A few French army officers had long recognized the possibilities in grouping tanks into division-sized units. On 15 February 1920, for example, General Estienne, at the Conservatoire National des Arts et Métiers, discussed the strategic advantages of a large, independent armored force. In July 1927, before the Center of Higher Military Studies, Col. André Doumenc described the tactical advantages of a large armored force. In 1928, Doumenc submitted to the General Staff a project for the organization of an armored division, similar to that which would later be realized by the Germans. While he was a member in 1930 of the tank technical section of the Department of Infantry, Lt. Col. Marie J.P. Keller wrote a special study on the need for a mechanized division that could “rupture” enemy defenses. When Lt. Col. Charles de Gaulle published his book in 1934 on a professional armored corps, his special army included three thousand tanks and six armored divisions. The official effort to create division-sized tank units, however, did not actually begin until the late 1930s, and the French did not form their first armored divisions until January 1940. Although the first light mechanized division was formed officially in 1935, its function, as will be discussed below, clearly differed from that of the later armored divisions.
Throughout the debates and discussions on forming large tank units, the military hierarchy believed it could not form the first armored divisions until sufficient medium tanks were available to equip an entire division. Since France had only a small number of medium tanks for almost the entire interwar period, the High Command saw no compelling reason to create large armored units, even though production of the B-model tank remained painfully slow. By January 1936, the army had only seven B-1 tanks, and this number slowly increased to seventeen in March 1936, thirty in February 1937, and thirty-five in September 1937. Following the decision to produce an improved version of the B-1, production of the B-1 tank ceased and that of the B-1 bis began. In May 1937, the French army had four B-1 bis tanks, and this slowly increased to seventeen in September 1937, thirty-five in January 1938, forty-seven in July 1938, and seventy-one in January 1939. The first model of the B-1 ter, another improved version of the B-1, arrived in April 1938, but only a few models were ever manufactured. By the beginning of 1939, there were only 107 B-model and fifty D-2 tanks. In comparison, France had 790 R-35, one hundred H-35, and eighty-nine F.C.M.-36 tanks. Beyond a doubt, the High Command could have formed—if it had wanted—armored divisions with the infantry tanks. The army’s leaders, however, were reluctant to strip tanks from the infantry, whose needs remained most important in their eyes, to form an unproved and untested armored division.
Despite the sincere desire of the tank enthusiasts for a larger number of the B-model tank, its extreme complexity made it unsuited for mass production and resulted in its being produced very slowly. During the period between March 1936 and March 1937, for example, the French produced 13 B-1 medium tanks and 422 R-35 light tanks. Among other problems, the steering system for the B-1 was very exacting and delicate. Since the 75mm cannon was mounted in the hull, it could only be traversed by turning the entire tank. This required a very sophisticated steering mechanism. Also, the demand for better protection against antitank weapons led to a decision to increase the thickness of the armor plating in the B-1. In contrast to the 1921 plan for the medium tank to weigh a maximum of thirteen tons and the armor plating to be 25mm thick, the B-1 bis had armor plating of 60mm and weighed thirty-three tons. When the Consultative Council on Armaments agreed in January 1935 to increase the armor plating of the new model of the B-1 to 60mm, it added the stipulation that the increase in armor plating should not decrease the mobility of the tank. In fact, weight increases reduced the cruising radius from more than nine hours, to about five hours and fifteen minutes. The decrease in cruising range and agility was the practical result of the proponents of protection winning the debate against those of mobility. Throughout its development, constant modifications and slow production plagued attempts to field more B-model tanks.
The French High Command recognized the production difficulties with the B-model tank. In October 1930, the inspector general of tanks conducted studies on establishing specifications of a new tank that would use much of the technical experience from the B-1. If produced, this tank would have been designated the B-2. In 1932, two other prototypes, designated the B-3 and the B-B, were discussed. Even though the minister of war ordered prototypes of the B-2, B-3, and B-B tanks, none were ever produced. In late 1935 or early 1936, the inspector general of tanks began a study on a new thirty-five-ton tank, whose specifications resembled those of the B-2. In the same period, the Department of Infantry ordered a prototype of a new twenty-ton tank, which in no case would go beyond a total weight of thirty-five tons. The same request proposed that the new tank be called the G-1. The suggestion for a massive forty-five-ton tank also reappeared during this period, but was squashed when the director of infantry branded it as being “without real utility.”
Unfortunately for France, the army had begun in the early 1930s to reconsider the B-1 tank during the period when discussions at the conference on disarmament at Geneva indicated a possible limitation of twenty-five tons on the weight of tanks. This delayed the anticipated studies, since the improved B-model tanks might have weighed as much as thirty-five tons. Then, reductions in the 1934 budget slowed the development of a new medium tank. Finally, the massive resurrection of the light-tank program undoubtedly came at the expense of the medium tank. While the precise effect of these factors cannot be determined, the High Command’s reassessment of the tank program from 1932 to 1934 and its redirection thereafter adversely affected the long-term development of the French tank and undoubtedly had a greater effect than the disarmament talks or the budgetary restrictions. Instead of producing a better medium tank, it produced hundreds of infantry support tanks. Had the High Commission aggressively pursued the development of a new tank in the early to mid-1930s, a better tank than the B-1 might have been produced, but by the middle of 1938, it was too late to begin development and production of a new medium tank. The crises of 1938 underlined the need to have large amounts of modern equipment as soon as possible, and France’s previous experience indicated that the development process might take several years. Her insistence on having sufficient medium tanks before she formed her first armored division, however, gave her little choice but to await production of sufficient B-model tanks. She refused to untie herself from the anchor of the “perfect” weapon.
The center of the discussion on forming and organizing large armored forces was the Superior Council of War, which was very deliberate in its consideration of armored divisions. General Gamelin stated in an April 1936 meeting, “The problem of constituting…large [tank] units has been studied in France since 1932; the development of the antitank weapon has caused the renouncing of this conception.” While the 1932 field tests had seriously delayed the creation of large tank units, subsequent tests had also not furnished evidence to support the creation of such units. Gamelin repeated other important findings from those tests when he explained that a tank attack could succeed against a soundly constructed defensive system only if it were supported by a “strong” artillery, which would be used against enemy antitank weapons. He added that the German armored divisions seemed incapable of completing a rupture of strong defenses, and seemed most appropriate to attack weakly held defenses or to conduct an exploitation. Throughout the discussion, members of the council seemed to view large tank units as large groups of mass maneuver tanks, which performed essentially the same function as the mass maneuver tanks but at a higher organizational level. While the council did recommend in April 1936 forming France’s second light mechanized division, which was a mechanized cavalry division, none of its members argued for a more energetic program of testing or forming an armored division. As far as they were concerned, the ad hoc combination of battalions of mass-maneuver tanks was sufficient.
In late 1936, France slowly began to accelerate the creation of large armored units as she began to improve and modernize her force. The government had already reestablished two years’ service in the army in March 1935, and had also slightly increased defense expenditures in 1935, but the real rearmament of France began in late 1936, following the remilitarization of the Rhineland in March, the outbreak of the Spanish Civil War in July, and the German extension in August of military service to two years. With an increasingly threatening international environment, the government increased the defense budget from 11.48 billion francs in 1934 to 12.657 in 1935, 14.848 in 1936, 21.235 in 1937, 28.976 in 1938, and 93.687 in 1939. Although these increases lagged behind those of Germany, which was spending more than double these amounts in terms of percentage of the gross national product, France began what she believed to be a massive rearmament and modernization program. After all, Germany had to reconstitute her equipment almost completely, while France had a large quantity of equipment on hand in which massive sums had already been invested. As part of the armament program of September 1936, France proposed to create, among other things, fifty battalions of light tanks and twelve battalions of “heavy” tanks, evidently B-model tanks. The latter twelve battalions would equip two armored divisions of six battalions each.
The High Command also began more energetically in 1936 to investigate the possibility of large armored formations. In October, Gamelin “invited” the members of the council to study the question of forming an armored division. In November 1936, Daladier, the minister of national defense, ordered the study of the possible future employment of armored divisions. But the High Command made no move to create such a division immediately. At the Riom trial, Daladier insisted that the military hierarchy could have formed an armored division in 1936 if “they” had wanted. His comment illustrated his belief that such decisions were clearly in the realm of a technical, military question and beyond the proper authority of the civilian minister of war. The military leaders, however, would not rush into an unproved concept, and they would not create an armored division until there were sufficient B-model tanks in France’s inventory.
On 15 December 1937, the Superior Council discussed in detail the organization of an armored division. Gamelin began the discussion by explaining that the type B and D tanks were tanks for a mass maneuver and referred to these tanks as “heavy” tanks. While he accepted the need to organize armored brigades in peacetime, he noted the difficult question of whether it was necessary to organize the brigades into divisions. Gamelin argued that it was easier to split up a division into brigades than it was to form a division from brigades, a fact that supported the need to form an armored division. He concluded by noting that a large tank unit could be used in a “powerful” counterattack, but that it could also play an important role in the exploitation of a breakthrough, as well as in a flanking maneuver. General Dufieux offered a word of caution. He explained that France would not have enough B-1 tanks to form six tank battalions until the beginning of 1939. He believed it was dangerous to group all the heavy tanks in one division. Gamelin insisted, nonetheless, that the existing “heavy” tanks be organized into a brigade so that special studies on the employment of the armored division could be conducted. The final decision of the council was to constitute a special group under the inspector general of tanks to study the proper composition of the armored division.
In February 1938, the minister of national defense ordered the constitution of the special study group recommended by the Superior Council of War. General Martin, the inspector general of tanks, was placed in charge of the study. His selection as director of the study group ensured that the ideas he initially expounded after the 1933 Coëtquidan maneuvers and reiterated in his January 1936 presentation at the Center of Higher Military Studies would remain an integral part of French doctrine. Tragically, when France needed a man of great vision and imagination, she received instead an officer who was content to apply the ideas and methods of the past. The study included the creation of a large armored unit in 1938 for the first time in France. The unit was supposed to include one battalion of B-1 tanks, one battalion of B-1 bis tanks, one battalion of D-2 tanks, and two battalions of infantry. It would also include one battalion of 75mm and one battalion of 105mm artillery. Extensive field tests were scheduled for March 1938 and October 1938, but they were interrupted first by the annexation of Austria by Germany and then by the Munich crisis. Such tests could not be conducted in a time of international tension. Thus the major results from the special study remained almost purely theoretical. No armored division came from the study group’s effort.
General Martin cannot be accused of completely lacking any perception of the future possibilities of armored warfare, since his final report did contain a proposed organization for an armored division. He suggested it include two “demi-brigades,” which were units smaller than the normal brigade and which included no regiments. The division consisted of four tank and two infantry battalions, half of which would be assigned to each demi-brigade. Two 105mm artillery battalions, with the shorter-range howitzer, provided organic fire support. The special study group also wrote a draft edition of a manual on the employment of armored divisions, but the provisional manual followed the directions of Gamelin by limiting its focus to the “general case of operations conducted against an enemy imperfectly installed on terrain and disposing of reduced means.” The study group produced five copies of the draft notice in October 1938 and sent copies to Gamelin and Georges. The concepts in the draft were highly reminiscent of those employed with the tanks for mass maneuver. If the several crises in 1938 had not interrupted the planned tests, more progressive concepts might have emerged. With the unimaginative Martin as the study-group leader, however, such a development was not likely.
With the international situation becoming more threatening in 1938, the Superior Council of War discussed the capabilities of the tank more intensely in December 1938. After beginning the discussion by noting how the partial mobilization of September 1938 prevented annual maneuvers and field trials, Gamelin said that numerous studies had been made and the potential of the tank was clear. He then referred to armored divisions as “rare and precious” units, and stated that one armored division had to be created in 1939, with three ultimately being created. Gen. L. A. Colson noted, however, that due to matériel limitations, the first two armored divisions could not be formed before the beginning of 1941. In the midst of the discussions, Gamelin explained his perception of how the large armored units should be employed. He said their best use was in an action “on the decisive point in the battle,” and explained that an armored division could achieve a “result which could not be attained in the past.” While Gamelin obviously accepted the possibility of large tank units playing a crucial role in future battles, he nevertheless believed it was “impossible to constitute in actual fact in times of peace, this large unit with all the means for which it may have need in all the phases and forms of its action.” In peacetime, the armored divisions should have only “strictly indispensable elements,” and, according to Gamelin, any other solution was “premature.” Except for Gen. Pierre Héring, the other members of the council agreed. Héring insisted that the armored division should be organized in a fashion where it could operate in a more autonomous manner; this required additional artillery, communications, supply, air defense, and maintenance support.
While the differences in organization might seem nothing more than a debate over resources, they revolved around entirely different notions of how an armored division should be employed and of how a future war might be fought. From Héring’s viewpoint, greater combat and support capability could enable a tank division to operate in an independent and much more fruitful fashion. He insisted the armored division should not be “obliged to advance in successive jumps according to the classic process of slow attacks based on the movement of infantry and artillery.” From Gamelin and the other council members’ viewpoint, the armored divisions operated in the same manner as other tanks for a mass maneuver. While a division was much larger than the other mass-maneuver organizations, which were battalions or ad hoc groups of battalions, their concept envisaged it being employed in a step-by-step and carefully controlled fashion within the methodical battle. Though Gamelin proved himself a stronger proponent for armored forces than some of the other members of the council, he failed to recognize the capability of the armored division to restore maneuver to the battlefield. He still considered the tank units to be most valuable for their ability to add considerable firepower to an attacking force and for their potential for crushing any opponent in their path, and he believed a tank division provided more firepower and mass than any other organization.