PART ONE—
PRIVATE ENTERPRISE IN DECLINE, 1928–1936

In the early 1930s French manufacturers, engineers, and politicians tried to rejuvenate a languishing aircraft industry—and failed. This episode in the history of French aviation, which began with the creation of an Air Ministry in 1928 and ended on the eve of the Popular Front in 1936, is commonly forgotten today. Aviation enthusiasts prefer to celebrate the heroic exploits of French aviators in the 1920s, when fliers like Jean Mermoz and Henri Guillaumet captivated the public by stretching the airmail routes to Dakar, Buenos Aires, and the far side of the Andes. Retired aircraft workers, by the same token, tend to remember the political dramas of later years—the Popular Front, the Resistance, and the Liberation. Yet however obscure the early 1930s may be in the folklore of aviation, they were fateful for the industry. Technical advances and the specter of German rearmament created a formidable challenge for French industrialists—to build versatile, high-speed airplanes by the hundreds. To meet this challenge, aircraft manufacturers had to modernize their methods, enlarge their firms, and, in the end, restructure their industry. Change of this sort did not come easily in France, where mass production and industrial concentration were much less advanced than in Germany. In fact, most of the major aircraft manufacturers evaded rationalization during the early 1930s, and in doing so they unwittingly lost valuable time for rearmament and made their firms vulnerable to new forms of state intervention.

The early 1930s were pivotal for labor relations as well. At the beginning of the decade aircraft workers, like most laborers in France, had little means of defending themselves against hostile foremen, autocratic employers, and the threat of unemployment. Aircraft unions were little more than the fantasies of a handful of militants; in the metalworking world of which aviation was a part, the labor movement still had not

recovered from the strike defeats that had crippled the left in the early 1920s. Employers ruled as they pleased. But after 1933 conditions began to change. By the end of 1935 militants in a number of plants were clearly rebuilding a labor movement—cementing loyalties, drawing up grievances, sorting out rivalries, and creating a moral climate for militant action. By the time Léon Blum came to power in 1936, the political ethos of the Popular Front had penetrated into the plants.

One—
The Failure of Industrial Reform

Aircraft manufacturers in interwar France were quick to invent new airplanes but slow to convert to mass production. This combination of technical prowess and business conservatism had deep roots in a country where industrialization had done little to diminish the prominence of specialty trades, artisanal production, and the family firm. In the early days of aviation this penchant for small-scale, high-quality production was a blessing. Between 1906, when the Voisin brothers established the first factory for making airplanes for sale, and 1914, when more than twenty small firms had entered the business, France quickly emerged as the world leader in aviation.^[1] Much the same story could be told of French automaking. But by the early 1930s French aircraft builders were losing ground to American and German firms better equipped for aggressive marketing, corporate mergers, and mass production. As early as 1928 nearly everyone connected with French aviation recognized that French manufacturers were slipping behind. But if the malaise was obvious, its remedies were not. Business and government officials spent nearly a decade battling with one another over how to revitalize a troubled industry.

Despite the predilections in France for entrepreneurial caution, it is puzzling that French aircraft manufacturers had such difficulty in the interwar period modernizing their industry. For one thing, French manufacturers had already made one successful effort to mass-produce aircraft—during the First World War, when French firms employed two hundred thousand workers to build more than fifty-one thousand airplanes and ninety-two thousand motors, an achievement that made France the leading producer of aircraft matériel in the war.^[2] During the 1920s, moreover, several French firms—Potez, Bréguet, Hispano-Suiza, Gnôme-et-Rhône—remained major competitors in the world market. In

the French business community at large there were several entrepreneurial visionaries—men like Ernest Mercier and Auguste Detoeuf, who proselytized in behalf of scientific management, industrial concentration, and a new partnership between business and the state to stimulate growth. These neoliberals dissented openly during the early 1930s from the dominant values of laissez-faire liberalism and called instead for a planned, mixed economy.^[3] Yet despite a strong start in aviation and a business culture at least partially open to new ideas, the French failed to rejuvenate the industry in those years. Government policy, military doctrine, budgetary austerity, and entrepreneurial strategy all conspired to freeze the industry in a structure ill suited for a rearmament effort.

Industrial Drift in the 1920s

The troubles in French aviation began with demobilization after the First World War. Peace cut short the demand for warplanes, and new markets for aircraft only slowly emerged. Throughout the 1920s the industry floundered. In 1919 employment shrank 50 percent to one hundred thousand workers and then plummeted in 1920 to an astonishing fifty-two hundred workers.^[4] Many firms managed to survive by investing war profits shrewdly and making other products until demand for airplanes revived. As a result, by the late 1920s at least twenty-three companies were still in the business of building airframes in France, and about ten companies were making motors—fewer firms than the forty of 1918, but a sizable number nonetheless.^[5] Since nearly all these companies made their profits on military aircraft, builders in this troubled industry appealed to the state to keep them alive.

For the most part, government officials obliged. The army had served as the industry's main customer since before 1914, and despite the emergence of commercial aviation state orders remained the lifeblood of the companies after the war. Eager to preserve the industry as a military resource, the government followed a politique de soutien , or support policy, of dispersing orders widely to keep firms afloat. Such a policy had the virtue of maintaining excess capacity in the event France had to rearm for another war; it was also consistent with the military's desire to avoid becoming too dependent on any one firm for supplies.^[6] A support policy, of course, could easily degenerate into pork barreling, as was the case in the notorious stock-liquidation scandal, when some manufacturers were alleged to have bought back surplus airplanes at a pittance from the government to sell them at premium prices abroad. Questions of collusion between state bureaus and the companies kept the industry in the newspapers throughout the 1920s; it was commonplace to assume

that airplane companies depended as much on government connections as on the talent of entrepreneurs.

Not that the industry lacked for talent: by the late 1920s two generations of remarkable builders had emerged to dominate the aircraft business. Several of the prewar pioneers, men like Louis Bréguet, the Farman brothers, and Fernand Lioré, still ran firms and served as chieftains of the industry. Typically these pioneers had trained in elite engineering schools and worked in machine construction or the new automobile industry before throwing themselves into aviation. Louis Bréguet, for example, after graduating from the Ecole Supérieure d'Électricité, worked for a while in his family's electrical machine firm before turning to airplanes in 1905. Louis Blériot established an automobile headlight factory before taking to the air. When Fernand Lioré, a polytechnicien with several years' experience in the chemical industry, saw Blériot fly in 1907 in the fields of Issy-les-Moulineaux outside Paris, he gradually began to convert his own new automobile accessories firm into an airplane company. Aviation became a passion for men with just the right blend of money and madness to adapt swiftly to a novel technology.

During the First World War a second generation of builders entered the business alongside the early adventurers—young men like Henry Potez and Marcel Bloch, who after graduating from the new Ecole Supérieure de l'Aéronautique in Paris joined the army engineering corps and used their positions to launch a firm of their own. Emile Dewoitine, a brilliant young engineer, had a similar start in Toulouse; during the war he rose to the rank of technical director for Pierre Latécoère's company before breaking away to set up his own Toulousain firm in 1920. But whatever their differences in experience, both generations of builders, the pioneers and the younger newcomers, possessed that special combination of qualities it took to succeed in the business—creativity, ambition, an obsession with flight, a measure of greed, and, not least, a capacity to cultivate contacts in the ministries. If the world war taught these airplane manufacturers anything, it was that military orders made a firm.^[7]

Not surprisingly, then, builders and politicians looked to the state for remedies when the ills of the aircraft industry became apparent in the late 1920s. By 1927 employment had risen to eleven thousand.^[8] Only a few firms really prospered. Bréguet, Potez, Farman, and Lioré et Olivier remained competitive internationally, but many firms just limped along. Blériot's director, for instance, warned parliament in 1926 that despite the firm's recent technical successes it might soon have to close for lack of orders.^[9] The irregularity of demand and a chronic shortage of capital—few bankers could stomach the risks of the aircraft business—

plagued nearly every firm in the airframe sector. The major engine firms, especially Gnôme-et-Rhône and Hispano-Suiza, flourished reasonably well building motors for a host of client firms; but in the end their health, too, depended on the vitality of the airframe business.

Nor did the general condition of French aviation give cause for comfort. Throughout the 1920s the fledgling French air force that had emerged from the war remained little more than a stunted stepchild of the army and navy. Military conservatives and their parliamentary allies stymied every effort to create an autonomous air force.^[10] Commercial aviation languished as well. French fliers may have been winning trophies and grabbing headlines, but the French airlines appeared to be less of a match for their rivals abroad. The German aircraft industry, which had been prohibited from building military aircraft by the Treaty of Versailles, spent the 1920s developing the world's most advanced commercial aircraft to supply an aggressive group of German airlines.^[11] By 1928 Germany boasted a network of domestic air routes covering sixty thousand kilometers, which made the paltry three-thousand-kilometer domestic network of France seem in comparison like a blank wall, as one despairing journalist put it.^[12] More worrisome still, the new British and German semipublic airlines, Imperial Airways and Lufthansa, threatened to squeeze French airlines out of the new international air routes opening up in Asia, the Near East and the Americas. When Charles Lindbergh beat his French competitors in the race to master the Atlantic in 1927, French journalists announced to the public what experts had been fearing for years—that French aviation was fast losing its competitive edge.^[13] By 1928 every airplane accident, business scandal, or setback at the international air races only deepened the awareness of a dreadful crisis in French aviation. Henry Paté, a deputy from Paris, captured the prevailing mood of frightened frustration when he introduced the air budget of 1928 to his colleagues in parliament: "Once again we cry: French aviation is gravely ill. In its general organization, its technology, its industry, in supplying military and naval units, and in developing its commercial airlines, it suffers from profound troubles which have become chronic, will soon kill it, and which in any case have now made it inferior to aviation abroad."^[14]

The New Air Ministry

By the summer of 1928 ideas were circulating in parliament about how the state could rejuvenate aviation without jeopardizing the private status of the industry. The most serious effort to chart a policy came from a commission of the Conseil National Economique (CNE), the quasi-governmental body that a center-left government, the Cartel des

Gauches, had created in 1925 to bring together spokesmen from business, labor, and the state. Raoul Dautry chaired the commission. As a polytechnicien and director of the Compagnie du Chemin de Fer du Nord, Dautry bridged the worlds of engineering, business, and state administration. A Freemason and political centrist who stayed clear of party rivalries, he embodied the Saint-Simonian optimism and political pragmatism that had become common in progressive business circles since the war. Like many technocrats of the late 1920s, he viewed industrial rationalization and a closer partnership between business and the state as the keys to economic rejuvenation.^[15]

In his CNE report Dautry offered a hardheaded diagnosis of commercial pathology. He said the government's support policy, its failure to promote research, and its haphazard subsidies all conspired to weaken French aviation. As a remedy, Dautry's commission called for policies that would enhance state regulatory powers without encroaching on the autonomy of private firms, including state subsidies for research, state-guaranteed loans for private companies, improvements at the Ecole Supérieure de l'Aéronautique, and the creation of a Superior Air Council to coordinate air policy. Dautry drew attention to Germany, where a subtle partnership between the state, the big banks, and the companies had enabled the builders to make strides. "From the example of foreign countries," he argued, "it would seem useful to make a small number of creative firms prosper and to steer industrial manufacturers toward series [mass] production."^[16] But he stopped short of recommending a way to reorganize the industry, calling for industrial concentration but refraining from saying how it was to be done. Dautry's commission, which was itself divided over whether the state should take over the airlines, proved astute in analyzing the illness but equivocal in prescribing a cure.

Meanwhile support grew in parliament for creating an Air Ministry. A parliamentary coalition stretching from the conservative Pierre-Etienne Flandin to the Socialist Pierre Renaudel shared the view that a new ministry might help revive aviation. Since the war, aircraft questions had fallen under—and often between—the purview of four ministries: war, navy, commerce, and the colonies. Air policy had become a hodgepodge of programs; as the army's leading advocate for aviation, General Hirschauer, put it, "There were too many tensions, too many cliques, too much of a desire to be isolated from one another and to have one's own schools . . . , personnel, workshops, [and] experimental commissions."^[17] A single ministry, as many aviators had been arguing since 1920, could coordinate policy, economize funds, and pave the way for a bona fide, autonomous air force. But since an autonomous air force was precisely what military conservatives bitterly opposed, Poincaré's center-right government dragged its feet on the issue.

When on 6 September 1928 Poincaré's cabinet finally yielded to the notion of an Air Ministry, the irony of the circumstances escaped no one. Four days before, Maurice Bokanowski, Poincaré's commerce minister, died in a plane crash en route to Clermont-Ferrand, where he was about to address a gathering called to popularize the idea of air travel. The Socialist press took the opportunity to ridicule the government: "Criminal French aviation," the headlines of Le Populaire read, "has killed even its own leader!"^[18] An embarrassed cabinet approved the new ministry, and within days Poincaré named André Laurent-Eynac to run it. Amid the uproar opponents of an Air Ministry, not the least of whom had been Bokanowski himself, now mounted little resistance.

Poincaré's new air minister, Laurent-Eynac, a lawyer and parliamentarian, seemed ideal for traversing between the government bureaus and company boardrooms of aviation. He had flown in the war, chaired a subcommittee on aviation in the Versailles Treaty deliberations, and promoted air routes to French colonies as under secretary of aviation in 1921. A powerful insider in aviation circles, he had shown his gift for delicate brokering in 1924 when as under secretary once again in Herriot's Cartel des Gauches government he managed to keep the lid on investigations of the stock liquidation scandal. Once a Radical-Socialist, Laurent-Eynac had gradually migrated like so many of his colleagues to the respectable moderation of the center-right. As one prominent general said of him, "he knows the airline executives and the manufacturers in aviation thoroughly; he has confidence, firmness, tenacity, and savoir-faire. And something else valuable: he is viewed well by the press, he has the ear of parliament and is liked there, and he knows how to use his influence in the government."^[19] Poincaré's choice no doubt pleased both the aeronautical professionals looking for competence and the businessmen looking for a friend in high places.^[20]

Although Laurent-Eynac assumed command in temporary headquarters—the Air Ministry would eventually be located near the southwestern edge of town on the boulevard Victor—his first act as minister was a decisive step toward industrial reform: he appointed Albert Caquot as the ministry's technical and industrial director. Though only forty-seven, Caquot was already something of a legend in the world of French aviation, for it was he who had supervised the airplane procurement program in 1917 and 1918. The third son of a farming family in the Aisne, Caquot rose to fame, and eventually fortune, in a manner that seemed to vindicate the meritocratic ideals of both the Napoleonic and republican traditions. As a brilliant student he won entry into the Ecole Polytechnique in 1898, and after securing a post in the prestigious Corps des Ponts et Chaussées, he made a name for himself as a civil engineer before

the war. Then, as technical director for the under secretary of aviation, he coordinated the immensely successful effort to mass-produce airplanes during the final stages of the hostilities.^[21] By the 1920s Caquot—stern, dignified, decisive—enjoyed the enormous respect that only a man of extraordinary accomplishment as both engineer and administrator could have commanded in the select fraternity of the aviation business.

With a political mandate to address the purported crisis in aviation, the team of Laurent-Eynac and Albert Caquot hoped to revitalize the aircraft industry in three ways: by launching a new wave of innovation in airplane design; by decentralizing the industry, moving it to new locations away from Paris; and by streamlining and concentrating the industry's structure. Caquot, who as an engineer approached industrial problems as if they were technical puzzles, believed that the first goal, technological innovation, provided the key to the entire undertaking. French aeronautical design had suffered badly from state neglect; in 1928, for example, the French government spent 40 million francs on aircraft research, in comparison to the 118 million francs spent in Germany.^[22] Overnight Caquot made research the Air Ministry's highest priority; he immediately replaced the discredited support policy with what would prove to be the hallmark of the ministry between 1928 and 1932—a so-called prototype policy to promote research. Accordingly, the ministry set aside a large portion of its budget to pay firms to design airplanes for speed and ease of maneuver. To stimulate this activity, it would reimburse firms for 80 percent of the costs of new prototypes, however viable these new inventions proved to be. In addition, it would give bonuses for prototypes that offered new ways to enhance the speed or diminish the ascent time of an aircraft. In exchange for this support the builders were required to yield all patent and licensing rights to the state.^[23]

Although this last provision infuriated those builders who were accustomed to profiting handsomely by selling licenses to other firms, the prototype policy plowed a great deal of money into the companies. Moreover, Caquot's prototype policy conformed to an established tradition of state-sponsored research. Since the eighteenth century French governments had financed the invention of guns, cannons, and ships, and had even built state-owned arsenals, without undermining the arms business as a private, profit-making venture. During the First World War state sponsorship of research had gone hand in hand with what one historian has called the privatization of arms production: state contracts expanded the private sector in the arms business much more substantially than the state-run sector.^[24] As a veteran of the airplane procurement effort, Caquot symbolized the continuity of a tradition in which state engineers played a strong role in directing research without in-

fringing on the independence of private firms. He saw the new prototype policy in precisely this light—as a way to invigorate the industry within the framework of state-tutored private enterprise.

Caquot's efforts to decentralize the industry conveyed the same respect for the integrity of private firms. The vast majority of airframe and engine plants had been built in Paris and its western suburbs. Since the war, however, military officials had urged airplane manufacturers to set up new plants in southern and western France as a precaution against aerial bombardment in the event of another war. Moreover, advocates of decentralization argued that by building plants in provincial cities, employers would reap the benefits of cheaper wages. Yet by 1928 only a few plants had sprung up in the provinces. True, Pierre Latécoère and Emile Dewoitine had made Toulouse second only to Paris as a center for airplane construction, and the port cities of Nantes, Saint-Nazaire, Bordeaux, and Marseille served several firms profitably as places to make seaplanes. But for most companies the attractions of Paris remained irresistible. Close ties to other metalworking firms, the availability of engineers, draftsmen, and skilled workers, and proximity to the ministries made Paris a superb locale for building planes. Only a determined Air Ministry could reverse a natural inclination to expand the industry as a Parisian enterprise. Accordingly, Albert Caquot established a fund to subsidize the costs of starting or enlarging provincial plants. Again, in decentralization as in prototype building, Caquot's style was to steer, guide, cajole—but not commandeer.

In the same spirit Caquot approached the most vexing problem of all, the need to streamline the aircraft industry by reducing the number of firms. In November 1928 he and Laurent-Eynac sent plans for restructuring the industry to the Chambre Syndicale des Industries Aéronautiques, the employers' association that the pioneers of the industry had founded in 1908. Housed in elegant headquarters on the rue Galilée just south of the Arc de Triomphe, the Chambre Syndicale was the one institution that brought together employers in all three branches of the industry—airframes, engines, and accessories. This organization, Caquot hoped, would assume some of the burden for converting an overgrown gaggle of government-fed firms into a leaner group of major enterprises. With this goal in mind, Caquot and Laurent-Eynac called on the Chambre Syndicale to form groupements, or company groups, that would pull firms together into trusts and lay the foundation for future mergers. According to Caquot's plan, the groupements would bring about "the rationalization of the industry," a process of industrial concentration carried out by the builders themselves.^[25]

Caquot's program, with its focus on prototypes, decentralization, and groupements, produced a flurry of activity between 1929 and 1932, but the

results were disappointing. Prototypes proliferated: the Air Ministry sponsored more than two hundred contracts for innovations in airframes and engines. But as a Finance Ministry audit revealed in 1933, the Air Ministry handed out too many contracts to too many firms to build airplanes of questionable technical value. There seemed to be no grand design to the program. Auditors discovered, moreover, that Caquot's system of paying firms in three installments for their prototypes gave builders an incentive to abandon projects in midstream and husband the funds for other activities.^[26] Some strides were made—in 1931 French firms made the breakthrough to all-metal airframes, and by 1933 some prototypes achieved a speed of 350 kilometers an hour^[27] —but many planes failed; Villacoublay, the major testing field just southwest of Paris, became known as a prototype cemetery.^[28] Similarly, the decentralization policy produced only modest results. Louis Bréguet opened factories in Le Havre and Saint-Nazaire; Lioré et Olivier followed suit in Rochefort, as did Hanriot in Bourges, Amiot in the Norman town of Caudebec-en-Caux, and Loire-Nieuport in Saint-Nazaire.^[29] Overall, however, the industry remained vulnerable to German attack, and the engine sector was still completely ensconced around Paris.

Worse still, little came of the effort to streamline the industry. Two mergers did occur, with the encouragement of the Air Ministry—the creation of Loire-Nieuport and Potez-CAMS. But Caquot's strategy of groupements backfired. As early as the autumn of 1928 builders at the Chambre Syndicale warned Laurent-Eynac that the whole subject of industrial concentration "had to be approached with great prudence." They expressed a willingness to "envision mergers, but they would have to involve [state-financed] indemnities for the losers," something the Air Ministry opposed.^[30] By 1930 the Chambre Syndicale had gone ahead and helped firms construct two large groupements —the Société Générale Aéronautique, which brought together Hanriot, Amiot-SECM, Loire-Nieuport, the Société Aérienne Bordelaise, and the engine-building firm Lorraine; and the Groupement Aéronautique Industriel, which included Bréguet, Lioré et Olivier, Potez, and for engines, both Hispano-Suiza and Renault.^[31] But the companies created these groups for purposes quite different from those Caquot had intended. By design the builders had assembled firms with complementary specialties so that when the Air Ministry granted an order to the group, it was simply passed along to the firm best equipped to handle it. Firms in the company groups pooled some of their financial resources but left intact their own production facilities and administrative services. The Chambre Syndicale, moreover, created a liquidation fund, financed through the groups, to compensate weak firms that might otherwise fall into bankruptcy.^[32] In short, the builders used the company groups not to rational-

ize the industry by closing down plants and negotiating mergers but rather to protect themselves from the very process of concentration that Albert Caquot had proposed.

Thus, between 1928 and 1932 the Air Ministry managed to stimulate research but failed to remedy the structural flaws that lay behind the shortcomings of the industry. This failure to develop a more concentrated, rational manufacturing sector stemmed from three basic sources. The first was the contradictory nature of Caquot's efforts. Despite his desire for industrial concentration his policies actually militated against the emergence of a few robust firms. As auditors pointed out in 1933, his pricing policy, which allowed firms to add a flat 10 percent profit margin to their cost estimates, gave builders little incentive to cut costs or seek economies of scale.^[33] Caquot's stress on prototypes and decentralization, moreover, impeded the progress of industrial concentration. Pierre-Etienne Flandin, leading conservative rival to Laurent-Eynac for the Air Ministry in 1928 and a close associate of Louis Bréguet and the Farman brothers, criticized Caquot's policy on precisely these grounds. Flandin applauded the effort to promote research but opposed a policy that separated prototype building from series production. Caquot's approach, he argued, was simply financing a host of small builders who were eager to invent something but ill equipped to address the challenge of mass production. Better to entrust research to larger firms able to modernize their manufacturing procedures and lower their costs. Flandin identified the same fallacy in decentralization. This policy, he asserted, would only weaken the largest firms by depriving them of skilled Parisian workers, who did not want to be uprooted, and engineers, who thrived on contact with "that creative flame that animates the Paris region."^[34] What the industry needed, in his view, was the same spirit that "propelled General Motors in America"—a commitment to bigness, since aviation seemed to him destined to follow the commercial path of the automobile. Critics like Flandin believed Caquot had unwittingly perpetuated the support policy he had vowed to replace.

It was one thing to argue that Caquot's prototype policy should have linked research to a program promoting mass production; it was quite another to propose feasible ways to increase demand for mass-produced aircraft. Here lay the second impediment to industrial concentration—the failure of the Air Ministry to expand the market for aircraft, either by building a strong air force or by stimulating growth in the airline business.

When the air ministry opened in 1928, no autonomous air force came in on its coattails. Between 1928 and 1932 pacifism in parliament and traditionalism in the military kept air force enthusiasts at bay. Laurent-Eynac made a few modest gains: he created the post of air force chief of

staff and won juridical control over the air force, even though most air personnel technically remained under army and navy command. Laurent-Eynac's successor, Paul Painlevé, continued the effort; he established a Superior Air Council in 1931 and an Air Force Studies Center in 1932.^[35] But on the terrain that really mattered—military doctrine, battle planning, and the budget—army conservatives like generals Philippe Pétain and Maxime Weygand rebuffed the aviators. For them, an air force was simply an auxiliary instrument of a land-based army, valuable for reconnaissance, transport, and tactical bombing but illegitimate as a strategic fighting force. As veterans of a war that had been won on the ground, army traditionalists had little sympathy for the notion of massive, strategic bombardment that the Italian strategist, General Giulio Douhet, had popularized in his writings during the early 1920s. Moreover, Douhet's lurid images of devastated cities made pacifists on both the right and the left all the more hostile to "aero-chemical warfare" and the notion of a large, modern air force.^[36] Through 1932 military traditionalism, pacifism, and the pursuit of disarmament all conspired against any serious new program for building airplanes. Without an ability to distribute large orders, Albert Caquot could not entice builders into restructuring firms.

Nor was the Air Ministry successful in bolstering civil aviation as a market for aircraft. By the late 1920s five major airline companies, each heavily subsidized by the state, faced chronic financial difficulties and stiff foreign competition. In 1928 Raoul Dautry's CNE commission on aviation had been at odds over what to do: some members favored, and some opposed, consolidating the five airlines into a single national firm in which the state would participate as the major minority stockholder—the arrangement behind Lufthansa and Imperial Airways. For conservatives like Pierre-Etienne Flandin, this scheme raised the specter of a state monopoly. Amid the controversy Laurent-Eynac first wavered, then postponed, and finally evaded the issue. Instead of addressing the structural problems, the Air Ministry continued to subsidize the five airlines and pinned its hopes on a drive to negotiate favorable air routes abroad for the leading French firms.^[37] By 1933, however, the depression and Hitler's rise to power had so poisoned the atmosphere for air route diplomacy that even this feeble strategy had to be abandoned. The French airline business suffered accordingly, as did the demand for commercial aircraft.

If Caquot's policy was poorly designed for promoting industrial concentration, and if large orders were unavailable as a tool of industrial policy, might it still have been possible for the builders, as Caquot hoped, to consolidate firms on their own? Perhaps, but a third obstacle stood in the way of structural reform: the hostility of the manufacturers to state

intervention. Of course, every aircraft company was perfectly prepared to depend on the Air Ministry as its chief source of sales. The prototype policy, in fact, made builders as eager as ever to lobby privately in the corridors of the ministry for lucrative contracts. Georges Houard, the editor of the trade journal Les Ailes and a conservative conscience for the industry, was moved in 1930 to write in an open letter to the air minister, "There is an immoral rush toward the Air Ministry going on, not to give support, competition, and cooperation, but to extract from its formidable budget of two billion francs everything that such a sum placed at the disposition of the State would permit in the form of orders, subsidies, broker's commissions, honor, and other smaller advantages. Aviation? That's secondary! Business first!"^[38]

No doubt most of the moguls of aviation believed that there was something noble, and certainly patriotic, about building airplanes. But in the end it was a business; aircraft manufacturers were as loath as anyone else who owned a factory to sacrifice their autonomy at the behest of the Air Ministry. To do business with the state without losing control of the business—this was the delicate balance that every builder wanted. Yet with so many companies bedeviled by the problems of winning regular orders, finding long-term capital, paying a highly skilled work force, and keeping up with technical advances, it was a balance that was becoming more and more difficult to achieve. When Caquot and Laurent-Eynac sent the Chambre Syndicale their plans for stimulating mergers and acquiring the licensing rights of new planes, the builders promptly assumed a defensive posture. At the Chambre meeting of 4 December 1928 some members went so far as to proclaim that "we are headed toward the socialization of the industry" and that the builders ought "to get in touch with other Chambres Syndicales" to coordinate a response.^[39] Ironically, Caquot, who had no intention of undermining the private status of the airplane business, had inspired the builders to turn more resolutely to the Chambre Syndicale as an instrument of collective defense.

This odd blend of "asking for personal favors from people highly placed," as one company spokesman put it, while pursuing a strategy of collective resistance to state-led reform served to perpetuate the fragmented structure of the industry.^[40] Had one or two firms been strong enough to swallow up their competitors, the industrial concentration that Raoul Dautry, Albert Caquot, and many other observers were calling for might have occurred. Several leading builders were certainly aggressive enough to have entertained great ambitions for growth. Fernand Lioré, the Farman brothers, and Louis Bréguet had all invested heavily in the airlines, and Lioré in particular had become a major power in the boardrooms of commercial aviation by the early 1930s. Louis Bréguet

had experimented during the 1920s at his factory in Vélizy outside Paris with ways of adapting the basic principle of the assembly line to airframe production. He eventually abandoned the effort for lack of large-enough orders to make it pay.^[41] Henry Potez and Marcel Bloch were nothing if not ambitious, and they were clearly the ablest aeronautical engineers of the lot. By the early 1930s Potez had built, with hefty government subsidy, a major factory for large series in the small northeastern town of Méaulte. But none of these men could establish enough command over the market to become an oligarch in the industry. The one manufacturer who eventually did build an industrial empire, Marcel Bloch (later Marcel Dassault), was only getting back into the airplane business in 1929 after a decade of making money in real estate. In the early 1930s no one was in a position to become the General Motors of aviation of which Flandin dreamed.

Not only did state policy and the structure of the industry mitigate against change; the business culture of aviation did so as well. The aircraft business was a fraternity, a tightly woven network of men who shared in the great technological adventure of their time. They had all won places for themselves in aviation during the world war, and they continued to serve side by side as the stewards of a host of institutions—the Aéro-Club de France, the Comité pour la Propagande de l'Aéronautique, the Fédération Nationale Aéronautique, the Ligue Aéronautique de France, and the airports and testing fields that by 1930 had sprung up in every important municipality in France. A busy calendar of air races, flying shows, dinner gatherings at Maxim's, meetings at the rue Galilée, and sessions in the bureaus of government, to say nothing of family connections and school ties, made the airplane business a fishbowl where men like Potez, Bréguet, and Bloch were colleagues as well as competitors. The bonds of friendship, the pressures of dependence, and the benefits of collaboration were too strong to ignore. In such a milieu airplane manufacturers had little incentive to break from the fold. Instead they preferred to protect the autonomy of all the firms, pursue the strategy of groupements as a defensive measure, and, in so doing, preserve the status quo. As a result, by 1932 the industry was still as fragmented, as ill equipped for mass production, and as poorly prepared to compete with foreign rivals as it had been in the late 1920s.

The balance sheet of Caquot's reform effort between 1928 and 1932 mixed high promise with mediocre results, and in this respect the Air Ministry bore the stamp of the Tardieu era in national politics. André Tardieu, whose two governments between 1929 and 1932 shaped the national political debate, had been an early advocate of an Air Ministry in the twenties. Though a leading conservative, in 1929 he broke decisively with the right-wing orthodoxy of laissez-faire. He campaigned vig-

orously for a policy of national retooling—a program of state financing for industrial modernization that resembled the kind of state intervention that Caquot and Dautry were promoting.^[42] Between die-hard conservatives like Louis Marin, on the one hand, and the Socialist and Communist left, on the other, these neoliberals searched for new ways to use the financial leverage of the state to rejuvenate a capitalist economy. Air policy offered these reformers a splendid arena for putting these ideas to work since aircraft procurements, commercial subsidies, and air route diplomacy gave state officials considerable power in the industry.

Yet neither Tardieu nor the Air Ministry could overcome the obstacles to industrial reform. By 1931 Tardieu had proved unable to construct a political coalition to push national retooling through parliament. The traditional right refused to betray laissez-faire principles, and the reformist left resented Tardieu for stealing its thunder. Most Radicals, meanwhile, opposed his program for fear it would usher in the corporatist France of big business and bureaucracy they abhorred. In the end the interest groups that stood most to gain from Tardieu's retooling—big employer organizations, medium- and large-scale farming concerns, and, to some extent, the reformist trade unions—remained either too weak, too suspicious of the state, or too removed from parliamentary debate to lobby effectively for the program.

Likewise, Caquot and Laurent-Eynac lacked the political support for pursuing a coherent strategy of state intervention. Schemes for consolidating the airlines, much less the aircraft industry, aroused bitter opposition from the left and the right. During the air budget debate of 1930, for example, Socialists expressed fears that the effort to promote concentration in the industry would only create powerful trusts and enrich the banking cartels without solving the underlying problems in aviation.^[43] The right, meanwhile, shared Flandin's fears that state intervention would prevent a few healthy large firms from really flourishing. In short, the new Air Ministry lacked the broad base of political support needed to change both the level of demand for airplanes and the structure of the industry itself—the two steps required to revitalize French aviation. The malaise of the 1920s had inspired just enough political consensus to get an Air Ministry off the ground but not enough to launch policies that would break the impasse over industrial structure, military doctrine, and airline subsidies that had led to the malaise in the first place.

The lack of political allies, however, does not account fully for the weakness of Air Ministry policy; ideological convictions were important as well. Dautry, Laurent-Eynac, and Caquot proposed solutions to basic problems—an industry with a half-dozen large firms, an autonomous air force, and a national airline equipped to compete with Lufthansa. But they were hamstrung when it came to taking action. In the late 1920s and

early 1930s progressive businessmen and state administrators were still constrained by conventional notions of state intervention. They may have toyed with visions of consolidating the aircraft business "from above"; but when it came time for action, the men at the Air Ministry fell back on traditional methods. The prototype program, international business diplomacy, continued subsidies to the airlines, and tighter control over air policy—these were all half-measures that revealed the ambivalence reformers still felt about enlarging the role of the state in French industry.

Pierre Cot and the Second Round of Reform

Changing international conditions, however, and the return to power of the Cartel des Gauches paved the way for a fresh round of reforms in the aircraft industry in 1933. By then the French economy had finally succumbed to the Great Depression; French aviation had weakened accordingly. Hitler, moreover, had come to power, deepening French fears of a belligerent Germany. Military officials had little doubt that Hitler could, in short order, transform Germany's top-notch aircraft industry into a rearmament machine. It was in this context that Edouard Daladier, the emerging standard-bearer of the Radical Party, formed a center-left cabinet on 31 January 1933, which brought Pierre Cot to the helm of the Air Ministry. However lackluster the Radical governments of 1932 and 1933 proved to be for the country as a whole, Cot's brief stint at the boulevard Victor turned out to be a remarkable moment in the struggle to reinvigorate the industry.

Cot, in contrast to his predecessors at the Air Ministry, was willing to challenge the conventional boundaries of state intervention. His nonconformity in this respect was characteristic of many young politicians coming of age in the early 1930s. Born into a political family from the Savoie in 1895, Cot served in the combat artillery during the war and then trained in international law. After running unsuccessfully for the Chamber of Deputies in 1924 as a Poincaré conservative, he switched to the Radical Party. In 1928 he won a deputy seat from the Savoie. A brilliant orator with a streak of brashness, Cot soon became a rising star in the Radical Party. "Make a friend of him," Aristide Briand advised the British diplomat Philip Noël Baker in the early 1930s. "He will be premier of France not once, but many times."^[44] Cot quickly became a leading spokesman for the so-called Young Turks of the party. As good Radicals, the Young Turks were republican and anticlerical. But the world war had undermined their faith in economic liberalism; by the early 1930s they felt an attraction to antibourgeois, antiliberal and anticapitalist ideas that set them apart from the mainstream of the party. After 1933, when

the cleavage widened again between left and right, the Young Turks too would diverge: some, including Pierre Cot, Pierre Mendès France, and Jean Zay, would move left; others, like Bertrand de Jouvenel and Gaston Bergery, would move right. But between 1930 and 1934 they shared a common quest for a "rejuvenated" politics, a program to escape the "tiresome dichotomies" of collectivism and capitalism, left and right.^[45]

In this atmosphere of ideological inventiveness Cot tried to blend the moderation of party elders with the irreverence of his own generation. He affirmed the traditional aims of the party—as he put it, "to protect the independence of the individual and to permit the full development of the human personality"—and the need of the party to remain committed to its traditional constituency of small producers, peasants, artisans, and liberal professionals. With greater stridency than most Radicals, however, Cot condemned "the moral bankruptcy of the bourgeoisie," which he viewed as a grave threat to the individualism and independence he had vowed to defend. "Our bourgeoisie," he wrote in 1931, "knows not what to do: neither to adapt itself, reform itself, rule events, nor revolt against itself; it neither understands, nor commands. . . . The people of France no longer have the elite they deserve." He admired Socialists for proposing alternative policies, implying that Radicals ought to build bridges to the left: "If it is absolutely necessary to choose between socialism and capitalism in the name of individualism we choose socialism. We are Radical-Socialists and not Radical-Capitalists."^[46] In practical terms, as far as Cot was concerned, this meant nationalizing monopolies (in the Socialist tradition), sponsoring cooperatives (in the Solidarist tradition) and struggling against financial speculators (in the interest of small savers). Without offering a full-blown political program, he nonetheless sought a perspective whereby socialism, Radicalism, and technocratic interventionism might converge. That he could articulate such a position on the left fringe of the party while retaining respect from the party's old guard testified to his political skill.

As air minister, Cot promptly confirmed his reputation for boldness by rushing headlong into the nasty thicket of the airline question. In December 1932, two months before he took over the ministry, parliament passed a statute giving the state greater control over equipment procurement by the airlines. Most officials believed that the statute would affect only one company, Aéropostale, which the state was trying to salvage from bankruptcy. Cot thought otherwise. He viewed the statute as a device to reshape the entire airline industry. In April 1933, he addressed letters to each of the five airline presidents announcing that as a condition for further subsidies their firms must merge into a single company. Naturally, this directive, as one commentator understated, "raised a stir in aeronautical circles."^[47] But by 1933 state subsidies had

come to account for 80 percent of the income of the airlines, and the depression was only crippling the firms further.^[48] Faced with a choice between bankruptcy and the merger, the airlines complied with Cot's demand, and by June his efforts had given birth to Air France. State subsidies could now be distributed on a more rational basis, and routes added or eliminated, without concern for the short-term solvency of several firms. What Raoul Dautry had only imagined in 1928 became a reality through Cot's administrative guile.

Although Cot may have envisioned nationalizing Air France at some future time, he was careful to establish it as a private firm. His merger agreement allowed private stockholders to transfer their holdings in the constituent firms into comparable shares of Air France, whereas the state was to own only 25 percent of the stock. Despite Cot's efforts to reduce the influence of the major airplane manufacturers in the airlines, the builders managed to secure one-fifth of the directorships on the Air France board.^[49] Of course, the Air and Finance ministries had real leverage in the firm, through stock power and subsidies. But the new Air France was still a far cry from the nationalized firm it was eventually to become in 1945. Technically speaking, it differed little in structure from other companies—Crédit National, the Caisse Nationale de Crédit Agricole and the Compagnie Française des Pétroles—in which the French government had become an important minority stockholder.^[50] Still, Cot's success in breaking through the impasse in commercial aviation was no mean achievement considering how hostile some of the airline executives were to the merger.

Cot also made strides in winning more autonomy for the air force. Just as the depression had given him greater leverage with the airlines, so too another external shock, Hitler's rise to power, strengthened his hand in military aviation. Working closely with General Victor Denain, the air force chief of staff, Cot secured approval for a decree giving the air force its own general staff. Issued on 1 April 1933, the decree granted air force generals a command status equal to that of their counterparts in the army and navy; it also established a system of "air regions" supervised solely by air force officials. For the first time an air minister referred to the air force as l'armée de l'air .^[51] The festering argument over air doctrine, however, remained unresolved; Cot's decree simply identified three objectives for the air force—aerial operations, territorial defense, and tactical support for the army and navy—an ambiguous mission that sidestepped debate over strategic bombing.^[52] All the same, the decree gave the air force a measure of independence, and hence the Air Ministry a bit more legitimacy as a claimant for credits in the budget for national defense.

Fear of a rearming Germany also enabled Cot and General Denain to

win approval for the first of what would prove to be several plans for building new aircraft. Plan I, also known as the Plan 1010, called for constructing 1,010 airplanes over a two-year period to replace the outmoded frontline aircraft of the air force.^[53] The Superior Air Council approved Plan I in June 1933; parliament appropriated credits in 1934. According to the production schedule of the plan, France would have a refurbished air force by 1936, just about when General Denain expected Hitler might contemplate war.^[54] Plan I, then, marked the first serious effort at aircraft rearmament. It signaled a shift in the Air Ministry from a preoccupation with prototypes to an effort to balance research and mass production. This shift, in turn, gave the Air Ministry a fresh opportunity to use production contracts to encourage firms to decentralize their plants and rationalize the competitive structure of the industry.

Cot hoped that this change in the industrial priorities of the Air Ministry would open the door to reform, and he proceeded accordingly. In the Matériel Committee, where representatives from the air force and the Finance Ministry met regularly with Air Ministry officials, Cot spelled out his strategy for improving the industry. He insisted that the time had come for scaling down prototype production, for simplifying the most promising new models, and for equipping only the strongest firms for mass production. Henceforth, he argued, contracts should go only to "approved firms well experienced in manufacturing new equipment most efficiently from the point of view of simplicity and low cost."^[55] The production of new planes for training and commerce should be, as he put it, "as simple as possible and subject to free competition and minimal regulation." Cot also called for restoring closer connections between prototype building and series production by giving large firms, in effect, two contracts—one for the prototype, another for raw material and machinery to construct a small series of ten planes.

Through this shift in policy Cot and his colleagues in the Matériel Committee sought both to strengthen the supervisory role of the Air Ministry and to encourage large firms to function more competitively. This strategy required a sharper division of labor: the ministry would set production goals, define standards, and enforce contracts, and firms would meet these objectives as best they could. Under the new policy the military would reimburse firms for prototypes only if they were successful, thus shifting the risks of innovation from the state back to the enterprise—risks that only the largest firms could afford to assume. Feeble firms that had survived on prototype contracts alone would now, presumably, fall by the wayside. This tough stance aroused little debate in the Matériel Committee. Embarrassed by past errors and now pressured to meet the new deadlines of Plan I, government officials agreed in principle to promote series production in the largest firms.

If Cot found it was easy enough to secure official approval for a new industrial policy, achieving concrete results was another matter. To break the old habits of favoritism, to abandon the deeply institutionalized practice of keeping most firms afloat, and to overcome what Cot called "the spirit of routine," the Air Ministry needed money, political leverage, and time.^[56] All three assets were hard to come by in 1933. As long as the Radical Party remained committed to deflation as remedy for depression, and as long as the air force remained the weakest branch of the military, the Air Ministry would continue to operate within severe financial constraints. As a junior cabinet member of a timid government, Cot was in no position to change these conditions.

Nor was Cot in a strong position to prevail over Louis Bréguet, Henry Potez, and the rest of the aircraft fraternity. Although he charged pell-well into the world of aviation when he assumed his post at the boulevard Victor, even to the point of learning to fly, he remained an outsider to the industry. This status was obviously to his advantage if he hoped to shake builders and bureaucrats out of their complacency. But it would take time and support for an outsider to make headway—time to win allies within the industry and the bureaus of the Air Ministry, and support from the highest levels of the government, especially from the premier and the Finance Ministry. In an era of weak government and ministerial instability, time and support were not in the offing. Cot survived two cabinet collapses in the course of a year, but when the Daladier government fell after the bloody night of right-wing rioting on 6 February 1934, Gaston Doumergue put together a more conservative cabinet and made General Denain the new air minister. It thus fell to Cot's military associate, an air force insider with more conservative views, to implement Cot's program of reform.

Failure Exposed

The street fighting of 6 February triggered much more than a cabinet change; it marked the first fusillade in what some historians have called an undeclared civil war in France, a polarization of left and right that threatened the stability of the Republic until 1937.^[57] Right-wing extremists rebelled against the ineffectuality of the republican center. Their rebellion, in turn, prompted Communists, Socialists, and Radicals to forge a popular front against fascism at home and abroad. As the political temperature rose, many people gradually altered their views on foreign policy and national defense. Many leaders on the left drifted away from pacifism and toward an enthusiasm for national preparedness, whereas many of their counterparts on the right swung away from a traditional hostility toward Germany to defend appeasement. This re-

versal only began to evolve in 1934; it would take another three years to complete. But almost immediately after the February demonstrations it was easy to see that France, as the British journalist Alexander Werth wrote, "was in an acute state of 'nerves.'"^[58] Political polarization made the center-right cabinets of 1934 and 1935 all the more stalwart in their conservatism. Such rigidity could not help but affect the way government officials and businessmen addressed the unsolved problems of industrial reform.

This changing political climate made the new air minister, General Denain, all the less likely to adopt Cot's healthy disrespect for the conventional boundaries of state intervention. Denain, unlike Cot, had little interest in pondering the difficult questions of economic reorganization. He was a war hero and an energetic promoter for the air force. He had risen rapidly through the ranks, and by the time he became air minister at the age of fifty-four, he had acquired an easy familiarity with nearly everyone of political consequence in the world of aviation—generals, ministers and industrialists. Ambitious builders like Marcel Bloch and Henry Potez felt comfortable with the man.^[59] Denain no doubt cared deeply about Plan I and the fate of the air force, but his style was to grease the machinery of state-business relations rather than tinker with its structure.

The contrast between Cot's irreverence and Denain's accommodation with the business world became apparent in the way Denain handled the airline question. Just when it seemed that the creation of Air France had brought some order to civil aviation, Denain mired himself in the Air Bleu affair. In July 1935 Louis Renault and two former administrators from Aéropostale, Beppo di Massimi and Didier Daurat, inaugurated Air Bleu, an airline providing postal service between six major cities in France. They designed the scheme shrewdly: Georges Mandel, the postal minister, granted them exclusive rights to domestic postal routes, and Denain gave them subsidies to buy a fleet of Simoun aircraft build by Louis Renault's aircraft division, Renault-Caudron. Thirteen months later, the airline collapsed. Just as Louis Bréguet and other unsympathetic officials at Air France had warned, Air Bleu could not compete with the overnight trains, which could carry the mail more cheaply. It was a fiasco in state-business collaboration worthy of the 1920s: the Air Ministry had once again embarrassed itself by supporting costly, unviable routes, and Denain found himself open to charges by the left that he enriched his friends at public expense.^[60] The Air Bleu affair did little for the effort to establish the ministry as an august authority safely removed from profiteers.

Of greater importance was General Denain's failure to advance the cause of industrial concentration. To be sure, the air minister made ges-

tures in this direction. In July 1934 Paul Dumanois, a distinguished state engineer who as industrial and technical director had assumed Caquot's role in the campaign for industrial reform, presented a new plan to the builders for revitalizing the airframe sector. His plan, addressed to the Chambre Syndicale, called for accelerating the drive to locate factories in the provinces, adopt modern machinery, and consolidate the airframe sector into five or six firms. Dumanois wanted genuine mergers, "a unification of administrations," and not the phony mergers that the earlier groupements strategy had disguised. To press his point, he warned the Chambre Syndicale that Joseph Caillaux, the chairman of the Finance Committee of the Senate, had lost his patience with the industry and had recommended "the purchase of foreign airplanes and the need for tightening credits and reducing orders from French industrialists."^[61]

The leaders of the Chambre Syndicale responded to the Dumanois plan with characteristic finesse. Henry Potez, Chambre president and by now the most accomplished builder in the airframe business, endorsed the call for concentration. He warned his colleagues, prophetically, that "if the Chambre Syndicale does not find a concrete solution, [the air minister] himself will proceed with his own program on account of the pledges made before the Senate."^[62] The Chambre Syndicale saw fit to follow Potez's heed. By the fall of 1934 builders had put together a new configuration of six groupements , which presumably would go further toward merging firms. In practice, however, this scheme fell far short of genuine concentration. Eight firms either dragged their feet or refused outright to enter the new groups. Firms that did cooperate still maintained their autonomy within their groupement , preserving their own fixed capital, production facilities, research laboratories, and administrative services. Each group created a central bureau to parcel out orders to member firms, negotiate contracts, and search for foreign clients.^[63] In effect the groups served more as marketing cartels than as central administrations, and in this respect they differed little from the two large groupements they replaced.

By the end of 1935 the structure of the industry had still scarcely changed: only two of the groups, Potez-Bloch-Cams and Farman-Mureaux-Blériot, even functioned in accord with the new plan of the Chambre Syndicale. By 1936 seventeen firms still preserved full autonomy behind the facade of the company groups.^[64] If a handful of the feeblest firms was beginning to fall by the wayside, the basic problem of fragmentation remained.^[65] The Air Ministry had once again failed, and employers had wasted what would soon prove to be their last chance to reorganize the industry on their own.

If Denain and Dumanois ran up against the same wall of resistance that Albert Caquot had encountered years before, at least they had an

opportunity to use the handsome production orders of Plan I to encourage encourage the strongest firms to enlarge. But here too Denain failed. Like his predecessors, he could not resist the temptations to spread out the largesse. In November 1934, for example, he gave contracts for building Bloch 200s not just to the two firms his staff had recommended for the job but to four, apparently to give more firms the benefits of new orders. In the fall of 1934 Denain and Dumanois clashed openly in the Air Ministry's Matériel Committee over which firms should be granted new orders.^[66] By April 1935 the practice of dissipating orders had become so common that an angry and beleaguered Dumanois told Denain one last time that everything should be done to concentrate orders in a few firms and when possible grant serial production contracts to the mother firm, that is, the company that had invented the original prototype.^[67] Soon thereafter Dumanois resigned. Years later he would accuse Denain of practicing favoritism in awarding contracts and of undermining the effort to decentralize factories.^[68]

Conflicts between Dumanois and Denain crystallized tensions that had been building up in the Air Ministry. As a proud member of the state aeronautical engineering corps, Dumanois epitomized the staff professional of the ministry—technocratic, distrustful of industrialists, eager to enforce standards. When in 1934, for example, he discovered that Renault had failed to build fifty motors according to specifications, he proposed to Denain that the contract be canceled, "having believed frankly that after the sixth of February some things would have changed."^[69] Denain preferred, however, to let the matter slide and continue Renault's contract as usual. Other conflicts of this ilk, pitting technical standards against company interests, led Dumanois to view the ministry as divided into factions—with Denain and his personal staff, on the one hand, willing to yield to business pressure, and professional engineers, on the other, alone committed to protecting the interests of the state. Though Dumanois obviously had his own axe to grind, he no doubt spoke for other engineers whom Denain had dismayed by failing to force industrialists to reform.

As long as the controversy remained within the ministry, little was likely to change. But when the airplane companies fell behind the schedule for Plan I, the bankruptcy of the reform effort became obvious to everyone privy to procurement policy. By January 1936 the industry had produced only 588 of the 725 planes scheduled for completion, and ministry officials predicted the builders would fall even further behind by the summer.^[70] Design changes had created some of the delay, but the main cause was clear: the industry, wrote a liaison officer in the Air Ministry, "was not equipped to respond to massive orders."^[71] A large cluster of undercapitalized firms that still relied on artisanal methods simply

could not build the airplanes of Plan I in the time required. In March 1936 the Air Ministry had to revise its deadlines so that by the end of the year the companies could deliver what had been originally anticipated for 1935.^[72] What made the delays doubly painful, moreover, was the knowledge that Plan I had become pitifully inadequate to the country's needs. As early as 1934 intelligence reports had concluded that French air superiority would disappear by the end of 1936. Worse still, the airplanes of Plan I would be outmoded as soon as they left the factories.^[73] Beyond Plan I lay the larger challenge of producing a new generation of airplanes in numbers several times larger than that plan had entailed. Plan I, which looked like a boon to builders in early 1934, exposed how little they had done to adapt to the needs of national defense.

The builders, of course, bore a hefty share of the blame for the failure to modernize the aircraft industry. Together they had conspired to avert the Darwinian struggle that a serious wave of mergers, acquisitions, and bankruptcies would have involved. They made little effort to lobby the government to invest in reequipping the industry. Since 1928 they had clung stubbornly to a strategy of collective defense—a strategy that, though effective in the short run, left them increasingly vulnerable to government critics. When Marcel Déat, the neo-Socialist and future right-wing extremist, became air minister in Albert Sarraut's government in early 1936, he expressed his exasperation with the builders, and no doubt that of many other people, in no uncertain terms in front of a Senate subcommittee: "I cannot have confidence in my manufacturers. The groupements are purely a facade, and the aircraft industry as such does not exist. You can't build on moving sand. . . . As for the manufacturers, they work when they please, practice blackmail, play with layoff threats and the rising price of raw materials. It really is, excuse the word, a filthy mess [pétaudière ]."^[74]

If in retrospect, however, it is easy to see that employers were short-sighted in resisting Caquot's invitation to restructure the industry on their own, at the time it clearly made sense from the builders' point of view to stick to their defensive strategy. Aircraft manufacturing was still as precarious a business in 1935 as it had been in the late 1920s. Handsome profits could be made, but only sporadically—with a big production order, a coveted export contract, or a successful prototype. Because the irregularity of demand made it hard to amortize the costs of plant and equipment, builders had little incentive to invest in factory expansion or the latest machinery; it seemed wise to minimize capital expenditures, especially in periods when orders were scarce.^[75] Most of the time companies tried to finance yesterday's payroll on tomorrow's delivery. Cash flowed, and builders lined their own pockets, but the companies remained financially insecure.^[76] In another business owner-manufactur-

ers might have sold out for more promising vineyards. But aviation was different: most manufacturers were in it to stay, having long since accepted its uncertainties for the chance to make their own airplanes. Men who derived such prestige from their products, and often enough real personal wealth, placed a high premium on their autonomy.^[77] Under these circumstances Malthusian business practices—dividing up markets rather than trying to expand them and compete for larger shares—made sense to these men in the short run, even though they ran the risk, as Henry Potez warned, of mortgaging their independence down the road.

Yet it would be a mistake to view entrepreneurial conservatism as the chief source of the ills of the aircraft industry. For one thing, the younger generation of builders—Potez, Bloch, and Dewoitine—had great ambitions, as their work in the late 1930s would eventually show. Even some of the older pioneers might have expanded more aggressively in the early 1930s under different market conditions, as Louis Bréguet had tried to do in the previous decade until he was stymied by government obstacles. What made Malthusianism a rational strategy was the willingness of the government to sustain it. Air ministers from Laurent-Eynac to Denain had all called for industrial reorganization; none took effective measures to promote it. From Caquot's prototype program to Denain's favoritism, the Air Ministry had preserved rather than transformed an outmoded industrial structure. Whereas builders followed a logic of short-term survival, government officials were stuck in a shortsighted logic of their own. Laurent-Eynac, Denain and even Cot all assumed that structural reform could be gradual, that it could accompany rather than precede rearmament, and that the builders could be persuaded to take up the cause of reform as their own. Cautiousness at the top, moreover, played into the hands of air force and Air Ministry bureaucrats, who did not wish to see any one firm emerge as too powerful a force in the industry. By the end of 1935 the bankruptcy of government policy had been laid bare—so much so that Denain came under heavy criticism on the floor of the Chamber of Deputies. "There really has been no industrial policy in the Air Ministry," Antoine Brocard, a center-right deputy from the Isère, said to Denain on 17 December. "You have pursued precisely the opposite of an industrial policy. . . . So it is not surprising that our aircraft industry, after having been overwhelmed by orders of considerable quantity, finds itself a prey to the worst difficulties."^[78] Deputies across the political spectrum applauded after Brocard's remarks, a sign of the rising tide of impatience in parliament over the impasse in aeronautical reform.

In criticizing Denain, however, Brocard neglected to say that industrial policy was itself a hostage of the same fiscal conservatism and military antipathy that had plagued every air minister since 1928. Despite

Plan I, budgetary allocations to the air ministry actually declined from 29 billion francs in 1933 to 24 billion in 1936, largely because of the deflationary policy that the conservative governments of Doumergue, Laval, and Flandin upheld.^[79] The high command, moreover, remained just as antogonistic to the air force as before. Under Denain's ministry the air force even lost some of the autonomy it had won under Cot: General Weygand reestablished the right of the army to command more than 86 percent of the aerial forces in the event of a war.^[80] In 1936 the air force still garnered only 18 percent of the defense budget.^[81] In contrast to Britain, where the Royal Air Force was able to win a serious commitment to aerial rearmament in 1935, in France financial and military policy would continue to hamper the Air Ministry until late 1937.^[82]

If conservative economic policy and hostility in the general staff prevented the government from investing in the modernization of the industry, might not the Air Ministry still have been able to force the builders to consolidate firms? In principle nothing was to prevent a crusader at the boulevard Victor from using the leverage of contracts to force firms to merge. But political realities stood in the way. Laurent-Eynac, Caquot, Denain, and the bureaucrats they presided over in the Air Ministry lacked the autonomy to carry out rigorous reforms. These men were too closely tied to the conservative business culture of aviation to overcome the resistance of the manufacturers. As Cot later wrote, "Most of the civil and military officials of the [Air Ministry] were honest people who would have resisted every bribe. But it was impossible to establish watertight bulkheads between administration and purveyors; social or family relations and friendships grew up and destroyed the absolute independence that should have existed." Cot understood how tough it was to break the habits of patronage: "The 'pioneers' of the aeronautical industry . . . thought that they had a lifetime or hereditary right to partake of the manna of state orders. Their bankruptcy would have been considered scandalous."^[83] Despite the importance every Air Ministry clearly attached to decentralizing the factories, consolidating firms, and modernizing production methods, air ministries lacked either the autonomy or, in the case of Pierre Cot, the political support of a strong regime to reorganize the industry.

Thus the failure of industrial reform had its roots in the ability of the builders to exploit an entrenched pattern of state patronage that government officials were either unwilling or unable to change. This pattern had a distinctly French character. In the United States a small number of large firms emerged in a competitive struggle to capture the market for commercial aircraft. In such a huge country there was a sizable demand for commercial aircraft, especially to carry the mail. By 1933, Boeing, Douglas, and Curtiss-Wright had already emerged as the leading firms;

civil aviation, not the military, propelled the industry. British airplane manufacturers, by contrast, were much more like the French: they depended directly on an Air Ministry that as a matter of policy supported an artificially large number of firms. But in contrast to France, in Britain there was a modest movement toward concentration in the aircraft industry, centered around the Hawker Company. More important, British manufacturers and government officials did not engage in a protracted struggle over industrial policy. The British Air Ministry did not try to play midwife to a process of industrial reorganization. Once rearmament had got under way after 1934, the ministry expanded production by subcontracting work to a wide variety of metalworking firms rather than restructuring the industry.^[84]

France in the 1920s and early 1930s was, as Stanley Hoffmann has argued, a stalemate society. Economic interest groups relied on weak regimes and stable bureaucracies to protect their established position. To be sure, France's centralized state bureaucracy had a good deal of potential power to implement change; but as long as the political balance of power vacillated innocuously between weak center-left governments and weak center-right ones, the predictable dependencies that had developed in the aircraft industry survived. For all the rancor between reformers in the Air Ministry and the builders in the Chambre Syndicale, in the end the reformers were stymied by a regime that as Hoffmann has written, "had plenty of brakes and not much of a motor."^[85]

Had airplane manufacturing been merely a commercial affair, politicians and bureaucrats might have tolerated the impasse over industrial reorganization a good while longer. But German rearmament kept everyone's feet to the fire. By the winter of 1935–36 even conservatives in parliament were expressing frustration with the recalcitrance of men like Louis Renault who were successfully rebuffing government promptings to move some of their operations away from Paris.^[86] Paradoxically, after spending years resisting the injunctions of Albert Caquot, Pierre Cot, and Paul Dumanois, manufacturers were beginning to exasperate even the die-hard defenders of laissez-faire. When the left-wing parties of the Popular Front coalition issued a call, as part of their electoral campaign in the spring of 1936, for the nationalization of the arms industry, the spontaneous impulse of most conservatives was to dismiss the notion out of hand. But as events would soon show, much of the confidence in the conventional framework of state-tutored private enterprise had disappeared. By squandering their credibility and resisting reform, the builders unintentionally set the stage for a revolution in state-business relations that would follow the coming to power of the Popular Front.

Two—
The Revival of Working-Class Militancy

In early May 1929 police informants in Communist cells throughout Paris reported that militants were laying elaborate plans for disrupting the air festival scheduled for 20 May in the Bois de Vincennes.^[1] The festival, sponsored by the Aéro-Club de France and the new Air Ministry, seemed ideally suited for an antimilitarist demonstration: air races and acrobatics would conclude with a mock air battle and the aerial bombardment of an artificial village. What better chance, Communists felt, to call attention to the horrors of "biochemical warfare," the "false pacifism" of French Socialists, and the "imperialist and antiworker aims of the bosses and the French State."^[2]

A few days before the air festival Communists at Farman, Gnôme-et-Rhône, Blériot and other Parisian aircraft plants passed out leaflets inviting workers to join the protest. In L'Humanité the PCF ran articles on the many aviators who had died for war or profit and on the "militarization" of factories preparing planes for the next "imperialist butchery."^[3] Meanwhile Le Figaro assured its better-heeled readers that the air festival would proceed as planned since police promised to arrest anyone caught so much as leafleting on the scene.^[4]

The authorities need not have worried. The Communist appeal fell on deaf ears; only seventeen demonstrators appeared on the festival grounds, and police promptly grabbed them, leaflets in hand. As thousands of Parisian families witnessed the most awesome display of aerial firepower since the war, no protest intruded. To call a demonstration and have nobody come was symptomatic of the isolation of the Communist movement and the political demobilization of the French working class in the late 1920s.

By early 1936, however, a different political climate had come to prevail in these same airplane factories. The labor movement had returned

to life. Although the shops of Blériot and Farman and Gnôme-et-Rhône were still a far cry from the Communist "fortresses" they would later become, militants were finally establishing viable union locals. Calls for wage hikes and collective bargaining became common again. The strike threat reemerged as a worker's weapon. Of course, this shift from the apparent quiescence of 1929 to the militance of 1936 was common to many industries in France. But nowhere was it more dramatic than in aviation. There workers suddenly found the means to challenge management in firms where employers had been used to ruling like feudal lords. In the course of a mere two or three years aircraft workers transformed their industry from a bastion of employer autocracy to what by the summer of 1936 would become a major battleground of industrial conflict. Changes in national politics, in Communist strategy, and in employment conditions within the industry would enable labor militants to close much of the gap that had separated them from workers during that abortive protest in 1929.

The Autocratic Firm

Few observers in the early 1930s would have predicted that aircraft workers would soon emerge as a vanguard of the labor movement. In contrast to Britain, where aircraft unions remained intact during the 1920s, in France aircraft unions failed to recover from the collapse of the labor movement after the First World War.^[5] Industrial demobilization in 1919, the schism of the CGT in 1921 into Communist and non-Communist confederations, and the postwar stagnation of the aircraft industry made it nearly impossible for militants to rebuild their unions. Although most airplane factories were located in cities that had a history of metalworking militancy—Paris, Toulouse, Bordeaux, Nantes, and Le Havre—workers remained at the mercy of their employers and a slack labor market. The strike records tells much of the story: aircraft workers launched only ten strikes in France from 1927 through 1933, eight of which failed.^[6]

One reason why aircraft workers had such a dismal record of collective action between 1919 and 1934 was that builders were able to maintain a stern, autocratic style of managerial authority. The aircraft industry lent itself well to employer autocracy. Men such as Henry Potez, Louis Bréguet, and Emile Dewoitine towered over their employees as owner-inventors—men who possessed the money, political connections, and especially the engineering prowess on which the life of the firm, and everyone's job, depended. These men had mastered the science of flight; more often than not they designed their own airplanes, working closely with the engineers, draftsmen, and craftsmen who built the prototypes.

Their authority was personal, direct, and anchored in their own expertise. Although most of the major builders established posh administrative headquarters in Paris to be near the ministries, they spent enough time in their plants to make their presence felt. Even if the employer was disliked or resented, as a builder he commanded respect. As one engineer put it, "when a man like Bréguet, . . . who had flown since 1909 and who then had invented planes every two or three years, stands in front of three hundred or four hundred employees, these people will follow him, a patron with a capital P."^[7]

Of course, following the lead of the patron came more naturally to some employees than to others. The top design engineers, factory directors, and test pilots enjoyed the most privileged status in an airplane company. They either shared in the tasks of designing and testing aircraft or, in the case of the factory managers, served as surrogate authorities for the patron . By the 1930s most of the men in the top posts in a company came out of the same engineering schools that the leading builders had attended, although one could still find such anomalies as M. Rowse, a brilliant autodidact who had spent much of his life as a chauffeur for Henry Potez's father before rising to become the son's plant director.^[8] By virtue of their posts and their backgrounds top personnel were not inclined to question the autocratic hierarchy of the firm.

Beneath these professionals at the summit of the company was a host of white-collar employees—structural engineers and aerodynamists, designers and draftsmen, accountants, purchasing agents, time study specialists, and a variety of other technicians, clerks, chauffeurs, factory guards, and secretaries—all of whom stood in more ambiguous relationship to the boss. The white-collar staff made up a larger portion of the work force in the aircraft industry than in any other part of the metalworking sector. In some firms up to 40 percent of the work force took part in research, design, planning, flight testing, and other operations at least one step removed from production. The complexity of airplane manufacturing demanded such participation. At Farman, for example, 150 engineering draftsmen stooped over desks drawing plans not only for the eleven thousand different pieces that made up the airframe of the Farman 221 but also for the nine thousand machine tools and patterns that workers would use to make them.^[9] The white-collar staff of a company worked mostly in large design studios, research laboratories, or planning offices where technicians converted designs into production procedures.^[10] Middle-class, paid by the month, and often highly trained, these people were identified as techniciens or employés , rather than as workers; but they lacked the high salaries, prestige, and authority that gave the top engineers and managers such privileged status in the companies. And like workers, they were subject to job insecurity because of

the danger of layoffs that loomed in a troubled industry like aircraft manufacturing.

Blue-collar workers, comprising from 60 percent to 80 percent of the work force, had the most contradictory relationship to the employer: they had enviable jobs, in comparison to the drudgery of the assembly line or the textile mill, but they were still subject to the heavy hand of autocratic authority. In the early 1930s airplane construction was still largely artisanal; skilled carpenters, metal fitters, machinists, and sheet metal workers made up the bulk of the work force. Some of these workers were lucky enough to work in prototype shops, where one could find "interesting, creative work," as one metalworker later recalled, and "where everyone was his own master: the foreman handed me a packet of designs and I made an entire airplane tail."^[11] Prototype shops had the aura of an inner sanctum, fenced off from outsiders (and industrial spies) who might glimpse the firm's latest model. Here workers enjoyed higher pay, greater prestige, and better working conditions than their counterparts in mass production. If the small-shop ambiance of the early days of aviation survived anywhere in the early 1930s, it was in these prototype shops, where skilled workers, technicians, and engineers cooperated in the process of building a new plane.^[12]

Even mass production had its artisanal flavor. At first glance the big assembly sheds of an airplane factory might have resembled the long, noisy halls of an automobile plant. But the similarity was misleading. After the First World War orders for airplanes were never of sufficient magnitude to justify the expense of adopting the assembly-line methods and employing the huge armies of semiskilled workers that eventually became common in the auto industry. From 1918 until the late 1930s aircraft construction was less industrialized than it had been during the First World War. Manufacturing the Farman 165 in the late 1920s, for example, required a work force in which only 21 percent of the workers were semiskilled or unskilled.^[13] Skilled workers predominated, not just in the toolmaking and maintenance departments of the company but also in the manufacturing process itself. Most of the time they worked in teams headed up by a chef d'équipe (team leader), or they labored side by side at machinist benches or assembly jigs. It was a looser form of work organization than the assembly line, allowing for a sense of companionship on the shop floor. "What camaraderie there was!" one worker recalled. "Everyone knew one another. We loved this work. First of all, it was clean. . . . I liked that. You were better off than in automaking."^[14]

A look at the floor plan of Bréguet's assembly plant in Bouguenais, near Nantes, reveals how highly skilled the production process remained even as late as 1937 (chart 1). The huge factory shed had three large work areas, each several hundred meters long and often echoing the

Chart 1.
Plan of the Bréguet factory at Bouguenais, 37,380 square meters. Derived from a blueprint deposited in 
the archives of the Société Nationale des Constructions Aéronautiques de l'Ouest. Not to scale. AN 99AQ 8.

deafening sounds of noisy machinery. In the front workshop (bottom of chart 1) several hundred skilled workers labored at forges, lathes, milling machines, drill and punch presses, metal-stamping, -shearing, -nibbling, and -stretching machines, and other metal-fabricating equipment to create the thousands of pieces that made up an airpláne. Some of these workers made tools, templates, and jigs that other workers used in the construction process. Others made forgings, castings, fittings, and metal sheets for the airplane itself. Moving deeper into the shed, past a narrow band of offices for storage and supervisory control, one entered a second enormous workshop, the partial assembly hall. Here fitters, welders, electricians, sheet metal workers, sheering-machine operators, and other skilled workers assembled the major subsections of the aircraft—wings, fuselage, cockpit, rudder, and fins. By the mid-1930s the modern fuselage had become a highly complex metal shell, composed mostly of aluminum and an aluminum alloy skin wrapped around a framework made up of a huge number of pieces serving as spars, bulkheads, and ribs. Although some semiskilled workers were involved in these intermediate stages of construction, the work called mainly for skilled work teams. In the third large hall, at the far end of the plant, the fuselage acquired the wings, tail, cockpit and other subassemblies that had been built in the second hall. Skilled fitters and sheet metal workers collaborated with less skilled riveters and assembly workers to complete these final stages of production. After assemblers mounted the engines, propellers, and accessory equipment on the airframe, the plane was rolled to the paint shops along the left side of the shed. Once painted, it was ready for the test flights and inspections required before delivery.^[15]

If skilled workers predominated in the tasks of building an airplane, and if they often labored in congenial work teams, the managerial hierarchy nonetheless found its way onto the shop floor. Regimentation diminished the artisanal character of the work. Unlike prototype workers, production workers received detailed instructions from the bureaux de méthode describing each step of their jobs, the tools to be used, and the time to be spent on each task.^[16] As a result, "interesting, creative work" and, above all, a sense of autonomy were rarer in the production shops. Workers were heavily supervised. In the Bouguenais plant, as in others, no major work area fell outside the line of vision of an office of contrôle . A bevy of inspectors from the company, the Air Ministry, and client firms often joined foremen in roaming the floors. In some cases policemen and military officials stood guard as well. Workers could escape surveillance only in cloakrooms and lavatories—or at the bicycle sheds, where militants had their best chance to call impromptu meetings. Even the orderliness of the shops conveyed the message of managerial control. Far from the dark and disorderly machine shops typical of the metal-

working trades, aircraft plants had a clean look—uncluttered, ventilated, and well lit.^[17] Workers undoubtedly found small, subtle ways to put their personal stamp on their work space and their everyday work rituals. But through it all, managerial discipline prevailed.

Much the same story could be told about engine-building firms like Gnôme-et-Rhône and Hispano-Suiza. Like airframe companies, engine firms divided production into prototype building and mass production, with the same consequences for the work process and shop floor experience. Engine building also involved a progressive assembly process, beginning with forgings and castings, proceeding to subassembly activities, and culminating in final assembly. At each stage skilled workers played the dominant roles, backed up by semiskilled workers who operated machines, inspected pieces, or participated in highly subdivided assembly tasks. In the early 1930s engine construction involved even fewer semiskilled workers than did airframes. All the same the ambiguities of shop floor experience—the tension between regimented work rhythms and highly skilled labor, between informal interactions and tight supervision, between a sense of participation and a sense of exclusion—posed problems for airframe and engine workers alike.

For both kinds of workers, the tension between artisanal work and regimentation found expression in the wage system as well. On the one hand, wage schedules in aircraft conformed to conventional differentiations in the metalworking branch between skill levels and between women and men. To build the Farman 165 in 1927, for example, the Farman brothers paid production workers anywhere from 6.10 francs an hour to 3.10, depending on the skill level and gender of the employee (table 1). These wages reflected prevailing rates in Parisian metalworking since employers hoped to minimize turnover. In Bordeaux, Nantes, Toulouse, and other regional cities aircraft employers paid local rates, usually about 20 percent lower than Parisian wages.^[18] Even during the depression years, after 1930, skilled workers in aviation suffered less from pay cuts than many workers, partly because the depression hit the industry less brutally than most other sectors, partly because in metalworking skilled workers remained scarce.^[19]

Yet at the same time many employers in the aircraft industry were beginning to experiment with wage incentive systems that further undermined the artisanal nature of skilled work. Although aircraft work was poorly suited to conveyor belt assembly or time-and-motion studies, employers found that they could make more systematic use of output quotas and bonuses to push workers to move faster. The Emerson, Halsey, and Rowan incentive systems all provided formulas for calculating wages on the basis of hourly or daily production rather than through simple piece rates or hourly wages. In aviation employers tended to favor

the Rowan system since its formulas ensured that workers would earn no more than double their normal hourly wage by working harder.^[20] Wage incentive systems were unsettling for workers: one could earn more, but only by stepping up the pace and taking the risk of antagonizing fellow workers.

What threatened workers most, however, was job insecurity. The demand for labor fluctuated with the ebb and flow of the market, as annual work force figures show (table 2). The drop in employment in 1932 and 1933, reflecting a decline in commercial sales and stagnation in government orders, had the added cruelty of throwing workers into the precarious labor market of the Great Depression. The irregularity of Air Ministry orders was certainly a hardship for the manufacturer. But it also posed a direct threat to the livelihood of workers; it put them all the more at the mercy of the boss. It was no wonder, then, that between 1929 and 1933 strikes were almost unheard of in the industry.^[21]

Employer Paternalism

Workers encountered employers schooled in the two dominant traditions of managerial practice that still prevailed throughout France—authoritarianism and paternalism. Despite a few breakthroughs in factory legislation after 1890 and the presence of the Conseils des Prud'hommes (labor relations tribunals), which mediated between individual employers and workers, the authoritarian side of labor management had diminished little in France since the early nineteenth century.^[22] From the early 1920s through the mid-1930s aircraft manufacturers disciplined their wayward workers harshly: troublemaking, union organizing, and laziness could all lead to abrupt dismissal. As one worker recalled, Emile Dewoitine made a habit of shouting "No Soviets! No Soviets!" when he saw two or three workers together.^[23] In aviation, as elsewhere, employers spent the 1920s dismantling the gains workers had won during and shortly after the First World War—the eight-hour day, a shop steward system, and collective bargaining. Like most French employers in this period, aircraft manufacturers resisted collective bargaining for fear that it would diminish their authority and reduce their leverage in the labor market.^[24] Just as they saw even mild forms of state intervention as an attack on their independence, so too they abhorred trade unions as a challenge to their authority.

Paternalism, which also had sturdy roots in nineteenth-century managerial practice, both tempered and reinforced this authoritarian style, although the nature of paternal authority varied considerably from one firm to the next.^[25] Henry Potez pursued the paternalistic approach most

aggressively by running Méaulte, a small community in the department of the Somme, as a company town. Scarcely more than a village in 1920, Méaulte soon burgeoned when Potez used state subsidies to build housing and municipal services for the peasant workers he recruited for his factory.^[26] As in other company towns of France, workers in Méaulte came to depend on the firm for medical care, social services, and recreation as well as for an hourly wage. In no time workers and their families came to view Potez with that special combination of admiration, fear, and resentment typical of paternalistic relationships.

Méaulte, however, was exceptional. Most aircraft plants were located in cities where workers lived autonomously in working-class neighborhoods and rode bicycles, buses, autos, or, in the case of Paris, the metro some distance to work. In Marseille, Toulouse, Bordeaux, Nantes, and Bourges, and in many of the plants in the suburban western side of Paris, the trip usually took workers to the edge of town, where an airplane factory stood out conspicuously amid the landing fields of the company. This separation between factory and community limited the repertoire of paternalistic controls managers could employ. Even so, several practices had become common in most aircraft plants by the early 1930s. Employers sponsored sports teams, outings, and recreational clubs for their employees. Louis Renault in particular took a keen interest in what he felt were the positive effects recreation had on productivity.^[27] Social-welfare services, ranging from family allowances to health and retirement programs, became widespread in the metalworking sector generally.^[28] Above all, employers held out the promise of job promotions to strengthen workers' loyalty to the firm. Carpenters, mechanics, and skilled metalworkers who began building planes before or during the war often rose as high as chef d'atelier and in rare instances even chef de production . Foremen often came up from the ranks, as did instructors in company apprenticeship and retraining programs. Skilled workers in prototype shops could sometimes hope to win promotion to jobs in the research shops and testing laboratories.^[29] To be sure, the promotion system had its limits as a means of integrating workers into the firm, especially since some of the most competent workers felt competing loyalties to unions, Communist cells, or simply to a working-class ethic of keeping one's guard up against management.^[30] Moreover, as aircraft technology became more sophisticated and companies more complex, the meteoric rise of the ambitious worker was slowly becoming a thing of the past. Modest promotions, however, remained an important instrument of managerial control.

Paternalism was not confined to an employer's relationship with workers; it colored the way management dealt with engineers and technicians as well. In the hierarchy of the firm, engineers and technicians straddled

the boundaries between labor and management. Technicians, at the lower end of this middle-level occupational stratum, had much in common with skilled workers. They shared similar frustrations over the ambiguity of their status, the modesty of their salaries, job insecurity, and the limits to their autonomy in offices, workshops, and labs. Engineers, too, had occasion to grumble about the boss. Salaries, work procedures, and substantive matters of aircraft design all raised issues that potentially pitted engineers against the builder.

Aware that engineers and technicians occupied a strategic position in the firm, employers in aviation as in many industries encouraged them to identify themselves as a professional elite rather than as conventional employees. Some employers even viewed engineers as potential mediators in the war between labor and capital—an image Georges Lamirand captured in the treatise he published in 1932 on the social role of the engineer.^[31] Drawing on Maréchal Lyautey's famous conservative tract on the social role of the French military officer, Lamirand promoted the paternalistic notion that engineers could serve as mediators between workers and employers and eventually eliminate the conflicts between them. Loyal to management yet attuned to workers' concerns, engineers would combine both humanitarian zeal and disciplined rationality to reconcile both sides in the enterprise. With this outlook in mind, the faculty at the Ecole Supérieure de l'Aéronautique added courses on labor relations to the curriculum, and the staff at the Union des Industries Métallurgiques et Minières (UIMM), the employers' association, began publishing reports on the special problems engineers faced in all branches of metalworking. By the mid-1930s the pressures of the depression and the revival of the labor movement would make the engineer's identification with management an even more urgent concern.

Aircraft manufacturers made use of an additional method for promoting loyalty not available to employers in other industries—the cultivation of an esprit de corps based on participating in the excitement of aviation. The mystique of flight, the appetite of the public for tales of aerial conquest, the legendary heroism of pilots, inventors, and mechanics, and the patriotism associated with building French airplanes all contributed to a sense of purpose in the industry. To some extent this esprit de corps emerged automatically in a factory: no doubt many prototype workers took some pride in the successful test flight of a new airplane. A sense of being part of something important may even have filtered into the production shops as well. After all, in the 1920s and 1930s airplane construction represented the quintessential blend of advanced engineering and skilled craftsmanship. There was no shortage of writers, publicists, and politicians willing to portray aviation as a moral or patriotic enterprise. The utility of the ethos of aéronautique was not lost on employ-

ers, who encouraged employees to attend the biannual salon de l'aéronautique , where the wares of the company went on display.^[32] On the shop floor employers frequently posted signs to instill a feeling for the seriousness of the undertaking. As one poster read:

In our work an error, an omission, or bad workmanship can cost the life of one or more people.

Anyone who makes a mistake must report it. To commit an error is a forgivable and correctable mistake. But to conceal it is a crime.^[33]

The message here was twofold: the obvious one was to beware the dire consequences of error; the subtler one was to take pride in one's work, laboring, as one journalist said after visiting an airplane factory in Bordeaux, "without error, without weakness."^[34] Employers wanted their workers to perform as a labor aristocracy. As one former chef d'atelier recalled, managers encouraged workers to acquire an esprit d'équipe (team spirit) and an esprit de maison (company pride) and in so doing to view themselves as a laboring elite immune to the hostility that workers commonly felt toward their bosses.^[35]

To what degree workers actually embraced this ethos of company pride is hard to say. Since employers seasoned their paternalism with ample doses of coercion, it is impossible to know whether a cooperative employee felt genuine loyalty to his boss. Many workers, moreover, took pride in their work because discipline and good workmanship had an important place in their own craft traditions. Many metalworkers in interwar France paid homage to the cult of the métallo , a traditional sense of pride that derived, as one worker put it, from the belief that the metalworker "can [do] and knows how to do everything."^[36] The image of the earnest, muscular metalworker, with hammer in hand and eyes to the horizon, had become a standard icon in the trade union press. Employers obviously profited from this tradition; prototype shops thrived on the initiative of experienced, talented workers. But an artisanal tradition could also inspire other loyalties—toward unions, parties, or a vision of workers' control. Similarly, workers could easily become captivated by an esprit de l'aéronautique without necessarily feeling company pride. What is clear, however, is that even if workers did not sell their souls to the builder, even if they resented his authority, they had little choice but to comply with his rules. In the autocratic aircraft companies of the early 1930s workers lacked the means to mount a collective defense.

The Marginality of the Unions

The pressure of paternalism, the vigor of managerial discipline, the large proportion of engineers and white-collar employees in the plants, and the newness of the industry made the challenge of building trade

unions formidable indeed. Economic conditions during the early 1930s only compounded the difficulty. The contraction of the industry from 1931 to 1934 weakened workers' bargaining position and made the usual threat of dismissal all the more real. At the same time the hunger of the Air Ministry for new prototypes kept a high proportion of the work force involved in the prototype shops rather than in the production shops, where resentments against management, and hence the appeal of a union, were likely to have been greater. The bitter acrimony in the labor movement between Communist and non-Communist factions made organizing in the plants tougher still.

Under these conditions few workers were willing to risk dismissal and the blacklist, the price commonly paid for a life as a trade union activist. Although police records are spotty, it is safe to say that in aircraft factories, which in most cases employed from 150 to 1500 workers, there were rarely more than a half-dozen employees during the early 1930s who devoted themselves to the clandestine life of Communist cell meetings, tract writing, and the virtually futile effort to recruit comrades into the Confédération Générale du Travail Unitaire (CGTU), the Communist-backed labor confederation.^[37] Socialist militants were even rarer, except in Nantes and Saint-Nazaire, traditional centers of strength for the Socialist Party, the Section Française de l'Internationale Ouvrière (SFIO). The archives have left scarcely a trace of any anarchists or revolutionary syndicalists; although a few of these holdouts against wartime reformism and postwar communism may well have worked here and there in the aircraft industry, most of them probably lost their jobs after the war. No doubt many aircraft workers felt some loose connection to a tradition of working-class militancy—through memories or folklore about the old prewar CGT or the big metalworking strikes of 1919; and surely many workers voted for Socialist or Communist candidates in the national elections of 1928 and 1932. But during these in-between years before the Popular Front, as historian Yves Lequin has described them, only the rare militant dared to jeopardize a job in the aircraft industry for an overtly political life.^[38]

Few as they were, militants in aviation nonetheless managed to prepare the ground for the big unions that would eventually mushroom in aviation after 1935. Most activists were either unitaires , that is, affiliated with the Communist-backed CGTU, or confédérés , affiliated with the non-Communist CGT. The latter adhered to a more moderate, nonrevolutionary vision of the labor movement. For them, Léon Jouhaux's famous Minimum Program of 1918 still served as a doctrinal beacon. His program called for collective bargaining, the eight-hour day, labor representation in national economic planning and in factory management, and the nationalization of key industries.^[39] Though most confédérés voted Socialist, they clung to the tenets of the old CGT charter of Amiens of

1906, which affirmed the absolute independence of the labor movement from political parties. In the world of aviation, confédérés did best in two settings—in the shipbuilding cities of Nantes and Saint-Nazaire, where confédéré metalworking unions already predominated, and in the airlines and airports. Even in the inauspicious climate of the early 1930s a number of airline personnel and civilian employees in military air installations joined the CGT.^[40] But confédéré militants found it much more difficult to woo blue-collar workers in the aircraft industry. In this respect confédéré experience in aviation reflected the national pattern: the confédéré wing of the labor movement had its strongest support in white-collar France, especially among civil servants. For the most part it lacked firm roots in the private, industrial sector.

As a rule, unitaire militants had a better foothold, such as it was, in airplane factories. Like the confédérés, the unitaires did best in areas where their metalworking union was already well established. For the CGTU this meant Paris. By far the most important center for aviation, Paris offered that rare aircraft worker who was also a unitaire militant a viable radical culture within which to fashion a life as a union militant. A vast expanse of working-class neighborhoods in the east end of the city and in the red-belt suburbs gave militants access to organizations, newspapers, leaders, and friendships that served to sustain one's faith in a working-class movement.

Consider, for example, Henri Jourdain's recruitment into the CGTU in the early 1930s. Jourdain, who was the son of poor agricultural workers in the Centre and was to become a prominent leader in the PCF after the Second World War, began his career as a trade union militant while working in 1932 as a metal fitter in Wybot's airplane prototype shop at the Porte de Saint-Cloud, on the western edge of Paris. One of his fellow workers, a young man named Louet, was a member of Communist Youth. He "oriented me," Jourdain later wrote, "toward the revolutionary movement. I admired and envied the extent of his knowledge and his culture. A very convincing man, it was he who recruited me into in the CGTU." Once signed on, Jourdain discovered a whole new world of serious, committed people. At his first union meeting, "in a small bistro in Boulogne," he met Alfred Costes, the head of the CGTU's Parisian Metalworking Federation, "another of these métallo autodidacts with extensive knowledge." Self-taught himself—he had already developed a passion for cycling, theater, and the cinema—Jourdain dropped his night classes in industrial design at Arts et Métiers to enroll in the PCF's new université ouvrière , where he studied political economy, history, grammar, and spelling. By 1933 he had become one of three active unitaires in his factory, and though he did not join the PCF until the fall of

1936, he quickly became ensconced in the associational life of the Parisian Communist movement.^[41]

Although not everyone in the CGTU had Jourdain's intellectual appetite, he was a typical aircraft unitaire in being young, skilled, and eager to become a part of something larger. No CGTU membership lists or even membership figures are available; but from the few brief glimpses of militants that appear in police reports, it seems safe to surmise that most of the aircraft unitaires before 1935 fit this description. The four major unitaires in Dewoitine's factory in Toulouse, for example, certainly did: one was twenty-two years old and worked as a technician and inspector; another, thirty-nine, was an inspector of finished pieces; a third, twenty-seven, was chef d'équipe in the wing-building shops; and the fourth, thirty-seven, was a skilled fitter. They all had been born in towns not far from Toulouse or Perpignan.^[42] The preponderance of young, skilled, and native workers in the thin ranks of the unitaire unions paralleled the emergence of a new generation of leaders at the national level in the PCF—the rise of young men like Maurice Thorez, Jacques Duclos, and Benoît Frachon. In the aircraft industry, as elsewhere, what the CGTU local lacked in size it partially made up for through the youth, daring, and dedication of its members.

Part of what sustained the CGTU, of course, was its backing from the Communist Party. Although many militants, like the young Henri Jourdain, were not party members (at least not yet), enough of them were to ensure PCF control. Since 1924, and not without prodding from the Comintern, the PCF had staked its future on the factory cell as its principal form of grass-roots organization. Typically, a factory cell and a factory local of the CGTU came to have overlapping memberships. By 1926 police informants had discovered active cells in a number of aircraft factories in Paris—at Potez, Farman, Hispano-Suiza, Blériot, Renault, Gnôme-et-Rhône, and Bréguet.^[43] In all these cells militants tired to master the usual repertoire of political skills—recruiting members, attending meetings, organizing protests, and issuing tracts, leaflets, and newspapers like Les Ailes rouges or L'Aviateur, which wove together the party line and local grievances. Militants met with one another almost daily, usually in the haven of a reliable café. Secrecy was essential, ideological conformity the rule.

In the provinces the lives of unitaire militants fit the same basic pattern, though the strength of the movement varied from place to place. In Méaulte, Potez's company town, a unitaire militant could scarcely survive. But in cities like Toulouse and Bordeaux the PCF and the CGTU metalworkers union provided enough of a base to keep a tiny cell alive. In Bourges, for example, aircraft militants derived support from a rela-

tively stable union that militants had established during the war at the government arsenal. This base, coupled with an unusually strong cluster of woodcutter's locals, made it possible for militants at the Hanriot airplane factory at the airport in Bourges to persevere in what otherwise would have been a more politically isolated setting.^[44]

Strict discipline, however, was as much the bane as the boon of the CGTU. It gave militants what every trade union needed—leaders, newspapers, instructional pamphlets, modest funds, and, not least, comradeship and a network of reliable contacts crucial for clandestine work. Doctrinal orthodoxy had its virtues as well. It provided militants with a ready-made language for propaganda, a revolutionary vision, and a sense of common purpose—all of which helped them to endure a period of retrenchment in the labor movement.

But ideological conformity so corseted militants that it hampered their ability to work effectively as trade union organizers. By adhering to the Communist strategy of class against class, militants had to propagate ideas that were often at odds with the short-term interests of the aircraft rank and file. The class-against-class line of the late 1920s and early 1930s derived from the Soviet view that capitalism had entered into a crisis, or "third phase," which was intensifying class conflict within nations and accentuating the threat of an "imperialist war" against the Soviet Union. As a result, the Comintern called on militants throughout the Communist movement to agitate against war and repudiate Socialists and moderate trade unionists as "social fascists." Militants were encouraged to use tough tactics—insurrectionary protests, political strikes, and violent demonstrations—to destabilize the European democracies and wean workers from the moderate left. The strategy was eventually to have catastrophic consequences in Germany; in France it simply isolated militants, provoked repression, and kept the PCF and the CGTU in steady decline.^[45]

The crippling effects of the class-against-class strategy were especially transparent in aviation. As in every other branch of French industry, militants in aviation went out of their way to distinguish themselves from reformists by stressing their own uncompromising opposition to militarism, employer paternalism, the rationalization of production, and the pernicious role of the capitalist state. In cell tracts and factory leaflets aircraft militants constantly reiterated these themes. At Gnôme-et-Rhône for example, militants passed out leaflets claiming that "Gnôme-et-Rhône and the government that is at its service are one and the same. And this is why the economic struggle only worsens, marching ineluctably toward war."^[46] The CGTU's metalworking newspaper, Le Métallurgiste, hammered home a similar message under the headline "How 'They' Are Preparing for War!":

[Bréguet at Villacoublay] is a company that also takes a very active part in the establishment of the "circle of iron and fire" that the world imperialists are forging around the USSR. [Bréguet does so] through its technicians who, being highly experienced, train instructors for foreign aeronautical services, and through the manufacturing of special tools that aircraft construction requires and which can only be made by very experienced workers. Poland, Romania, Serbia, and all the states bordering on Russia have recourse to the Bréguet company.^[47]

Antimilitarist rhetoric also provided Communists with a vocabulary for attacking employer paternalism. L'Humanité described Potez's factory at Méaulte as "a military regime in every sense of the term."^[48] Even the sports programs fit into the framework. "To make young workers accept all this forced exploitation," a Communist journalist wrote in L'Humanité, "the big industrialists in the Paris region try to enslave them in their factory sports clubs, in camping trips and vacation camps controlled by the Comité des Forges. It is in fact the militarization of the factories."^[49] At Blériot's plant outside Paris unitaire militants made a similar attack on the company soccer league, which "the bosses use . . . to cultivate in young workers an ideology conforming to their interests." These militants urged workers to join a CGTU league instead.^[50] Like employers, Communists also viewed recreation as morally uplifting—precisely the reason they hoped to get soccer out from under the bosses' control.

The same logic compelled unitaire militants to reject almost every manner of cooperating with the state. In contrast to the confédérés, whose leaders participated enthusiastically in the Conseil National Economique, the League of Nations, and the International Labor Office, Communist unitaires condemned all cooperation as "class collaboration." The CGTU enforced this principle all the way down to the shop floor. In 1929, for example, a painter in an aircraft factory in Bordeaux earned what must have felt like a scolding when he innocently suggested that the union ask a labor Ministry inspector to investigate unhealthy working conditions in the plant. The painter hoped that as a result of union efforts, the ministry might force the company to establish a special drying room for painting as well as give each painter two liters of milk daily. The local unitaire leader in the metalworkers union, a militant named Casta, scoffed at the notion. Labor inspectors, he said, "would scarcely heed such a request," and he went on to remind the painter that "it is only the workers themselves who ought to take the necessary action by expressing their grievances to the boss, via the union."^[51] Casta, like most unitaires, felt that the union must insulate workers from the reformist mentality an appeal to state officials would have nurtured.

This is not to say that unitaire militants were anarchists, rejecting the legitimacy of any state. On the contrary, apart from a few revolutionary

syndicalists who had survived the "Bolshevization" of the CGTU in the 1920s, most CGTU militants were either Communists or sympathizers who admired the Soviet Union as a workers' state. Their appreciation of the potentialities of state power made them all the more hostile to what they viewed as the inevitable collusion that prevailed between industrialists and state officials under capitalism.

In the aircraft industry, however, a radical rejection of the capitalist state made it difficult for militants to attend to the immediate grievances of their fellow workers. State officials, after all, had a good deal of say over basic conditions in the industry. The Labor Ministry, with its small army of labor inspectors scattered throughout the country, had the power to enforce legal regulations at the workplace, modest though they were. What is more, the Finance Ministry and the Air Ministry had considerable leverage with the builders. During the left-wing interlude of the Cartel des Gauches in 1924, the government began to require that aircraft contracts include labor clauses stipulating limits on the use of foreign labor and procedures for raising wages during the period of the contract. Labor clauses were expanded in 1930 to provide further protections for workers. Even Albert Caquot, whom the builders considered well disposed to their interests in these matters, felt obliged as industrial and technical director to enforce labor clauses.^[52] Since the Labor, Air, and Finance ministries were all becoming increasingly involved in labor matters, militants could ill afford to ignore the implications state intervention had for their strategies; and as the ministries became more involved in air policy, Communist hostility to the state became a heavier burden for local militants to bear.

Likewise, staunch antimilitarism could be a liability for a militant in aviation. At one level, of course, antimilitarism spoke to a pacifist sentiment widely shared in interwar France. At the same time, however, opposition to military production ultimately conflicted with the obvious stake aircraft workers had in the military procurement orders of their factory. Aircraft workers in the early 1930s may have harbored some misgivings about building weapons of war, but by contemporary standards it was decent work, good pay, and a way to practice a craft. A prototype shop in depression France was a far cry from a munitions arsenal during the First World War, where war-weary assembly-line operatives did indeed respond to many a pacifist appeal. Not surprisingly, then, the antimilitarist campaign of the CGTU failed to win much support in the aircraft industry, especially during the depression, when commercial aviation could hardly be viewed as a plausible substitute for the military market.

The CGTU campaign against rationalization, moreover, failed to speak directly to the problems aircraft workers faced in their industry. During the late 1920s and early 1930s the CGTU stepped up its efforts

to recruit semiskilled workers in the big metalworking plants of automaking, munitions, and machine construction. Although Communist militants were no more opposed to rational production than they were to the state per se—Lenin, after all, had hailed Taylorism as an asset to a modernizing Soviet economy—they opposed capitalist rationalization and stepped up their attack against the kind of work organization that prevailed in factories like Citroën and Renault.^[53] In aircraft manufacturing, however, workers were plagued by different issues—job insecurity, the volatility of the airplane market, and the instability of their firms. To solve these problems, Communist leaders would have had to approve defense spending, which on antimilitarist grounds they opposed. As a result, the PCF and the CGTU had little to say about the real question of rationalization in the aircraft industry—the need to streamline the industry and modernize its production.

CGTU doctrine, by reflecting the class-against-class strategy and the priorities of militants in other metalworking sectors, prevented aircraft unitaires from functioning effectively as trade union organizers. Hostility to reformist tactics, while preserving the revolutionary integrity of the CGTU, rendered militants useless as day-to-day advocates on the shop floor. Isolation, moreover, made militants easy targets for repressive employers. Henri Jourdain survived as a unitaire leader at the Wybot airplane factory just long enough to organize a single protest, a one-day symbolic strike on May Day, 1933, in which only five or six workers took part. Afterward, he recalled, a friend "told me that I was going to be fired, and in order to have some chance of being hired elsewhere, he advised me to quit first instead."^[54] Jourdain was lucky: he escaped into a job in another metalworking firm just in time. Like many of his unitaire comrades, Jourdain remained a militant on the move, struggling to give life to a sterile strategy and keep one step ahead of the blacklist. It was a testimony to human tenacity that the CGTU survived at all.

Toward a Popular Front, 1934–1936

After 1933 a change in the political climate enabled militants to begin making headway in the industry. The new atmosphere had its origins across the Rhine. Hitler's rise to power in 1933, by provoking the French government to rearm and left-wing parties to establish a united front against fascism, had a larger effect on the lives of aircraft workers than anyone could have foreseen at the time. His thinly veiled efforts to rebuild the Luftwaffe forced the French Air Ministry to come up with a rearmament plan of its own—Victor Denain's Plan I for building 1,010 combat aircraft. Suddenly, in the depths of the depression, aircraft companies began to expand. In 1934 total employment in the industry jumped by 23 percent, and by 1935 the work force had grown to thirty-

two thousand employees, double what it had been in 1933.^[55] Since companies expanded without modernizing their methods of production, skilled metalworkers became more and more in demand. For the first time since the war aircraft workers watched the terms of the labor market shift decisively in their favor.

Because Plan I called on the industry to build large numbers of all-metal planes, the work force not only expanded in 1934 and 1935; it became more homogeneous as well. The vast majority of aircraft workers by 1935 labored as fitters, sheet metal workers, welders, lathe operators, drill press operators, and machinists, the crafts that formed the backbone of the metalworking federations of the CGTU and the CGT. As all-metal planes displaced the wood and canvas planes of the 1920s, the proportion of carpenters in the work force declined, as did that of the women, who had often been hired to sew canvas. The shop floor became a more exclusively male and metalworking domain.^[56] Plan I, moreover, brought thousands of workers into production sheds rather than into the more privileged, less alienating world of the prototype shop. By tightening the labor market and expanding the most industrialized part of the airplane factory, Plan I created a more favorable environment for building a labor movement within the industry.

Just as important, however, was the shift to a united-front strategy in the Communist Party. In the spring of 1934 two developments converged to encourage the PCF to abandon the bankrupt strategy of class against class. First, within France, the right-wing riot of 6 February 1934 inspired grass-roots militants in the CGT, the CGTU, and the Socialist and Communist parties to collaborate in organizing counterdemonstrations against what they believed was a fascist threat from the right. After staging successful mass marches throughout France on 12 February, local militants set up joint vigilance committees to continue their combat against the right-wing leagues. This ecumenicalism at the local level gave those national Communist leaders who had become disillusioned with the class-against-class strategy a chance to argue for a united-front strategy. Meanwhile, in Moscow, Soviet leaders and representatives from various national Communist parties began to explore ways to ally with Socialists and trade union moderates throughout Europe in a common crusade against fascism. Faced with the prospects of fascist enemies in Japan, Austria, Italy, and above all in Germany, Stalin began to entertain a foreign policy of allying with the Western democracies. By May 1934 a sea change in Communist strategy was well under way. Maurice Thorez, the head of the PCF, soon made overtures to the Socialists, and some months later to the Radicals, in hopes of forging a popular front. By 14 July 1935 the efforts of all three parties would redound in an enormous street demonstration in Paris when, as Thorez described it, "half a mil-

lion men and women marched from the place de la Bastille to the place de la Nation through the old Faubourg Saint-Antoine, rich in revolutionary memories."^[57] Within eighteen months of the bloody riot of 6 February 1934 antifascism had brought the French left back to life.^[58]

The shift to a united-front strategy in 1934 quickly filtered down to the factory level. In Toulouse, for example, confédérés and unitaires at Dewoitine and Latécoère began sending representatives to each other's meetings in the summer of 1934, just when national federation leaders began negotiations in Paris.^[59] That summer too L'Union des métaux , a confédéré metalworking newspaper, began publishing a few articles presenting unitaire positions on key issues, a practice that gradually gave the paper an overtly bipartisan cast.^[60] Behind the facade of cooperation, of course, rivalries remained fierce. But at least militants in both camps could channel the energy they had previously devoted to attacking each other into a more productive struggle to combat employers, rebuild unions, and reunify the labor movement.

More significant still was what the shift to a united-front strategy did to the tactical mentality of unitaire militants in aircraft factories. The change from "class against class" to "the struggle against fascism" freed local militants to march in closer step with a rank-and-file work force whose day-to-day interests were tied to production for national defense. Although it took time to wean the party of its doctrinaire antimilitarism, by 1935 the PCF had come around—as had Stalin even sooner—to supporting the principle of a strong French defense.^[61] Aircraft unitaires now found it easier to embrace the bread-and-butter interests of their fellow workers. What is more, the CGTU dropped its taboo against lobbying government officials. At a meeting of the Paris machinists union on 31 October 1934, Alfred Costes of the CGTU called for joint CGT-CGTU delegations to pressure the Labor Ministry to intervene more aggressively in behalf of workers. This openness to state intervention marked so dramatic a change of heart in the Communist camp that a confédéré could not keep himself from saying to Costes, "But not long ago the unitaires reproached us for always ringing doorbells at the ministries."^[62] This stunning reversal of tactics in the CGTU—an acceptance of military production and a willingness to bargain with the state—removed the major doctrinal shackles that had hampered unitaire militants for years.

The first signs of a change in the way the CGTU approached its work in the aircraft industry could be seen as early as May 1934. At a two-day conference in Paris national officials of the CGTU's Metalworking Federation assembled aircraft militants from a number of companies to establish a national campaign to address workers' grievances in aviation. Wages and hours dominated the discussion: delegates attacked the

Rowan system as a disguise for wage cutting, and they condemned "the collusion between the government . . . and factory directors in various plants" in lowering wages and demanding overtime.^[63] To plead their case publicly, delegates devised a plan for presenting five demands at the upcoming international air show at Bourget: a guaranteed minimum wage of seven francs an hour for skilled workers; wage adjustments for women, youth, and the unskilled; an end to overtime; the creation of factory delegates for health and safety; and paid vacations to replace the unpaid annual plant closings that were common in the industry. Delegates concluded the conference by taking steps toward founding a section industrielle de l'aviation within the Metalworking Federation. With the expansion of the industry at last under way, aircraft militants were now beginning to win recognition within the federation. And with the CGTU shifting toward a more pragmatic style of working-class advocacy, militants focused their attention on the bread-and-butter issues that troubled their comrades on the shop floor.

Meanwhile signs of a revival of working-class militance appeared at the factory level. On 12 February 1934 automobile and aviation workers at Renault's plant in Billancourt staged their largest protest since 1926. The trouble started when they overturned the snack wagons that managers had brought into the shops to reduce tensions that had been building since the right-wing riots of 6 February. After the initial mayhem workers settled into a grève tournante in which shops struck briefly, one after the other, a tactic designed to avoid lockouts and reprisals.^[64] A few days later, normalcy returned to Renault, but within a week trouble erupted nearby at the airplane factory of Lioré et Olivier, which had just received a large order for seaplanes from the Air Ministry. When managers began hiring more workers, however, technicians and workers already on board demanded wage hikes. At first managers listened, but then they delayed. On 20 February eight hundred workers sat in their shops during lunch hour and then walked out well before closing time, leaving only fifty draftsmen behind. A lockout followed. Three weeks later, work resumed at the old wage rates.^[65] Strictly speaking, the protests at Renault and Lioré et Olivier came to naught. Still, workers had broken the collective inertia that had long prevailed in aviation. Events like these, moreover, were beginning to close the gap between militants and unorganized workers.

Naturally enough, the "red belt" of Paris served as the site for the initial renewal of labor protest. Yet in the provinces, too, workers began to rediscover a capacity for mutual defense. In Toulouse police informants reported in August 1934 that the CGTU had sixty-six members at Dewoitine and fifteen at Latécoère, whereas the CGT claimed about forty members in the two plants—no stampede, but real progress for the

unions nonetheless. When Dombray, the factory director at Latécoère, fired three militants for trying to build a Communist cell in the plant, workers responded in force. Representatives from both the CGTU and the CGT formed a delegation to demand their reinstatement and "to make Dombray understand," as one militant put it, "that when a worker enters his factory, he need not leave his politics in the cloakroom."^[66] Dombray refused to rehire the three Communists, even after the Conseil des Prud'hommes, the local labor tribunal, ruled in their favor. More than three hundred workers at Latécoère then petitioned management in the militants' behalf—to little avail. An autocratic employer like Pierre Latécoère could still have things his way, but not without provoking some collective resistance.^[67]

Two months later, in October 1934, Emile Dewoitine took it on himself to try to clip back this budding militancy in Toulouse. On 22 October he fired a leading militant in his factory for allegedly slowing down production. At closing time half of the five hundred workers at the factory gathered around the company gates to organize a response. Each shop designated a representative to be part of a delegation to confront the boss. But Dewoitine was already several steps ahead of his workers. The day after the firing, a police spy assigned to the plant wrote the local prefect that Dewoitine had been kept current about the activities of militants working in his plant and was "quite disposed to rid himself of them when the favorable circumstances presented themselves." A fortnight later, after consulting with no lesser authority than the air minister, General Denain, Dewoitine fired two more militants. Police informants correctly predicted that since only 150 to 250 workers could be counted on to follow the union, the militants would refrain from calling a strike. Again the boss prevailed. But the CGTU and the CGT kept up a tireless propaganda campaign to win the reinstatement of the workers.^[68] In Toulouse, as in Paris, reprisals against militants had the effect of purging plants of a few troublemakers at the cost of instilling a sense of injustice on the shop floor.

The revival of militancy continued to work its way through the industry in 1935—so much so, in fact, that by June 1935 employers assembled at the Chambre Syndicale to construct a collective defense of their own.^[69] An epidemic of wage demands had begun at Gnôme-et-Rhône, where Paul-Louis Weiller, the no-nonsense director of the company, had found it prudent to grant workers a 7 percent wage hike and a fourteen-franc bonus for night work—all in the hopes of heading off a strike. These concessions prompted foundry workers at a rival firm, Hispano-Suiza, to stage a walkout for a similar bargain. Trouble then surfaced at Blériot. Faced with the first rash of wage demands to appear in the industry since the mid-1920s, employers at the Chambre Syndicale swiftly

improvised a common strategy. Members pledged to resist the temptation stemming from Plan I to hire away skilled workers and technicians from one another. They then agreed to renew their efforts to standardize wages, vacation policies, and other personnel practices in hopes of undermining union efforts to play one firm off against another to win benefits. How faithfully the builders stuck to this bargain is hard to say; the Chambre Syndicale had no way to sanction its members, and in an industry like aircraft manufacturing there was always the hope of passing wage hikes on to the state. One thing is clear: for the first time workers had forced employers to close ranks behind a common labor policy—one more sign that Plan I and the united-front strategy were having a corrosive effect on the autocratic structure of authority in the industry.

Plan I impinged on employer autocracy in another way as well: it raised the government's stake in the productive capacity of the aircraft industry, and this change, too, had an impact on labor militancy. A highly publicized strike at the Société de Duralumin plants in Rive-de-Gier, near Lyon, brought this lesson home to militants and employers alike during the summer of 1935. Metalworkers at Duralumin walked out in late June when managers tried to quicken the pace of production by introducing the Bedaux wage incentive system. Workers, led mainly by CGTU militants, agreed to let the local justice of the peace mediate. Management, however, refused. Workers then appealed to the Air and Labor ministries since most of the duralumin, an aluminum alloy, produced at the plant went straight to the aircraft manufacturers. With Plan I under way, Air Minister Denain intervened, dispatching an official from the ministry to arbitrate the dispute in short order. Within hours after Denain's man arrived at Rive-de-Gier, management retreated. The company dismantled the Bedaux system and raised wages modestly. Although no friend of labor, Denain could ill afford strikes at a firm that produced nearly all the duralumin, and much of the aluminum, that the aircraft industry needed.^[70] Rearmament made the Air Ministry a powerful partisan of social peace in the industry, and everyone knew it.

The growing involvement of the state in the aircraft industry enabled militants and left-wing politicians to bring labor issues out of the confines of a company's personnel office into the forum of public debate. In December 1935, for example, Emile Dewoitine announced that he would soon be forced to lay off one hundred of his six hundred workers unless more orders from Plan I were forthcoming. By making his case public, Dewoitine unleashed a wave of protest directed toward Denain's Air Ministry. Business and labor leaders in Toulouse asked local deputies and senators to lobby the air minister for more orders. Even the prefect entered the act. "It should not escape your attention," he wrote Denain, "that in addition to the social repercussions that can result from throwing

many workers into unemployment, it would jeopardize the National Defense to let go specialized workers whose training has taken so long and who could not be replaced without a long and costly apprenticeship."^[71] Meanwhile a Socialist member of the Chamber of Deputies, Emile Berlia, took the occasion to lobby for labor reforms. It is not enough, he told the Air Ministry, just to save jobs at Dewoitine when political repression and working conditions within the plant were severe. The Air Ministry, Berlia felt, ought to give workers a guarantee that their working conditions would improve.^[72] Faced with an avalanche of appeals, Denain instructed Fernand Lioré, the industrialist, to redirect recent orders to his groupement toward Dewoitine. Once again a builder got what he wanted—more orders—but in the process the politics of the industry were becoming a triangular affair involving employers, state officials, and labor.

By the spring of 1936 the combined effects of company expansion, state intervention, and working-class militancy had broken open the closed doors that only three years before had allowed the builders to preside over their workers as omnipotent lords. Aircraft workers had by no means created a powerful labor movement; nor could anyone have had a premonition of the resurgence of militancy to come. Still, in the two years following 6 February, workers had forced employers to contend seriously with labor protest for the first time since the world war. Even if most aircraft workers had not joined unions, they were now consistently defending the right of militants to do so. At Saint-Nazaire 612 of 717 workers at the airplane factory of Loire-Nieuport struck for two days to win back jobs for fired militants.^[73] Trade union activists were no longer isolated mavericks. At Bloch's factory near Paris about 20 percent of the workers had joined the CGTU or the CGT by 1935. By early 1936 Marcel Bloch, as tough a boss as any, found himself locked in negotiations with the leading Communist militant in his plant over wages, hiring policy, union rights, and paid vacations.^[74] Amid all this activity, moreover, militants continued to promote the notion of a popular front. In March 1936 the "unity of the working class against fascism" became something much more palpable than a slogan when unitaire and confédéré leaders formally consolidated their confederations into a new, unified CGT. Reunification made it easier for a worker to join a union local and feel part of a larger movement. By the spring of 1936, after fifteen years in the wilderness, trade unionism had finally returned to the factory sheds and prototype shops of the aircraft industry.

As workers recovered their capacity to act collectively, they added a new dimension to the crisis into which the industry had fallen by 1936. Plan I had exposed how poorly equipped this ramshackle collection of companies still was to carry out a rearmament drive. Labor militance, limited though it still was in 1935, posed new questions for employers

and officials in the Air Ministry about how much they could demand of workers without undermining social peace in the industry. If in falling behind production schedules employers were increasingly tempted to resort to speedups, wage incentives, overtime, and night work to meet deadlines, they now ran greater risks than before of provoking employee resistance—as occurred, for example, at Courbevoie, where Bréguet's workers tried to voice their objections to la minute économisée .^[75]

The revival of labor militance, moreover, increased the political risks of laying off and rehiring workers with the alacrity that employers had been used to in the early 1930s. Because the Air Ministry had not yet determined what kind of production program would follow Plan I, something odd began to happen in late 1935. Even though the aircraft industry as a whole continued to grow, some companies, like Dewoitine in Toulouse and Loire-Nieuport in Saint-Nazaire, began to run short of orders.^[76] A badly structured industry and a poorly organized rearmament effort made layoffs, paradoxically, a continuing danger in an expanding industry. By 1936, however, workers were much less apt to stand by idly while a foreman, a builder, or a bureaucrat in the Air Ministry determined the fate of their jobs. More and more, militants and their budding unions provided a way for workers to shout back about the issues that really irked them—job insecurity, the boss's arbitrary power to hire and fire, reprisals against militants—forms of injustice that had too long prevailed in an autocratic industry. The emergence of factory unions gave workers a way to speak up about conditions just at a time when employers were losing their credibility as an elite fit to run the industry on its own.

Militants undoubtedly profited from the employers' failure to consolidate the industry into a small number of modernized firms. Years of resisting reform had exposed builders to public criticism, eroding their authority all the more. No one had taken over their firms—at least not yet—but between 1928 and 1936 ministers, state engineers, labor inspectors, prefects, and even local politicians had all come to encroach on the builder's right to command his firm as he pleased. And just when employers were losing prestige, popular frontism gave militants both the tactical means and a compelling social vision with which to compete with the employer's authority. Left-wing spokesmen for the Popular Front, whether Communist or Socialist, ex-unitaire or ex-confédéré, offered workers an alternative orientation rooted in the language of French left-wing tradition. As Maurice Thorez said in a speech at the Communist International, "We, the great-grandsons of the sansculottes of 1792, the soldiers of Valmy, deny . . . Colonel de la Rocque [the right-wing leaguer], whose great-grandfather was in the army of Condé and of the king of Prussia, the right to speak in the name of our country."^[77] By

blending patriotism, antifascism, and left-wing radicalism, the Popular Front movement opened up a new way of thinking about what it meant to build airplanes for France. It offered workers a different frame of moral reference to compete with that of the boss. In 1935, when the Popular Front coalition of Communists, Socialists, and Radicals called for the nationalization of war industries, the left-wing challenge to the builders became explicit. For the politically aware aircraft worker, it now became clear that the Popular Front movement might have radical implications for life in an airplane factory. Thus, the revival of labor militance after 1933 meant more than what a few factory skirmishes might otherwise have suggested. Because it coincided, and indeed was itself bound up with, rearmament, a production crisis, and the movement for a popular front, working-class militance in the aircraft industry posed a threat to the very foundation of the builder's authority. No one could have recognized this challenge in the spring of 1936. By summer it would be apparent to all.