid" to abandon this and other national technologies. In a separate statement, the CGT called for the publication of reports that would "re-establish the truth which is indispensable to the defense of French atomic energy... The CGT's engineering and white-collar worker section will not hold back in its efforts to ensure that France remains independent in the energy sector."[91] Others used even stronger language to denounce the intrusion of Westinghouse into French industry. "What some are calling the `guerre des filières' is a booby-trap," cried the PSU representatives. "It's really a war between international trusts orchestrated by one of them: Westinghouse." They continued: "What could Westinghouse's intrusion into the closed world of bourgeois businessmen and technocrats which governs us mean, other than the brutal manifestation of American imperialism in our midst. Elsewhere, it kills by war; here, it seeks to reduce us to the state of an economic colony. Let us not be dupes: the French government is not neutral in this affair. It's an accomplice. It's the enemy of the workers. [emphasis in original]"[92] For them, clearly, the government could not be counted on to produce a reasonable solution. It was, instead, colluding with private industry to orchestrate an American takeover of France.

Boiteux declares the end

The situation finally exploded in mid-October at Saint-Laurent 1, the pride and joy of the gas-graphite program. The reactor had been operating for several months and had already produced a billion kilowatt-hours of electricity. CHECK! On October 16, Marcel Boiteux, accompanied by Robert Hirsch and Francis Perrin, went to the site for the official inauguration of the reactor. During his press conference he congratulated the site's teams on their success, declaring that Saint-Laurent was the best of EDF's reactors. Too bad, he added, that the gas-graphite system was not commercially viable. From then on, he said, EDF would be building light-water reactors under an American license (see Figure 9.2).[93] This announcement sent a shock wave throughout the nuclear program, the government, and the press. Everyone knew that this was the direction in which the program was headed, but no one thought that a formal decision had been reached.

------------ figure 9.2 here ------------

Figure 9.2: Marcel Boiteux gives a press conference at Saint-Laurent on October 16, 1969. EDF's official caption for this picture is "Inauguration of the plant at Saint-Laurent-des-Eaux." Gas-graphite supporters would have titled this picture "Boiteux announces the termination of the French system." (photographer: M. Brigaud. Courtesy Photothèque EDF).

Reactions in the press were mixed. Le Monde's Nicholas Vichney was soberly jubilant. Acknowledging the technical success of Saint-Laurent, he followed Boiteux in emphasizing its economic drawbacks. Noting that EDF had been interested in American reactors for some time, he could not resist a dig at the CEA: "Opposing this realistic attitude was the hope of finding in nuclear plants support for energy independence and the [CEA's] desire to operate the type of plant it had developed."[94] In contrast, Pierre Juin writing in the Communist daily L'Humanité was scandalized. His front page article featured a photograph of Saint-Laurent with a caption describing the site as a "prestigious French achievement."[95] Saint-Laurent's technical success, Juin wrote, might lead one to expect that "the top brass of [EDF] and the [CEA] who piloted specialized journalists through the vast construction site of Saint-Laurent on Thursday would be overjoyed. Well no." He went on to impute the decision not just to EDF, but to the government more generally. "In his press conference last Monday, Mr. Ortoli, minister of industrial and scientific development, had declared that France's nuclear policy would be fixed at the end of the year... Mr. Boiteux, however, could not hide that the case had already been heard." Juin continued bitterly: "During a lightning interview, which only allowed for a half-dozen questions, Mr. Boiteux affirmed that all countries were now oriented toward light water reactors and that, as a result, it would be tasteless to obstinately pursue our own technology in the restricted space of the French hexagon." The decision had been the result of pressure by foreign monopolies and would seriously endanger French independence.[96] In a similar vein, the caption of one Canard Enchaîné cartoon read "U.S. Go Ohm!"[97] The satirical weekly's tone and message resembled that of L'Humanité, but the Canard noted that the decision had not officially emanated from the government. "Pompidou is slowly rushing to decide nothing," sneered the weekly. "For the moment, he is still in training. After all, the French system is the general's gadget. Got to treat that carefully. The dear old gentleman might take offense."[98]

Indeed, Boiteux's announcement even surprised Pompidou, who had not officially taken the decision, as well as some of EDF's board members, who had thought that matters were still up for debate. Explaining himself a few days later, Boiteux emphatically denied that he had announced any sort of decision. He had merely stated that because the economic success of Saint-Laurent was less certain than its technical success, the future of the gas-graphite system remained uncertain. It was "regrettable that his words were given the political meaning that they were."[99] Journalists had misinterpreted his responses to their questions. He had said that "there was no reason to regret what had been done in this domain, [since] the effort poured into the `gas-graphite' system fit into the logic of the nation's history, but that today the fact nonetheless remained that nuclear plants were too costly, and it was only right to question whether an Establishment like EDF should continue to build them."[100] He had merely indicated that EDF had a preference for a light water system. The press had not mentioned that he had referred all final decisions to the government. Delouvrier expressed his support for Boiteux and emphasized that the government had not yet made a decision, though everyone hoped that this would soon occur. Tourgeron and other labor union representatives again objected to management's pro-light water position, but with less verve than before.

Disclaimers notwithstanding, Boiteux's statement was widely understood to signal the end of the gas-graphite program. For CEA employees, the first of the rumored layoffs confirmed this signal : the same day that Boiteux held his notorious press conference, 98 cleaning ladies sub-contracted to the Saclay research center were let go. The next few days saw several more layoffs, all branded by the unions as violations of the labor agreements drawn after the 1968 protests. On October 23, Saclay's director returned from a trip to find the site's unionized personnel up in arms. He refused to revoke the layoffs. Four days later, 700 Saclay employees launched a series of strikes that would last for over a month.

CEA strikes

In order to understand these strikes, we must briefly go back in time to 1968. During the nationwide protests that year, numerous CEA engineers and technicians had joined unions.[101] Like demonstrations elsewhere in the country, the 1968 sit-ins at the CEA focused on democratizing the workplace and loosening the institution's decision-making hierarchies. From the perspective of the protesters, the results had been somewhat mixed. They had obtained new administrative bodies which, at least in theory, made room for broader participation in managing daily workplace affairs. But as the October 1969 layoffs indicated, not all CEA directors had taken well to these new structures. Further, as the termination of the gas-graphite program showed all too clearly, the administration had no intention of including the personnel in programmatic planning, not even in the cursory style to which EDF's board had devolved.

Still, the 1968 sit-ins had provided a brief opportunity for many CEA workers, technicians, and engineers -- particularly in the research centers -- to enact, in however small a manner, the cross-class solidarity that the national labor unions had been advocating (see Chapter 4). The most extensive of these sit-ins had occurred at Saclay, where the working population consisted primarily of engineers and technicians.[102] At Marcoule, 1968 apparently did little to change the relationship between workers and engineer-managers described in Chapter 5. But the fact that protests occurred across many CEA sites showed unionized employees that they shared sentiments with their colleagues (even those of different hierarchical status) at other sites. In 1969, this new, somewhat shaky alliance amongst engineers, technicians, and workers across multiple sites resulted in protests that combined the practices and goals of labor unions with those of engineers. The "interests" of the different groups became indistinguishable in the ensuing strikes. These combined demands to halt the layoffs with calls for greater employee participation in management and challenges to the termination of the gas-graphite program. The 1969 CEA strikes, in other words, fused issues of national technological policy with issues of social relations.

The initial group of protesters at Saclay included five hunger strikers. These men demanded the revocation of all layoffs. Echoing the tones of 1968, they presented their case as a moral issue, a matter of basic social equity:

On that note, the five men installed themselves in Saclay's labor union offices on the afternoon of October 29, only to be evicted a few hours later when 240 gendarmes stormed the site. For the next two and a half weeks, the men would continue their hunger strike in a nearby Protestant temple.[104]

On October 31, news leaked that the administration would announce another 2000 layoffs on November 12. The unions responded by broadening their demands and intensifying their strike actions. Though they continued to express outrage on behalf of the cleaning ladies and other sub-contracted workers, protests now focused primarily on the issue of nuclear policy. The strikes continued through the end of November and spread to all of the CEA's research and production centers.[105]

Echoing earlier arguments, strikers denounced the termination of the gas-graphite program and the threat of an American industrial takeover.[106] They contested the assertion that gas-graphite reactors were not competitive and argued, furthermore, that "profitability [was] not the only important criterion."[107] National independence had to count too -- especially independence from the United States. Never had the threat of American capitalism loomed larger. "We are in the process of losing our national independence; we are on the path to under-development and colonization."[108] (see Figure 9.3) French plants, asserted the protesters, were equivalent in quality and cost to American ones, and nuclear research had been a source of French pride for decades. Even the British nuclear industry admitted this: at one press conference, union leaders noted that one British engineering journal had said that the French had a "natural flair" for nuclear technology and science.[109]

The problem, said the unions, lay in the fact that the government had not handled either industrial or research policy properly. "Such an important decision...should be preceded by consultations with employee representatives, not announced on the fly by a bureaucrat, no matter how highly-placed he might be."[110] Now only the unions had the nation's welfare firmly in sight: "Our goals are clear. We are in favor...of funding research which will ensure the intellectual, economic, and social future of an entire people and guarantee its independence."[111] Although nuclear weapons did not form a significant issue in the strikes, three of the unions could not resist a jab at the military program: "Strangely, military applications, which constitute the least important part of nuclear research, are not in the least affected by restricted funding. The government talks of national independence when atomic bombs are involved; at the same time, it is liquidating our national industry, which is the measure of true independence and the source of progress and well-being."[112]

------------Figure 9.3 here ----------

Figure 9.3: The Confédération Général des Cadres, the white-collar workers union, issued tracts which imitated American $1 bills during the 1969 strikes. Note the alterations made to the front of the bill (all in English): "What is good for Westinghouse is good for France"; "In EDF we trust"; "Business is business"; "MB" for Marcel Boiteux; "PWR" for pressurized water reactor. The following explanation was printed on the back of the tract: "To confront American technology, the COMMISSARIAT A L'ENERGIE ATOMIQUE can and must -- in the national interest -- become a powerful and diversified group in which employees are involved in managing the firm and in which employment is guaranteed. Because of the current energy crisis, the Commissariat, which is at the origin of nuclear development in France, must acquire greater responsibilities in the nuclear arena, as well as a sufficient budget and the necessary personnel. A new personnel policy must be defined, one which involves the resumption of recruitment." (Tract courtesy of Jean-Claude Zerbib).

On November 14, the CEA administration finally reinstated the cleaning ladies, and the hunger strikers stopped their protest. If the administration hoped that this concession would end the strikes, however, it hoped in vain. That same day Pompidou formally announced the termination of the gas-graphite program for the foreseeable future. Although the "foreseeable future" clause was intended to leave open the possibility, at least rhetorically, that the gas-graphite system might find favor again some day, no one paid it much attention. The CEA strikes continued to intensify. On November 17, somewhere between 4000 and 6000 protesters descended on the Place des Invalides in Paris and marched past the Eiffel Tower (see Figure 9.4). Strikes continued in the provinces too. According to CGT statistics, 90% of Marcoule personnel were on strike between November 14 and 18.[113]

------------Figure 9.4 here ----------

Figure 9.4: CEA protesters march past the Eiffel Tower, historical symbol of French technological glory. (photographer: Philippe Mousseau, Lumifilms. Courtesy CFDT archives)

These strikes combined typical employment matters with issues of national industrial and research policy in a seamless web. The heterogeneity of the issues raised during the strike doubtless derived in part from the heterogeneity of the strikers themselves, who ranged from the blue-collar workers at Marcoule to Saclay research scientists and engineers. Doubtless realizing that purely political tactics would have little effect in a debate whose terms were defined by its dominant participants as economic -- and, by extension, apolitical -- a group of union engineers, scientists, and technicians prepared a counter-report on the relative merits of the competing nuclear systems.

Economic comparisons, union-style

One major difference between the union report and those written by advocates of the American system lay in how each group posited the relationship between technological development and politics. American system advocates, as we have seen, sought to remove what they derisively called "political" considerations from the decision-making process. Union advocates of the French system, on the contrary, sought to bring such considerations back into the process. Paralleling but also extending the points Horowitz had made in his report, the union document argued that the importance of political considerations in nuclear energy policy derived from the fundamental uncertainty of the data on nuclear power.

A major source of this uncertainty, said the union report, were differences in the financial and technological conditions under which power plants operated in the United States and France. This gave an artificial advantage to the American system in at least four ways. First, the amortization period for reactors in France was 20 years, while in the U.S. it was typically 30 years. Since a shorter amortization period penalized plants with higher capital costs, this difference unnecessarily disadvantaged gas-graphite reactors. Second, the capacity factor used in the calculations differed in the two nations: 6800 hours/year in France versus 7500 hours/year in the US. This too penalized French reactors. Third, price comparisons between nuclear and conventional power in the two countries operated under different principles. In the U.S., for example, the price of conventional fuel included the transportation cost to the power plant door. French pricing, by contrast included freight costs only to the port of entry. Taking port-to-plant transportation into account would raise the price of French conventional fuel by 0.37 c/kWh and therefore make the gas-graphite system more attractive. Fourth and finally, were France to engage in the "draconian" precautions that the U.S. took to reduce pollutant emissions, the price of conventional fuel would increase even further, perhaps by as much as an additional 0.85 c/kWh. None of these factors, said the unions, had been included in EDF's calculations.

Indeed, claimed the report, "the capital costs announced by EDF are incomprehensible and incoherent."[114] For example, the figures used to represent the capital costs of light-water reactors did not include the fact that two such reactors in the U.S. had gone 30 and 90 per cent over budget respectively. Combining this with the spectacular reduction in gas-graphite costs achieved by Saint-Laurent 2 yielded equivalent capital costs for the two reactor types. The calculations presented in the report's appendix concluded that an American light water reactor would produce electricity costing 2.93 centimes/kWh to 3.08 c/kWh, while the electricity from the second Fessenheim gas-graphite reactor would have cost 2.91 c/kWh.

Still, the union report argued, such numbers had limited value: "all the plants on which current economic comparisons, and therefore decisions, are based are `theoretical' plants."[115] Reliable numbers for fuel cost, use rate, operational costs, and amortization would only come from more extensive operational experience. Further, it was impossible to tell how the numbers used by EDF and PEON had been derived, since the actual calculations remained hidden. And finally, current economic studies were based only on the direct cost of the reactors, without taking into account the investments that either nation had already made in the technological system that supported each reactor type (which included fuel manufacturing plants, treatment plants for spent fuel, research infrastructure, the military functions of reactors, and more). "The unannounced but implicit abandonment of this system, into which the CEA and EDF have poured considerable investments, is therefore completely incomprehensible. This is no longer a technological choice, but a deliberate political choice, contrary to the most basic interests of the nation [emphasis mine]."[116] In essence, the unions used economic reasoning to show that the decision had indeed been political.

How did the unions view the politics of the situation? For them, the decision to terminate the gas-graphite program represented a capitulation to capitalism -- American capitalism in particular. "Everyone is aware of the concerted offensive launched by American industrial consortia to get hold of the French electromechanical [industry]," said the report. Pompidou's announcement merely confirmed the "Americanization of the French electronuclear [program]."[117] But the report did not argue that politics should have been left out of the decision. Instead, it argued, the wrong politics had guided policy-makers. When the uncertainty of the economic data was taken fully into account, the resulting estimates were "sufficiently close for other criteria of choice (currency flow, capitalizing on existing investments, national independence, full employment) to be considered on the same plane."[118] Rather than base a decision purely on the politics of capitalist development, in other words, the government should have also taken social politics into account. And it should have weighted other elements (such as national independence) differently. In conclusion, the report called for the creation of a new commission -- composed of ministerial officials EDF and CEA management, and labor unions -- to re-examine the case.

The battle fizzles out

On the evening of November 20, a delegation of union representatives brought this report to a meeting they had managed to schedule with prime minister Jacques Chaban-Delmas. He listened to them but refused to revoke the layoffs. Nor did he offer much hope on the programmatic front. He did not refuse outright to consider the unions' proposals, but he made it clear that he would probably turn them down. Discouraged, three of the unions -- the CFDT, the CGT, and FO -- called for more strikes the following Monday. Those would be the last ones: as it became increasingly clear that the programmatic decisions would hold, the unions lost heart.

After the CEA strikes the guerre des filières faded quickly from public view. In January 1970 Boiteux shared the latest PEON report with EDF's board of directors. This report essentially re-iterated the points outlined in the "plan of action" sketched the previous year and specified that the Fessenheim site would house light-water, not gas-graphite, plants. More as a matter of form than anything else, the union board members objected that the termination of the gas-graphite program had not been finalized. Their arguments were futile, however, and soon union members turned their energies to struggles they thought they could win. For the rest of 1970, debates continued to rage in the board over the contracting and organization of the Fessenheim projects. Confrontations over the design itself had ceased.[119]

Given what we know about the two institutions, it seems ironic that the strikes protesting the termination of the gas-graphite program occurred at the CEA and not at EDF. Several factors may explain this apparent irony. For one thing, the termination of the program did not threaten EDF jobs. The utility's labor statute guaranteed against technological unemployment, so that at worst employees would need to learn new skills. In addition, both unionized and non-unionized proponents of the gas-graphite system had already had several opportunities to present counter-arguments in reports and board meetings. Though these arguments had little effect on top management, EDF employees did not feel excluded in the same way as their CEA counterparts. Furthermore, while switching to another technology did hurt the pride of those who had labored on the gas-graphite system, it did not threaten the foundation of the utility as it did that of the CEA. Top management did use the guerre des filières to try to reshape EDF's technopolitical regime, but even this did not threaten unionized utility employees as profoundly as the termination of the gas-graphite system threatened CEA employees. At EDF, unionized employees could continue to oppose the new regime by challenging its contracting practices. CEA employees had no recourse but strikes.

Finally, one additional factor must be considered. Those who might have protested the loudest -- the designers and workers at EDF's gas-graphite reactors -- had a more urgent task to attend. The day after Boiteux's announcement, in a dramatic coincidence of time and place, one of the most serious accidents the nuclear industry had seen so far caused a partial meltdown of Saint-Laurent 1. Engineers and workers spent a full year cleaning up the mess and putting the reactor back on line.

It was in this activity, rather than through strikes, that they expressed their reactions to the termination of the program. Rather than contest the decision by striking, they sought to contest its apparent meaning by working to repair their reactor. No discussion of the program's demise can be complete, therefore, without a look at the clean-up of Saint-Laurent 1 in 1969-70.


As we have seen, even proponents of the light-water system called Saint-Laurent 1 an outstanding technical success in the first few months of its operation. Its designers and workers proudly called it the most elegant and efficient of all French reactors. They argued that the exceptionally well-planned and efficiently executed construction phase showed that nationalized companies should indeed lead France's technological efforts.[121] Even more proudly, they noted that at 480 MW, Saint-Laurent 1 was one of the most powerful reactors in the world. When it went on line in March 1969, it promised to help "defend the colors of gas-graphite"[122] by proving that the French system could compete not only with conventional plants, but also with nuclear power in other nations. Workers were prepared to put in long hours to help it succeed. Time, said one man, did not count: he once worked 36 hours in a row without sleeping just to get a job done, and remembered his boss coming by at 2 a.m. to bring his shift "a snack and a pat on the shoulder and say how you guys doing?" The work atmosphere was "very friendly, very convivial. We worked hard, but for love, eh?"[123]

The reactor appeared to hold the technopolitical key to the continuation of the gas-graphite system.[124] As such, the significance of its success for those who designed and operated it was both political and personal: their time, energy, and skill had made it France's most important technological achievement. It is not difficult to understand, therefore, why Saint-Laurent employees experienced Boiteux's announcement as "a stab in the back."[125]

------------Figure 9.4 here----------

Figure 9.4: The core of Saint-Laurent 1was made up of nearly 3000 vertical channels, each of which contained 15 uranium rods. Each rod was encased in a metal shield and surrounded by a graphite shell. The fission reaction thus took place inside the core, producing a great deal of heat. Carbon dioxide cooling gas flowed through the channels in the core and absorbed this heat. The hot gas then flowed into the heat exchangers, where it converted water into steam; the steam powered the turbines (not shown), thereby producing electricity. The entire reactor was encased in a concrete pressure vessel. On top of that vessel sat the loading machine. Reactor operators used this machine in order to remove spent fuel from the core and load new fuel into it. The machine was guided by remote control with the aid of a huge calculator. Operators fed the calculator perforated paper control tapes which contained coded instructions that told the machine which channel to load (or unload) and how many fuel rods to load it with. The machine then executed these instructions automatically.

On October 17, 1969 -- the day after Boiteux's press conference -- loading machine operators began testing a new control tape (see Figure 9.4). As far as they knew, the loading machine contained five uranium fuel rods which it was about to load into an empty channel. In fact, however, the machine contained five, slightly thicker rods filled with solid graphite. Everything went smoothly until 6:32 a.m., when the last rod from the loading machine began sliding into place. Operators were puzzled when this rod protruded from the top of the channel. They thought that the problem might lie with the automatic control system, which had been acting up a little recently. They decided to override the automatic mechanisms manually, and by 6:58 a.m., they had managed to cram the recalcitrant rod all the way into the channel.

At 7:08 a.m., the terrifying siren of the reactor's alarm system blared. Because graphite rods were slightly thicker than uranium fuel rods, that last graphite rod had blocked the flow of cooling gas in the channel. In the absence of cooling gas, the uranium rods had begun to overheat. The uranium melted the metal cladding around the rods which then fused together, producing a meltdown (though only in that channel). Fortunately, the operators soon realized that something had gone amiss. By shutting down the reactor quickly, they managed to avoid an accident on the scale of the one at Chernobyl seventeen years later.

Nonetheless, the reactor suffered considerable damage. After the fuel rods fused together, metal shards had been blasted out of the channel by a sudden burst of pressurized cooling gas. In addition to the disaster caused by a melted channel, therefore, over 100 kilograms of contaminated debris littered the structure that supported the reactor core. Furthermore, the pipes which contained the cooling gas had been exposed to radiation. Before Saint-Laurent could go back on line, the contaminated debris had to be cleaned up and the damage repaired.[126]

For Saint-Laurent employees, the accident in their plant enacted the crisis in the gas-graphite program. Repairing the reactor became their way of handling both disasters. Doing so required a complex conflation of technological and cultural work.

Site employees routinely used the word "pollution" to describe the radioactive contamination of their reactor; on the most obvious material level, this pollution threatened the proper functioning of the reactor.[127] The pollution also posed a threat to a fundamental basis upon which Saint-Laurent employees constructed their identities as nuclear workers, engineers, and managers: it cast doubt on their ability to control the reactor. Meanwhile, the decision to terminate the gas-graphite system threatened their place in the great story of French technological glory. If the gas-graphite program no longer constituted the forefront of the French nuclear program, then they were no longer at the forefront of high technology work, and therefore no longer pioneers. Finally, the accident seemed to prove that top-level EDF managers had been correct to judge the gas-graphite program unsuitable for further development, and thus to validate Boiteux's decision at the very moment of its announcement. The best way for workers and engineers to meet these heterogeneous threats was not to go on strike, but to clean up the reactor. The reactor was not only in technical danger. It was also defiled by the implication that it could not perform its electricity-production duties properly. A quick and effective clean-up would restore its functionality and its reputation. On another level, the clean-up would serve a psychological function, providing a means for employees to redeem their skills.

Thus the technological and cultural dimensions of the clean-up were inseparable. If, for example, workers failed to repair the reactor, or did so poorly, or with many casualties, then the clean-up would not help them confront threats to their cultural identities. Undertaking the most challenging clean-up operation in the history of nuclear power would reaffirm their solidarity as nuclear employees, restore their identities as pioneers, and make sense of a world with no future gas-graphite reactors.

Even before the clean-up began, site employees attempted to prescribe its meanings. When reporting upon Boiteux's infamous speech, the engineer-editor of the site's newsletter did not refer directly to the termination of the gas-graphite system. Instead, he asserted that

Clearly, the editor was trying to minimize the damage. The accident did not threaten the working principles of their reactor. Employees should temper the rage they felt about the discontinuation of their design: rage served no purpose, and now only their success in repairing the reactor mattered.

Venturing inside the reactor

Cleaning the debris under the core posed a particularly thorny problem for the engineers in charge of designing the clean-up operation because the mezzanine was completely inaccessible by any sort of passageway. Initially, they thought about building a special remote control device that they could lower down into the mezzanine through the damaged channel.[129] After considerable debate, however, they decided that such a device would cost too much and take too long to build. Instead, they chose the "solution of direct clean-up by access through the mezzanine."[130] In other words, they decided to send people directly into the space under the reactor core to clear the debris and the contamination.

The radiation level in the mezzanine was so high that engineers estimated that a single hour there would expose workers to between three and six times as much radiation as they were normally allowed in one year. Engineers decided that no single employee should spend more than 12 minutes in the contaminated zone. This limit, coupled with the extremely dangerous conditions of the work space, meant that every movement would have to be meticulously planned.

Three elements were essential in this planning: dress, motion, and space. Because of the high levels of radioactive contamination, dressing properly involved donning multiple layers of shining white garments, putting radiation detectors on every part of the body, and hooking up breathing and communication apparatus.[131] Even wearing all this equipment, however, men could not expect to stay in the work space longer than a few minutes. And much needed to be done while they were there. They had to remove the arm of a remote control device that had fallen to the bottom of the channel during a previous rescue attempt, clear and scrub the flooring and structure upon which the reactor core rested, scrub the cells around the melted channel, and more.[132] Motions therefore had to be carefully choreographed and rehearsed on a replica.[133] Finally, the space in which the "intervention" would occur required preparation. In order to give men access to this space a tunnel had to be built and the whole space outfitted for human use. This required the installation of ventilation, lighting, signals, intercoms, and television monitoring cameras. Such arrangements notwithstanding, the conditions in this space remained extremely harsh: in addition to high radiation levels, the temperature hovered at around 35 to 40deg.C (95-104deg.F) with very poor air circulation.[134]

By April 1970, these preparations were complete and the time to begin cleaning had arrived. Figure 9.5 shows the space at the entrance to the tunnel where workers prepared for their entrance into the contaminated zone. From the lock chamber, the worker crawled up through the vertical tunnel using mountain climbing pitons and other equipment. Once at his workplace, he spent roughly ten minutes performing the motions he had rehearsed so carefully in the replica. These might involve removing a chunk of debris, scrubbing a surface, or any number of other small tasks. His time up, he then towed whatever debris he had removed back down the tunnel with him, dropped it off in its designated spot, and removed the several kilos of clothes and equipment from his body. Once he left the tunnel, the next worker could enter to perform his tasks. In this fashion, workers succeeded one another for "interventions" which lasted two or three hours each. Each working day consisted of two such interventions; the entire operation took three weeks. Approximately three hundred people participated in the operation in some capacity. Around one hundred actually entered the reactor.[135]

------------Figure 9.5 here --------

Figure 9.5: Entrance area to the tunnel. In zone A, someone monitored entrances into and exits from the reactor. Workers arrived in their standard work outfits: T-shirt, jacket, pants, socks, and tennis shoes, all made of white cotton. Here they donned additional clothing: two pairs of cotton overshoes, a pair of long sleeved cotton gloves, a pair of long sleeved vinyl gloves, and a pair of vinyl leg coverings that came up to their knees. They picked up two kinds of radiation detectors (dosimeters and film badges) and proceeded to zone B. There, each worker received a mask with a filter hooked up to an air supply and equipped with a microphone and a tiny speaker to allow him to communicate with the men watching him through the TV monitors. He then fit a white cowl over his mask and added a white overcoat with a hood that fit over the cowl and mask. A team of dressers sealed the seams of his outfit with adhesive tape and stuck radiation detectors all over his body: two on his head, one on his chest, one on his wrist, one on his crotch, and an additional detector somewhere else on his body which would sound an alarm if it registered a radiation dose over 2.5 rems. Thus equipped, the worker then entered a lock chamber where he got pressurized (the reactor vessel was not at atmospheric pressure). Off to the side of the lock chamber was another set of spaces through which the equipment that the worker needed entered the chamber (and through which the contaminated debris that he removed left the reactor).[136]

The physical conditions and motions of the clean-up cannot be understood without also examining the language used to narrate and explain the process. Employees used these narratives to articulate the meanings of their motions and assert their status in the French technological adventure. The most extensive and coherent of these narratives, entitled "Great Spring Cleaning" and published in the site's newsletter, is well worth quoting in full:

The astronaut metaphor evoked the ultimate male pioneer: the man who entered a space not made for men, who crossed a frontier previously thought unattainable, who shone as a symbol for the whole world of what other men could accomplish. Mountain climbers and spelunkers were also respectably male heroes. They too performed difficult physical feats under extreme conditions, and they did so with "courage." Equating the nuclear workers with symbols of heroic masculinity simultaneously reasserted and constructed the pioneering nature of their work.

The event itself was construed as a "spectacle," an enthralling performance that captivated performers and spectators alike. The "main actor" stood at the center of the show. His actions propelled the plot forward, and his predicament generated the emotional tension. The supporting characters fussed over him and sustained him in his trial. The emotion conveyed by the performance was subtle, elusive, contained in "gestures" and "looks." But the message of community and solidarity was clear enough. The participants were bound to each other by "sympathy," "kindness," and "team spirit." They formed a team, and belonging to a community involved in a common project filled them with "enthusiasm" and "zeal." The enormity of their task might cause "fleeting apprehension," but solidarity made them fearless. These images and meanings were repeated in many accounts of the clean-up, both before and after the mezzanine intervention.[138] The solidarity evoked by the process was such that not even the CFDT, the labor union most concerned with issues of workplace health and safety, raised the slightest protest over the methods used.[139]

Clearly, cleaning up Saint-Laurent 1 was not a purely technological event. It involved transforming physical motions into culturally and politically meaningful acts. The CEA strikes brought engineers and workers together to construct alternative technological scenarios and contest the techno-economic practices of light-water proponents. The Saint-Laurent clean-up brought (a different group of) engineers and workers together to construct alternative meanings for the termination of the gas-graphite program and for their role in the national order.

Twenty years later, the ways in which workers talked about the clean-up show the uniqueness of this experience for them and how deeply that experience etched some of these meanings in their minds. The associations between the accident and the abandonment of the gas-graphite reactor remained clear for all of them. As one worker put it succinctly, the accident came at a "politically unfortunate" time.[140] Another man mentioned a rumor, which circulated right after the accident, that Boiteux's announcement had indirectly caused the accident by making workers too jittery to concentrate properly.[141] In retrospect, the men involved experienced the clean-up as the last hurrah of the gas-graphite program, the last time they felt special. It marked the moment when everything changed.


I moved to France for the first time in 1975. One of my most vivid memories of the cultural landscape from that period was an advertising slogan that seemed to be everywhere. On suburban billboards, in newspapers and in magazines, on the radio and on television, the words appeared: en France on n'a pas de pétrole, mais on a des idées ("in France, we may not have oil, but we have ideas"). My parents and I found this a wonderful phrase. Repeating it and adapting it to different circumstances became one of our favorite games. At the time I neither knew nor cared that the phrase was part of EDF's ad campaign for its new nuclear program, whose other slogan was tout éléctrique, tout nucléaire ("all electric, all nuclear").

Just five years earlier, the guerre des filières had ended with a decision to build light water reactors with an American license. Between 1970 and 1973, EDF broke ground on four Westinghouse-licensed reactors -- a "modest" number, as prescribed by the 1970 PEON report. Any desire to remain modest was nipped in the bud by the 1973 oil crisis. In March 1974, Prime Minister Pierre Messmer announced a new energy plan calling for the launch of thirteen 1000 MW light water reactors in two years. By the time I began my research in 1989, France was obtaining between 75 and 80 percent of its electricity from pressurized water reactors,[142] and engineers were eager to tell me how the light water system had become francisé -- Frenchified.

The guerre des filières and its outcome were significant in many ways.

Most obviously, terminating the gas-graphite program and buying a license from Westinghouse involved a profound rearrangement of industrial and institutional relationships. This had deep repercussions for reactor designers, builders, and workers, whose roles and skills had to change in order to accommodate the licensing agreements and the new technology. The licensing agreement specified work and safety guidelines which blended with existing practices. The new prescriptions for work practices affected not only those operating the new reactors but also those in the older gas-graphite reactors (the last of which continued to operate through the early 1990s).[143] And as the new reactors went up in the 1970s, the first wave of anti-nuclear protesters contested their very existence.[144] Clearly, such changes would be worthy of a book themselves.

The triumph of the light water design marked the triumph of the "economist-managers"(Frost's term) or the "cost-benefiters" (Jasper's term): men who measured technological success by purely economic criteria. But as we have seen, choosing light water over gas graphite was, itself, far from a simple, purely economic process. Neither was it purely technological or "purely" political. Instead, it was all of these. Choosing light water involved a technopolitical reformulation of the standard for French glory and nationhood. It also meant advocating a France that would evaluate itself in terms of comparative economics -- a France measured according to a scale whose increments were set not by France alone, but by international institutions and conglomerates. Choosing a new technology not only symbolized but also constituted the choice of a new society, one that would be interdependent rather than independent.

Perhaps ironically, the victors' strategy for technopolitical success was to separate (discursively) technology from politics. In a sense, this meant (re?)inventing a technological determinism: creating a context in which there was such a thing as the best technology, asserting that the best technology would lead to the best future, and especially, defining new standards by which "best" would be judged. This process involved defining some elements of reactors systems as "technological" or "economic," and others as "political." It followed from this separation that the "political" elements were secondary to the "technological" or "economic" ones. The separation constituted a way of defining which elements should count in making a choice and which should not; which criteria were relevant, and which were not. In the course of this process, one entire system came to be defined as "political," the other "economic."

Retrospective accounts of 1969 reflect these fault lines. Here is a typical statement:

The gas-graphite system was thus the political system. The light water system was not.

But the effort to separate technology and politics during the guerre des filières was neither neat nor successful. Truly separating technology and politics required program leaders to disentangle the gas-graphite system from French identity:

In this interpretation, questioning the nationality of these systems was thus both a strategy for depoliticizing the choice and a means of undermining the legitimacy of the gas-graphite system as a symbol of French glory. Witness the response of a former CEA official to Simmonot's question about the "French system":

Rather than portray British gas-graphite reactors as similar to French ones, this speaker alleged that the French "copied" elements of the British system -- a far less glamorous picture, which subverts the technical, and therefore the symbolic, value of the gas-graphite system. At the same time he gave two seemingly purely technical reasons for the failure of the gas-graphite system. Thus the move to separate technology and politics was closely tied with efforts to disentangle the gas-graphite system from French identity and to create a technologically determinist explanation and Zeitgeist.

If the move to separate technology and politics succeeded, however, it did so only at one discursive level. For in a sense, the very effort to disentangle the gas-graphite system from French identity meant that the discourse of nationalism continued to matter in the nuclear program. During the guerre des filières, French identity was not removed from reactors altogether; instead, it was transferred onto breeder reactors. Similarly, later arguments in favor of change involved claims that even if purchased from the Americans, the light water system could be "Frenchified" (and today engineers and managers claim that the system has indeed been francisé). The emphasis had merely shifted, from making a French technology to making a technology French.

Furthermore, as we have seen, the effort to separate technology from politics was by no means uncontested. In a variety of ways, unionized CEA employees, Saint-Laurent workers and engineers, EDF labor leaders, and groups of design engineers in both EDF and the CEA all protested not just the light-water design, but the exclusion of politics from technological choice. Taken together, these protests themselves manifested the hybrid nature of technological activity. These various (and overlapping) groups employed hybrid means of enacting their protests: strikes which used economic evaluations of programmatic choices, a clean-up which enacted the cultural and political meanings of nuclear work.

If the effort to separate technology and politics was itself a strategy to gain dominance over programmatic choices, resisting that dominance involved resisting the separation. The technologies and practices of the nuclear program remained what they had always been: hybrid entities through which men wrestled for control over their lives and their nation.

[1]Speech by Edmond Alphandéry, April 5, 1996 at the XIe Colloque historique de l'AHEF entitled "La Nationalisation de l'Electricité: Nécessité Technique ou Logique Politique?" Georges-Henri Soutou discusses the standard story of the battle of the systems in "La logique d'un choix: le CEA et le problème des filières électronucléaires," Relations Internationales hiver, 68 (1991): 351-377. Versions of this story can be found in Peter Pringle & James Spigelman, The Nuclear Barons (New York: Holt, Rhinehart and Winston, 1981) and Philippe Simonnot, Les nucléocrates (Grenoble: Presses Universitaires de Grenoble, 1978); more commonly, this is the story that circulates among the technological and industrial communities in France. Several scholars have debunked various aspects of this myth, including Soutou, cited above; James Jasper, Nuclear Politics: Energy and the State in the United States, Sweden, and France (Princeton: Princeton University Press, 1990); Frost, Alternating Currents.

[2]One former CEA agent was both surprised and skeptical when I assured him that contesting the abandonment of the French system had been a major purpose for the 1969 strikes.

[3]For more on these various projects see Soutou, Lamiral, Pierre-Henri Floquet,
*; Bupp & Derian.

[4]In 1964, for example, the PEON commission -- a government advisory group composed of officials from EDF, the CEA, and private industry -- issued a report speculating that light water reactors might have lower investment costs than gas-graphite reactors; it emphasized, however that pursuing the light water option would mean either depending on America to supply enriched uranium fuel, or building an expensive enrichment plant that would negate the still uncertain cost advantage of the light water system. France, therefore, should stay on the gas-graphite track (Simmonot, pp. 237-245). In forwarding this recommendation to prime minister Georges Pompidou, Gaston Palewski (the minister in charge of atomic affairs) urged him to make a rapid decision to engage in a "massive" development of gas-graphite reactors. The consequences of the choice, he said, mattered not only for France but also for the rest of Europe. Were France to give up its native reactors, "le retard technique et technologique qui s'ensuivrait ferait sentir rapidement ses effets dans notre economie comme dans notre politique." (G. Palewski to Georges Pompidou, 4 juillet 64. CEA archives, box F3-24-25). Pompidou approved the PEON plan.

[5]This argument is made by Bupp & Derian, p. 49 passim. Many other writers have supported this analysis, including Frost, Jasper, Saumon & Puiseux, Pringle & Spigelman, and several of Simmonot's interviewees assert that light-water reactors ended up costing far more than the initial estimates suggested.

[6]It's clear that EDF managers and engineers had been making unofficial inquiries into this possibility since at least 1965. Letter from Bienvenu to Melese, 30 december 1965. CB1.25.

[7]Decelle to Hirsch, 7 march 1966. CB66.8.

[8]"Groupe de Travail Commun CEA-EDF sur les Filières à Uranium Enrichi", 4 mai 66 and note from A. Decelle to R. Hirsch, 4 mai 66. CB66.11.

[9]CA # 240, 27 mai 66, p. 37. "ces informations....sont plus spectaculaires que fondées, car, bien entendu, il n'a jamais ete question nulle part d'un abandon quelconque de la filiere uranium naturel, gaz, graphite."

[10]CA # 244, 25 nov. 66; CA # 249, 10 mars 67.

[11]Nicholas Vichney, "La centrale nucléaire EDF3 de Chinon est arrêtée pour six mois," Le Monde (2 dec. 1966), p. 7; Vichney, "Les incidents survenus à la centrale de Chinon amènent à poser le problème des rapports entre l'Electricité de France et le Commissariat atomique," Le Monde (24 janvier 1967), pp. 1, 17.

[12]"Le crêpage de Chinon," Le Canard Enchaîné (7 dec 1966), p. 2. "Bref le matériel nucléaire bien de chez nous ne tient pas le coup....Des têtes vont rouler, citoyens!"

[13]CA 244, 25 nov 1966.

[14]In fact, Vichney at least had mentioned all these points, but only in passing; evidently he did not foreground them in the way that EDF's leaders would have liked.

[15]CA 245, 22 dec 1966. "atteinte portée au prestige de l'Etablissement."

[16]In an effort to reassure the personnel, Massé wrote a letter in January 1967 in which he confirmed that EDF's nuclear teams had carried out their duties with "un dévouement exemplaire." Building reactors was a difficult task, he affirmed, and getting more so all the time. But he also warned that the personnel could expect some structural change within the institution, which, he hastily added, was only normal in such a large establishment. The Direction de l'Equipement would soon receive union representatives to discuss these changes. Pierre Massé to André Decelle, 12 jan 1967. CB papers.

[17]"Communiqué commun CEA et EDF," 25 janvier 1967. CEA archives, F3-24-25.

[18]"Un communiqué intersyndical à la Presse, après le récent article de M. Nicolas Vichney," memorandum (unsigned) from département de sûreté et de protection du secret, 22 fev 68. CEA archives, F3-24-25.

[19]Letter Hirsch to Hubert Beuve-Mery, 24 avril 1968. Beuve-Méry responded in a conciliatory tone in Beuve-Mery to Hirsch, 30 avril 1968, but he did not agree to publish a correction. CEA archives, F3-24-25. "manque de bienveillance coutumier du redacteur [Vichney] en face des difficultes inherentes au developpement de l'en nucl...cette publicite, sur un incident qualifie de mineur dans l'article lui-meme, ne peut que compliquer la tache de l'industrie fcse, actuellement chargee de la construction d'un reacteur de meme type a l'etranger."

[20]Transcript of extract of emission TV: diffuse le 7 mai 1968 a 22h05 dans la serie "Demain Commence Aujourd'hui" d'Henri Polad et J. C. Bollardot. Enquete on prochaine evolution des necessites d'energie electrique. CEA archives, F3-24-25. "`graves' dans la mesure ou elles risquaient de mettre en cause a la fois la competence de l'ind fcse et la comp des ing's des differents organismes qui ont ete appeles a construire ces differentes centrales. Il s'agissait donc d'une question d'interet national."

According to one of the CEA's public relations officers, Vichney had developed a personal vendetta against the CEA. The journalist had called one morning to inquire about a plan that Le Monde had to devote a full page to Phénix, the breeder reactor whose construction had recently been decided. Upon learning that the CEA had decided to abandon the article plans, Vichney turned hostile. According to the officer, ""C'est donc la guerre," repartit aigrement V, "et c'est pour cela que je n'ai pas recu de lettre d'invitation a votre prochain dejeuner. Eh, bien, nous allons rire!" Sur ce dernier membre de phrase, la voix se fit plus sarcastique....Je fis remarquer que ce dejeuner etant une affaire plus amicale que professionnelle, une absence d'invitation ne signifait pas une rupture des relationss professionnelles. "Je le pense bien" repliqua Vichney d'une voix tres grincante. "D'ailleurs je ne serai pas venu a ce dejeuner, car j'ai considere comme intolerables les paroles peu aimables de M. Goldschmidt a mon egard au cours de la precedente rencontre." Vichney apparently accused the CEA of not following through on its promises to give Le Monde special access to its news and promised "lendemains qui chantent." According to the CEA's public relation service, Vichney had been insisting on obtaining exclusive scoops on all of the CEA's major projects. But when the CEA granted those scoops, bitter feelings often resulted because Vichney would do nothing with the information that he had received, even after CEA scientists and engineers had taken tremendous trouble and time to arrange exclusive meetings and interviews. In sum, "Les incessantes attaques de NV contre le CEA, sa polemique avec les syndicats, le dev't de sa campagne d'aide aux projets de M. Ambroise Roux, et aux menaceurs d'eclatement du CEA, ont rendu progressivement difficile la collaboration peniblement elaboree." Correspondence between Goldschmidt, Renou and Benoit, Service des Relations Publiques, 14 mai and 22 mai 1968. CEA archives, F3-24-25.

[21]CA 249, 10 mars 67, p. 18.

[22]ibid. "une instance judiciaire dont l'issue ne pouvait être que lointaine et incertaine"

[23]"accorder une prime a la mediocrite." ibid., p. 21.

[24]ibid., p. 29. "politique d'accroissement rapide des puissances de rech de caracteristiques nouvelles, a permis d'arriver, en 4 reacteurs, a la competitivite, alors que nos amis anglais, meme s'ils ont pour le moment une disponibilite superieure a la notre, n'y sont pas arrives en 16 reacteurs.

"La competitivite a laquelle on est ainsi arrive, il ne pense pas que les incidents d'EDF 3 la remttent en cause. Ce qu'ils vont couter, en effet, ne represente pas un pourcentage tres considerable de la depense initialement prevue.

"Il faut arriver maintenant a augmenter la disponibilité du materiel et, pour cela, faire un grand effort dans le sens de l'amelioration de la technologie. C'est la lecon qu'il faut tirer avant tt de cette malheureuse affaire.

"Il faut, dans l'avenir, en liason avec les constructeurs, renforcer les specifications de tous ces materiels et, surtout, renforcer les controles tant a l'echelon du constructeur qu'a l'echelon d'EDF....

"[I]l faudrait trouver le moyen de lier les constructeurs aux indisponibilites, non pas leur en faire supporter toute la charge mais... afin de les rendre solidaires et conscients a l'egard du materiel qu'ils construisent."

[25]CFDT, Federation EGF. "Direction de l'Equipement: La Situation de l'Equipement Nucleaire," 5 mai 67. CB papers. "attitude ambigüe de faiblesse d'EDF vis-a-vis des Constructeurs dans l'applications des contrats" These same tensions were being played out in the internal restructuring of the utility. Managment tried to streamline and centralize reactor design and operation by reorganizing various divisions. Unions interpreted these reorganizations as an effort to further reduce the responsibilities of rank and file employees. Cite piles of documents from CB papers.

[26]ibid. "entreprises nationalisees qui representent un acquis social et economique auquel nous tenons tous."

[27]Jean Cabanius, "Rapport du Groupe de Travail Placé sous la responsabilité de Monsieur Cabanius (EDF)" (25 janvier 1967), p. 2. Courtesy of Alain Beltran. "Les considérations politiques portant notamment sur l'approvisionnement ou la fabrication en F d'ur enrichi dépendent des Pouvoirs Public et ne seront pas évoqués dans cette étude. Le rapporteur s'est strictement limité à l'aspect industriel, technique, et économique."

[28]J. Horowitz, Dir. des Piles Atomiques, CEA, "Examen des filières electro-nucléaires dans le contexte français actuel" (1 février 1967), p. 1
*. Courtesy of Alain Beltran. "exposer les enseignements que l'on peut tirer du programme en cours d'exécution"; "apprécier à nouveau les possibilités de la fil gg."

[29]Cabanius, "Rapport du Groupe de Travail," pp. 4-5.

[30]Horowitz, "Examen des filières electro-nucléaires," p.

[31]Cabanius, "Rapport du Groupe de Travail," p. 9. "Il y donc dans la fil u-n-g-g, une source d'incidents qui peuvent etre tres lourds, non pour la securite des personnes, mais sur la longueur et la frequence des arrets, et par suite sur le coefficient de disponibilite a attribuer a une series de centrales integrees dans un complexe energetique."

[32]Cabanius, "Rapport du Groupe de Travail," p. 10. "Cette formule laisse a la charge d'un constructeur d'ailleurs puissant les incertitudes techniques. Son acceptation est certainement fondee sur une grande confiance dans la qualité technique de la fourniture. La tendance qui s'est ainsi instaurée sera probablement irréversible."

[33]Horowitz, "Examen des filières electro-nucléaires," p. 10. "Les catalogues relatifs a la `chaudiere nucl' sont precis en ce qui concerne la limite de fourniture, mais n'indiquent pas une decomposition des prix par composante; ils ne permettent donc pas une analyse technico-économique détaillée. De toute manière, comme le reconnaissent les promoteurs, ces prix ne correspondent pas au cout de quelques centrales prises isolement car c'est anticipant l'effet de serie que la percee du nucleaire a pu etre realisee aux E-U"

[34]They based their calculations of the cost of the pressurized water kilowatt hour on a meeting they had with Framatome, the company that had managed the construction of the Chooz reactor under license to Westinghouse. They had also tried to get figures for the cost of building a boiling water reactor, but no French company or consortium had a license with General Electric yet. Memorandum from Jules Horowitz to the Administrator-General, "Conditions de construction de centrales à eau ordinaire par l'industrie française. Reunions du 7 juillet avec Framatome-Westinghouse, du 13 juillet avec Alsthom-GECO." 27 July 1966. CEA archives, F 6-13-20.

[35]Only in the appendices to his report did Cabanius discuss (in half a page) the fact that the American estimates were extrapolated from a very limited operational experience. He also used the appendices to elaborate, in over two pages, on the technical uncertainties of the gas-graphite system. "Annexe III: les données actuelles et les perspectives futures des centrales classiques et nucléaires," part of Cabanius, "Rapport du Groupe de Travail."

[36]Horowitz, "Examen des filières electro-nucléaires," p. 26. "Des arguments g'aux d'ordre tech et econ ne fournissent pas des elements suffisants pour un choix entre PWR et BWR; celui-ci devrait donc decouler de considerations propres aux contexte fcs et en particulier des consequences qu'aurait dans l'un ou l'autre cas le recours a des licences americaines. Il convient de veiller a ce que l'Ind fcse, en attendant une decision, ne s'engage dans un effort et des liens couteux et eventuellement inutiles ou prematures."

[37]Horowitz, "Examen des filières electro-nucléaires," p. 25. "La poursuite du programme adopte par le Veme Plan se jusifie... pleinement. Elle doit en effet assurer, en plus de la reduction du cout du kWh, une economie appreciable en devises, la securite d'approvisionnement pour le combustible, la valorisation d'investissements importants effectues par le CEA, l'Industrie, et l'EDF et la production en toute independance du Pu necessaire aux programmes civils et militaires."

[38]Cabanius, "Rapport du Groupe de Travail," p. 25. "permettra a nos constructeurs regroupes d'affirmer leur valeur tech, d'acquerir des references qui ont un poids tres grand pour l'exportation et pour les accords avec d'autres constructeurs europeens et de participer a la grande confrontation industrielle de la prochaine decennie."

[39]memorandum, P. Ailleret to J. Renou, R. Hirsch, F. Perrin, P. Massee, and A. Decelle, 20.2.67. "Au Dela du Consensus Cabanius-Horowitz," CEA archives, F3-24-25.

[40]Direction de l'Equipement, "Position d'EDF devant les probleme nucleaires actuels (mise à jour juillet 1967 du texte du 29 mai)," CB papers. The franco-belgian and franco-swiss projects were discussed in EDF CA 254 (22.9.67) and EDF CA 257 (22.12.67). De Gaulle signed the official authorization for the franco-belgian project in December 1967: Présidence de la Republique, Secretariat Général. 8 dec 67. "Conseil restreint du 7 dec 67 relatif au programme nucléaire civil. Relevé des Décisions." CEA archives, F3-24-25.

[41]Direction de l'Equipement, "Position d'EDF devant les probleme nucleaires actuels (mise à jour juillet 1967 du texte du 29 mai)," CB papers. "Fondee, en tant qu'Etablissement industriel et commercial, a defendre son point de vue quant aux considerations economiques touchant a ses interets et a ceux de l'industrie fcse -- dont elle se sent solidaire en tant qu'Etablissement public -- l'EDF ne pourrait cependant que s'incliner devant les considerations d'ordre politique qui s'opposeraient a ce projet."

*check sources and offer a variety: Simmonot, JFP interview with Boiteux, Massé. See also Goldschmidt on this?

[43]Simmonot, #9, p
*. "Schumann, ce n'est pas un économiste, mais un homo politicus type. Palweski était du même genre, et il était lui aussi à fond pour la ff: c'est lui qui l'a vendue aux Espagnols. Schumann ne raisonnait pas dans les termes de l'ind. int'le. In ne reconnaissait pas le fait international."

[44]Transcript of interview with Maurice Schumann conducted by Alain Beltran, Martine Bungener, and Jean-François Picard on June 1, 1981. Transcript courtesy of Jean-François Picard. "Après avoir très attentivement étudié le dossier, j'ai attribué une priorité absolue au surrégénérateur."

[45]ibid. "l'abandon pur et simple de la filière française ne se justifiait pas; en tout état de cause, avant la francisation des filières à eau légère, qu'il s'agisse du PWR ou du BWR."

[46]ibid. "a eu -- j'allais dire: des complices -- des avocats à l'intérieur du CEA, qui invoquaient comme arguements contre la filière française, les dangers qu'elle comportait. Or, les études auxquelles j'ai fait procéder démontraient que les accidents et les retards étaient au moins aussi nombreux dans les filières étrangères." See also Maurice Schumann, "La politique électronucléaire de la France," speech given to the meeting of the groupe X nucléaire, 6 March 1968. CEA archives, DEDR-DIV 219 DPA.

[47]Transcript of interview with André Decelle conducted by Jean-François Picard and Alain Beltran on May 4, 1981. Transcript courtesy of Jean-François Picard.

[48]EDF, CA 253, 12.9.67.

[49]"M. André Decelle, directeur-général de l'EDF donne sa démission. Partisan de l'exploitation de la filière uranium enrichi, il était en désaccord avec le gouvernement." Le Figaro, 12 sept. 1967. Find p #?

[50]EDF, CA 253, 12.9.67, p. 9. "graves contre-verités." "determiner avec le maximum d'objectivite et dans un esprit scientifique ou se trouve, dans cette affaire, l'interet du pays." "particulierement choque de voir cet effort se transformer, pour l'opinion, en une espece de conflit passionel."

[51]Soutou discusses this part of the story at greater length in

[52]If need to support or articulate more, see EDF CA 257, 22.12.67 (#s 36, 38)

[53]See most of the interviews in Simmonot, Les nucléocrates. Also Hecht interview with Marcel Boiteux.

[54]EDF CA 257, 22.12.67.

[55]Anonymous memorandum, Ministère des Finances, "Propositions de questions à inscrire à l'ordre du jour de la Commission Consultative pour la Production d'Electricité d'Origine Nucléaire," 15 mars 1967. CB papers.

[56]Commission Consultative pour la Production d'Electricité d'Origine Nucléaire, Groupe de Travail Général, "Rapport de Conjoncture" (no date); "Prix des fuels a moyen et long terme" (9/11/67); "Hypotheses de travail" (27.9.67); "Hypotheses de travail concernant le developpement nucleaire" (8.11.67); réunion du 15 sept 1967, "Compte rendu de l'acitivite du gpe de trav gal" (redige par Jacques Gaussens). All in CEA archives, F6-13-20.

[57]Groupe de Travail Général, "Rapport de Conjoncture" (no date).

[58]Memorandum, Pierre Tanguy to M. le Directeur des Piles Atomiques (Jules Horowitz), "Observations sur le Projet de Rapport soumis à la Commission le 29 février," 28.2.68. CEA archives, DEDR DIV 219, DPA.

[59]Memorandum, Pierre Tanguy to M. le Directeur des Piles Atomiques (Jules Horowitz), "Observations sur le Projet de Rapport soumis à la Commission le 29 février," 28.2.68. CEA archives, DEDR DIV 219, DPA.

[60]Memorandum, Jules Horowitz to the Administrateur Général, 23.2.68. CEA archives, DEDR DIV 219, DPA. "La variation des prix aux E-U, les sacrifices que l'AEG et Siemens reconnaissent avoir du faire recemment pour obtenir leurs 1eres gdes commandes et les difficultes que rencontre l'ind belge dans les affaires de Doël et de Tihange illustrent bien la distinction qu'il convient de faire entre le cout reel d'une entreprise et le prix qu'il est nec de consentir pour s'imposer sur certains marches."

[61]JB/MW, le 6 nov 68. "Remarques concernant le rapport PEON d'avril 1968". Anon, though Mr. Vendryes is handwritten at top. Clearly written by someone at CEA, though. CEA archives, DEDR DIV 219, DPA. "En concl de lecture on constate, comme d'ailleurs cela est bien connu:

"1. que les differences entre fil's sont de l'ordre de grandeur des incertitudes.

"2. que l'on `pousse' l'eau leg et que l'on `fait un sort' au gg sur les bases economiques suivantes: ce que l'on gagne sur les investissements sera tres certainement perdu sur le comb et le seul parametre qui fait pencher la balance est un cout d'exploitation plus faible pour l'eau leg (21,7 F/kW-an c'est a dire 0,3 c/kW-h au lieu de 33,9 F/kW-an c'est a dire 0,5 c/kW-h).

"On peut se poser la quest de savoir si l'element de decision reel n'est pat tt simplement une grande difference de confiance dans la fiabilite des deux fil's (ce qui vient de l'etranger semble avoir toujours bcp plus d'attraits pour les esprits fcs mais attn aux reveils penibles)"

[62]Massé, Aléas, p. 136. CHECK!

[63]Letter from Ambroise Roux to Jean Couture, 29 mars 1968 (signed also by Baumgartner, Blancard, de Calan, Gaspard, Glasser, Jouven, Malcor), accompanying "Note pour la Commision PEON", 28 mars 1968. CEA archives, DEDR DIV 219, DPA. "Pour s'en convaincre, il suffit d'imaginer ce que serait la position de l'industrie fcse au sein d'un Marche Commun ou, le nucl ayant reussi, l'industrie allemande dominerait ce secteur."

[64]Reported in memorandum, Pierre Tanguy to M. le Directeur des Piles Atomiques (Jules Horowitz), "Observations sur le Projet de Rapport soumis à la Commission le 29 février," 28.2.68. CEA archives, DEDR DIV 219, DPA.

[65]Memorandum, Jules Horowitz to the Administrateur Général, 23.2.68. CEA archives, DEDR DIV 219, DPA. "il a ete pratiquement impossible de faire prendre en consideration de façon concrete le contexte national, la continuite technique, les dangers de la dispersion et du surequipement dans un marche qui restera longtemps assez etroit, bref le cout veritable au niveau du pays, sans parler de souci de creer une Industrie nucl fcse vraiment majeure pouvant traiter d'egale a egale avec les plus grandes Stes europeennes. J'ai essaye, a plusieurs reprises, de provoquer une discussion sur ces importants prob's industriels; les representants de l'Ind qui siegent a la Comm demeurent sur une prudente reserve."

[66]Memorandum, Pierre Tanguy to M. le Directeur des Piles Atomiques (Jules Horowitz), "Observations sur le Projet de Rapport soumis à la Commission le 29 février," 28.2.68. CEA archives, DEDR DIV 219, DPA. "opposition venant aussi bien des industriels qui refusaient de fournir la moindre donnee et du Comm du Plan qui a toujours marque sa preference pour des schemas abstraits multiples."

[67]Quoted in Simmonot, p. 248. "Il est vain d'espérer atteindre...une indépendance totale.... On peut définir le potentiel d'ind. économique comme la capacité de maintenir à long terms et sur le plan international, la compétitivité de l'économie sans que cette capacité se trouve à la merci de centres de décision dont la collectivité nationale ou plurinationale, ne serait plus maître."
*check actual report!

[68]Picard et. al., pp. 199-200.

[69]Editorial, "Le Rapport Couture," Revue Française de l'Energie 201 (mai 1968), p. 435. " raison essentielle de cette unanimité tient, croyons-nous, dans le fait que des hommes de bonne foi, venant des horizons les plus divers, devainet finalement tomber d'accord dans l'analyse d'une quesiton aussi complexe, du moment qu'elle était entirèrement dépolitisée, et ramenée à l'analyse objective des vrais problèmes qui se posaient."

[70]CB/HB, "Les Filières Nucléaires," 4 juillet 68, p. 10. CB papers, CB68B.11. "SL n'aura servi à rien! La fil g-g, qui etait a meme de profiter au maximum de l'experience accumulee et, peut etre, de lutter avec qq chances de succes contre la fil amer, va se trouver balayee comme un fetu."

[71]CB/HB, "Les Filières Nucléaires," 4 juillet 68, p. 12. CB papers, CB68B.11.

[72]Claude Tourgeron, "La production d'électricité d'origine nucléaire en France," économie et politique janvier 1969, p. 2. "constitution de stes nat'les qui affranchiraient cette industrie des pressions conjuguées des grands monopoles capitalistes et des états-majors militaires."

[73]Tourgeron, "La production d'électricité d'origine nucléaire en France," p. 13.

[74]MB/CH, "Politique des Reacteurs Nucleaires," p. 6. CB papers. "la filière de l'avenir." "sans nul doute," "reprend son souffle dans l'attente d'une nouvelle percee -- celle des surregenerateurs -- a laquelle elle consacre tous ses efforts de recherche-developpement."

[75]MB/CH, "Politique des Reacteurs Nucleaires," p. 7. CB papers. "melant l'intelligence fcse a l'experience amer, de construire un reacteur `francisé'."

[76]memo, CEA, EDF, "Programme d'action dans le domaine des centrales electronucleaires," 21 avril 1969. Approved by both Boiteux and Hirsch with identical letters to each other, dated 24 and 25 avril respectively. CB papers.

[77]EDF CA, 274, 25.4.69, p. 9. "La realisataion du 1er r a eau legere se ferait dans le cadre d'une licence gale, de facon a tirer des Amer's le maximum de connaissances sur la fil que l'on aura choisie, mais sans aller cependant jusqu'a la garantie conjointe du licencieur et du licencie."

[78]EDF CA, 274, 25.4.69, p. 14. "Ce n'est pas sans une certaine tristesse qu l'on voit l'AEG et Siemens reunis placer une centrale en Hollande, alors que l'industrie nucl a demarre en Allemagne bcp plus tard qu'en F. Il est temps certainement que les pays se reprenne s'il ne veut pas etre depasse et domine."

[79]memo Cabanius to Directeur de la REN1, Chef du SEPTEN, Chef du SEGN, 29 nov 68; memo Direction de l'Eq to Directeur de la REN1, Chef du SEPTEN, Chef du SEGN, no date, probably late 1968; SETPEN memo, DG/MGo, "Centrale Nucleaire a Eau Legere, Organisation du travail SEPTEN-REN1," 13 dec 1968. CB papers.

[80]Simmonot, pp. 254-259.

[81]De Gaulle resigned after losing a referendum vote in April 1969; this loss is widely interpreted as an aftershock of the 1968 strikes. For more on French government politics in this period, see Robert Gildea, France since 1945 (New York: Oxford University Press, 1996); Serge Berstein, La France de l'expansion. I: La République gaullienne, 1958-1969 (Paris: Seuil, 1989); Jacques Chapsal, La vie politique sous la Ve République. Tome 1: 1958-1974 (Paris: Presses Universitaires de France, 1981, 1993).

[82]EDF, CA 276, 27.6.69, p. 17. "sont payees depuis longtemps par le contribuable"

[83]EDF, CA 276, 27.6.69, p. 23. "tous les chiffres, cites au cours de la discussion, et qui s'appuient sur les resultats d'experiences, de realisations et d'appels d'offres recents, prouvaient que la fil a eau ordinaire etait la plus economique et qu'elle etait la moins couteuse en investissement. C'est la raison pour laquelle elle a ete choisie."

[84]"L'avenir du Commissariat à l'Énergie Atomique," Cadres et Profession (Mensuel de l'Union Confédérale des Ingénieurs et Cadres CFDT) 234, Juin (1969): 12. "Les difficultés auxquelles le Commissariat doit faire face aujourd'hui ne proviennent nullement d'echecs techniques, mais d'un manque de politique de la recherche et de politque industrielle du gouv."

[85]"L'avenir du Commissariat à l'Énergie Atomique," Cadres et Profession : 13.

[86]Westinghouse Electric International Company, "Westinghouse et un projet européen pour l'industrie de la construction electrique," 24 janvier 1969. Courtesy of Jean-Claude Zerbib.

[87]"L'avenir du Commissariat à l'Énergie Atomique," Cadres et Profession : 13.

[88]CGC, CFDT, CGT, CGT-FO, SPAEN, "Pour un Politique Française de l'Energie Nucléaire (Déclaration des Organisations Syndicales du Commissariat à l'Energie Atomique)," 8 Octobre, 1969. Flyer. Laponche papers. "soumises a des critiques systematiques et qui nous paraissent non fondees d'une partie de la presse, encouragee par le silence eloquent des directions du CEA et de l'EDF."

[89]ibid. "ne tient pas, contrairement a ce qui est ecrit quotidiennement dans la presse, dans le prix eleve des centrales nucleaires fcses, mais d'une part dans les prix de dumping realises par les petroliers sur les fuels d'autre part dans la structure actuelle de l'industrie electromecanique fcse en gal et de l'ind nucl en particulier."

[90]ibid. "doit se traduire tout d'abord par un dev de la fil gg dont la tete de serie, S.L. I, fonctionne dans d'excellentes conditions depuis son couplage au reseau et se situe en tete de toutes les centrales thermiques fcses."

[91]"Un passé qui est un exemple," Le Compagnon d'Energies Nouvelles (Journal des Ingénieurs et Cadres CGT du CEA) 7, supplément ndeg. 143 (1969): 1. "rétablir la verité indispensable a la defense de l'energie atomique fcse... Les sections d'ing et cadres CGT ne ménageront pas leurs efforts pour que la France reste independante en matiere d'energie."

[92]Les sections d'entreprise PSU de l'énergie atomique, "un seul adversaire, un seul combat," 1969. Flyer, Laponche papers. "Ce qu'on appelle `guerre des filieres' est un attrape-nigaud. Il s'agit en fait d'une guerre entre trusts internationaux orchestré par l'un d'eux: Westinghouse... Que signifie l'intrusion de West. dans le monde fermé des affairistes bourgeois et technocrates qui nous gouvernent sinon la manifestation brutale de l'imperialisme americain parmi nous. Ailleurs, il tue par la guerre, ici il cherche a nous reduire a l'etat de colonie economique. Ne soyons pas dupes: dans cette affaire, le gouv fcs n'est pas neutre. Il est complice. L'adversaire des travailleurs, c'est lui."

[93]A memorandum from Hirsch to the Minister of Industry provides further confirmation that Hirsch and Boiteux wowrked out the American light water plant together. The memo outlines their recommendation for how the decision should be shaped and worded. Hirsch to M. le Ministre du Dev't Industriel et Scientifique, 15 oct 1969. CEA archives F3-24-25.

[94]Nicholas Vichney, "Abandonnant la filière française, l'EDF affirme sa volonté de construire des centrales nucléaires de type américain," Le Monde, 18 october 1969, p. 1. "Mais à cette attitude réaliste s'opposaient l'espoir de trouver dans les centrales nuléaires le support d'une indépendance énergétique et le désir du commissarit à l'énergie atomique de faire exploiter le type de centrale qu'il avait mis au point."

[95]Pierre Juin, "Une réussite de la technique française...La centrale nucléaire de Saint-Laurent-des-Eaux risque de n'avoir pas de descendance," L'Humanité, 18 october 1969, p. 1. "presitigieuse réalisation française."

[96]Pierre Juin, "En moins de sept mois d'activité la production d'énergie de `Saint-Laurent-1' a dépassé un milliard de kilowatts-heures," L'Humanité, 18 october 1969, p. 6. "les états-majors d'Electricité de France et du Commissariat à l'Energie Atomique qui pilotaient les journalistes spécialisé, jeudi dans le vaste chantier de Saint-Laurent afficheraient une mine réjouie. Eh bien non." "Dans sa conférence de presse lundi dernier, M. Ortoli, ministre du développment industriel et scientifique, avait déclaré que la politique nucléaire de la France serait fixée à la fin de l'année....M. Boiteux, pour sa part, n'a guère pu dissimuler que la cause était d'ores et déjà entendue." "Au cours d'un entretien éclair, qui ne permit pas aux participants de poser plus d'une demi-douzaine de questions, M. Boiteux a affirmé que tous les pays s'orientent désormais vers les réacteurs à eau légère et que, par conséquent, nous aurions mauvaise grâce à nous obstiner à pratiquer notre propre technique dans l'espace restreint de l'hexagone français."

[97]Le Canard Enchaîné, 22 october 1969, p. 3.

[98]Le Canard Enchaîné, 22 october 1969, p. 4. "Pompidou se hâte lentement de ne rien décider. Pour l'instant, il en est à l'étape de mise en condition. Parce qu'enfin, la filière française, c'était un gadget du général. Faut manipuler ça avec délicatesse. Des fois que le cher vieux monsier se fâcherait."

[99]EDF CA 278, 24.10.69. "regrettable que l'on ait donne a ses propos la signification politique que l'on sait."

[100]ibid. "il n'y avait pas de raison de regretter ce qui avait ete fait dans ce domaine, les efforts consacres a la fil "gaz-graphite" etant dans la logique de l'histoire du pays, mais que lon n'en etait pas moins confronte aujourd'hui avec le fait que les centrales nucl's sont trop couteuses, et en droit de se demander si un Etablissment comme EDF ne devrait pas renoncer a en construire."

[101]Interview with Bernard Laponche,

[102]Laponche papers generally. Maybe describe this collection briefly? Not possible to cite individual documents.

[103] Bernard Gonel, et al., tract, 27 October 1969. Laponche papers. "Nous n'acceptons pas de jouer le jeu de la division du personnel entre agents du CEA (les nobles) et agents des entreprises extérieures (les parias) que veut nous imposer l'administration et par lequel une partie du personnel est tenté. Pour nous tous les travailleurs quels qu'ils soient ont droit à une vie décente...Nous refusons d'être les complices d'une sté hypocrite et lâche qui fait toujours payer ceux qui peuvent le moins se défendre....Nous refusons d'être les complices d'une sté répressive qui utilise tous les moyens, même ceux qui contreviennent à sa propre légalité, pour conditionner et intimider ceux qui en définitive sont à la source de toute richesse, les travailleurs."

[104]Force-Ouvriere - CFDT, "Conference de Presse, Hotel de Ville de Massy (Essonne)," 30 October 1969. Laponche papers.

[105]Strike days: October 10, 22, 27, 30, November 3, 6, 13-18, 19-24. Not all centers went on strike during each of these times. Marcoule was on strike for all of the November dates. For a complete list of which centers went on strike when, see "L'Action dans les Centres," Energies Nouvelles (CGT, FSM, Journal du syndicat national des travailleurs de l'énergie atomique) Decembre 1969: 3.

[106]CGT CFDT, CGT-FO, "Hier à Palaiseau, 1200 Grèvistes manifestaient...," 31 October 1969; CGT CFDT, CGT-FO, "Pourquoi la grève du 6 novembre," 5 November 1969. Tracts, Laponche papers.

[107]CGC CFDT, CGT, CGT-FO, SPAEN, "Appel aux cadres du CEA," 17 November 1969. "la rentabilite n'est pas le seul critere qui doive compter."

[108] CFDT CGT, CGT/FO, "Pourquoi l'Énergie Atomique en Greve," 6 November, 1969. Tract, Laponche papers. "Nous sommes en train de perdre notre ind nationale, nous sommes sur la voie du sous-developpement et de la colonisation."

[109]Force-Ouvriere - CFDT, "Conference de Presse, Hotel de Ville de Massy (Essonne)," 30 October 1969. Flyer, Laponche papers.

[110] Force-Ouvriere - CFDT, "Conference de Presse, Hotel de Ville de Massy (Essonne)," 30 October 1969. "Une decision d'une telle importance qui engage le Gouv doit etre precedee de consultations avec les representants des personnels concernees, et non annoncee a la sauvette par un fonctionnaire, si haut soit-il."

[111] CGT CFDT, CGT-FO, "Pourquoi une grève à l'échelon national au Commissariat à l'Energie Atomique? Pourquoi cinq agents du CEA en sont aujourd'hui à leur onzième jour de grève de la faim?," 1969. "Nos objectifs sont clairs. Nous nous prononçons... pour des credits a la rech assurant l'avenir intellectuel, economique et social de tout un peuple et garantissant son independance."

[112] CFDT CGT, CGT/FO, "Pourquoi l'Énergie Atomique en Greve," 6 November, 1969. Tract, Laponche papers."Curieusement, les applications militaires, qui ne constituent que la partie la moins importante de la rech nucl, sont les moins touches par les restrictions de credits. Le Gouv parle d'ind nationale quand il s'agit de fabriquer des bombes atomiques; en meme temps, il liquide notre industrie nationale, gage de la veritable independance, source de progres social et de bien-etre."

[113]"L'Action dans les Centres," Energies Nouvelles (CGT, FSM, Journal du syndicat national des travailleurs de l'énergie atomique) Decembre 1969: 3.

[114]CGC, CFDT, CGT, CGT-FO, SPAEN, "Comparaisons Économiques et Politique Industrielle dans le Domaine Électronucléaire," 20 November 1969, p.
*5? Copies of this report courtesy of Jean-Claude Zerbib and Bernard Boudouresques. "les couts d'investissements des centrales, tels qu'ils sont annonces par EDF, sont incompréhensibles et incohérents."

[115]CGC, CFDT, CGT, CGT-FO, SPAEN, "Comparaisons Économiques et Politique Industrielle dans le Domaine Électronucléaire," 20 November 1969, p.
*6? "toutes les centrales sur lesquelles sont basées les comparaisons économique actuelles, et donc les décisions, sont des centrales `théoriques.'"

[116]CGC, CFDT, CGT, CGT-FO, SPAEN, "Comparaisons Économiques et Politique Industrielle dans le Domaine Électronucléaire," 20 November 1969, p. 6. "On ne comprend alors plus du tout l'abandon, non annoncé, mais implicite, de cette filiere ou le CEA et l'EDF ont investi des moyens considerables. Il ne s'agit plus d'un choix technique, mais d'un choix politique delibere, contraire aux interets les plus elementaires de la nation."

[117]CGC, CFDT, CGT, CGT-FO, SPAEN, "Comparaisons Économiques et Politique Industrielle dans le Domaine Électronucléaire," 20 November 1969, p.
*0-1? "Nul n'ignore l'offensive concertee des groupes d'industriels americains pour s'emparer de l'electromecanique fcse." "americanisation de l'electronucl fcse."

[118]CGC, CFDT, CGT, CGT-FO, SPAEN, "Comparaisons Économiques et Politique Industrielle dans le Domaine Électronucléaire," 20 November 1969, p. 10. "suffisamment rapprochees pour que les autres elements de choix (sortie de devises, valorisation des investissements effectues, independance nationale, plein emploi) soient placees sur le meme plan."

[119]EDF CA: 281 (23.1.70), 286 (26.6.70), 287 (25.9.70).

[120]For a more complete analysis of this incident, see Hecht, "Nuclear Spelunking," Journal of Contemporary History (forthcoming).

[121]"Filière graphite-gaz, Problème de répartition des commandes," 19.10.65 and EDF, REN2, "La Politique Industrielle d'EDF," 25.11.65 (both in personal papers of Claude Bienvenu).

[122]JSL 6, Juin 1969, p. 2.

[123]Such sentiments were expressed in interviews I conducted with workers who had been working on the Saint-Laurent site since the construction period in 1965-69 (all interviews conducted on the site; interview date in parentheses): Serge Roullier & Jean-Claude Godineau (18/1/90); Serge Roullier, Félix Mazier, & Jean-Claude Contois (18/1/90); M. Delarue (23/4/90); MM. Mureau, Marlet, & Occhipenti (23/4/90).)

[124] Saint-Laurent workers and engineers expressed their hopes in a variety of forums: training sessions, workplace practices, and the site's local newsletter, the Journal de Saint-Laurent.


[126]This account of the accident and the extent of the damage is based on the following sources: Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Pollution du réacteur, analyse des signaux DRG," Dépannage du réacteur SL1, rapport ndeg. 1; Georges Lamiral, Chronique de Trente Années d'Equipement Nucléaire à Electricité de France (Paris: Association pour l'histoire de l'électricité en France, 1988) 2: 109-112.

[127]The use of "pollution" occurs in virtually all reports of the accident and its clean up. For example: Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Pollution du réacteur, analyse des signaux DRG," Dépannage du réacteur SL1, Rapport ndeg. 1; Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Mesures de pollution," Dépannage du réacteur SL1, Rapport ndeg.4 (27-1-70). Anthropologist Françoise Zonabend gives a fascinating analysis of the varying cultural meanings of words such as "pollution" and "contamination" in the contemporary nuclear industry in her La presqu'île au nucléaire (Paris: Odile Jacob, 1989).

[128]Ibid., p. 2.

[129]Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Pollution du réacteur, analyse des signaux DRG," Dépannage du réacteur SL1, Rapport ndeg. 1; Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Etat d'avancement des études et des travaux au 17.12.69," Dépannage du réacteur SL1, Rapport ndeg. 2.

[130]Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Choix fondamentaux pour la suite des opérations de dépannage," Dépannage du réacteur SL1, Rapport ndeg. 3.

[131]Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Mesures de pollution," Dépannage du réacteur SL1, Rapport ndeg.4 (27-1-70).

[132]Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Choix fondamentaux pour la suite du dépannage, édition revue et corrigée du rapport ndeg.3," Dépannage du réacteur SL1, Rapport ndeg.5, and "Programme du dépannage et de la remise en service de la tranche," Dépannage du réacteur SL1, Rapport ndeg.7.

[133] M. J. Grand & M. J. Hurtiger, "Aspect de radioprotection pendant les interventions de Saint-Laurent-des-Eaux," Bulletin de l'Association Technique pour la production et l'utilisation de l'Energie Nucléaire, 91 (1971): 48; Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Etat d'avancement des études et travaux, planning au 1er juin '70," Dépannage du réacteur SL1, Rapport ndeg. 13.

[134]Grand & Hurtiger, "Aspect de radioprotection..." and Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Etat d'avancement des études et travaux, planning au 1er juin '70," Dépannage du réacteur SL1, Rapport ndeg. 13.


[136]Grand & Hurtiger, "Aspect de radioprotection..." and Centrale de St. Laurent des Eaux (Electricité de France, GRPT C), "Etat d'avancement des études et travaux, planning au 1er juin '70," Dépannage du réacteur SL1, Rapport ndeg. 13.

[137]JSL, ndeg. 17 (May 1970).

[138]See for example the JSL: ndeg. 12-13 (December 1969-January 1970), p. 2; ndeg. 16 (April 1970), p. 1.

[139]Interview with Joël Sorin.


[141]Interview with Joël Sorin.

[142]The choice between pressurized and boiling water reactors could itself be the subject of an entire chapter. See Picard et. al, Jasper, and Lamiral for a variety of perspectives on this process.

[143]For more on these changes, see Bénédicte M. Vallet, "The Nuclear Safety Institution in France: Emergence and Development," Ph.D. dissertation. New York University, 1986.

[144]For French anti-nuclear protest, see among others Alain Touraine, et al., La prophétie anti-nucléaire (Paris: Editions du Seuil, 1980); Dorothy Nelkin & Michael Pollak, The Atom Besieged: Antinuclear Movements in France and Germany (Cambridge, MA: the MIT Press, 1981); Francis Fagnani, ed., Le Débat nucléaire en France: Acteurs sociaux et communication de masse, (Grenoble: Université des Sciences Sociales, Institut de Recherche Economique, 1977); Francis Fagnani & Alexandre Nicolon, eds., Nucléopolis: matériaux pour l'analyse d'une société nucléaire, (Grenoble: Presses universitaires de Grenoble, 1979).

[145]Cited in Simmonot, p. 65. "L'abandon de la filière au graphite-gas n'a pas été une décision politique, mais une décision technique; ç'a été une erreur de dire que c'était une décision politique; une décision politique aurait consisté à maintenir le graphite-gaz. L'abandon du graphite-gaz se justifiait par deux raisons: son exploitation n'était pas satisfaisante, et l'exportation était très difficile."

[146]Cited in Simmonot, p. 111. Simmonot's interviewees are anonymous, but speaker is probably Boiteux, who made a similar comment during my interview with him (date
*) (Simmonot describes him as an EDF "dirigeant. L'un des acteurs principaux de la décision. Dans la pleine force de l'âge. Massif de certitudes. Membre de Péon depuis 1967." p. 109). "On s'est finalement décidé pour la filière américaine, après avoir perdu quatre ans...L'explication [for this waste of time] est purement politique. On opposait une filière dite nationale à une filière dite américaine... qu'est-ce que cela signifie? Oublie-t-on que dans les centrales classiques au fuel il y a aussi des licenses américaines?"

[147]Quoted in Simmonot, p. 123. "Oh! Elle n'est pas tellement française que cela.

Des techniciens avaient convaincu des hommes politiques de la valeur de cette filière, en partie copiée sur les Anglais. Et ces hommes politiques étaient devenus encore plus acharnés...La filière française a deux graves défauts. D'abord elle utilise de l'uranium métal qui est un matériau instable et moins sûr que l'uranium enrichi, en cas d'incendie par exemple. Voyez ce qui est arrivé à Windscale (Grande Bretagne). Ensuite, l'emploi du gaz pose des problèmes difficiles; il faut installer un système de déchargement continu, alors que dans les réacteurs à eau, on peut se permettre de n'ouvrir la cuve qu'une fois par an."