The Madagascar Expedition of 1895 was the most disastrous colonial campaign of the Third Republic. Orchestrated by a number of interests, which included a clutch of deputies from the island of La Réunion intent upon annexation and Catholics who had never pardoned the Hova monarchy its 1868 conversion to Protestantism, the expedition got off on the wrong foot from the very beginning. This need not have been the case, for two months before October 1894, when the French took advantage of a rebellion against Hova rule over the island to present the Malagasy prime minister with a proposal to establish a French protectorate, an interministerial commission had been hard at work planning a military expedition in anticipation of a Hova refusal. So, when in November 1894 it was made official that Madagascar would be forced to accept a French protectorate at the point of a bayonet, military preparations were already well advanced.
Alas, this did not guarantee their efficiency. The source of the problems, the critics believed, stemmed from the decision to confide the expedition to the army rather than to the marines, which won the Madagascar contract simply because they had underbid the navy for the honor of attacking the “Red Isle” by thirty million francs.1 Although army units had been present during the conquest of Tonkin, and the Legion continued to serve there, army officers had had little experience in colonial campaigning since Mexico. The French foreign minister at the time of the decision to invade Madagascar, Gabriel Honotaux, later excused the mistakes of the war ministry, whose “organization is not made for that, which does not have the necessary contacts with the colonial world so that from one day to the next it can recruit from all over the globe the means which it needs.” Due to lack of experience, the army desperately underestimated the requirements of the Madagascar expedition, which helped to explain their low bid.
Nevertheless, it would be unfair to argue that the experience of other colonial campaigns was totally ignored by the committee that met in August 1894 to draw up a plan of campaign. Nor was their plan necessarily a bad one. In one month, representatives of the ministries of war, the navy, the colonies and the foreign office manufactured an invasion blueprint that, they believed, would require twelve thousand men to overcome a semi-organized Hova army of forty thousand. The objective of the operation was Antananarivo (Tananarive), the capital of the Hova people who dominated the northern half of the island, situated in Madagascar’s mountainous interior. Majunga on Madagascar’s west coast was chosen as the port of entry. Not only did it offer a large harbor, but also it stood at the mouth of the Betsiboka River, which, when combined with its tributary the Ikopa, provided a navigable route about 160 miles deep into the Hova heartland. Estimates put the number of porters and mule drivers required to support them for the remaining distance to Tananarive at eighteen thousand to twenty thousand. However, it was reckoned that this number could be reduced to five thousand by using the voiture Lefèbre—two-wheeled metal wagons invented in 1886, which came in kits and weighed about five hundred pounds when assembled. A final, and ultimately extremely controversial, decision taken by the committee was to depend essentially upon white troops, many of them from metropolitan units, to furnish two-thirds of the combatants, thereby reversing the proportions of white and native troops used in Dahomey. Worse, many of these white troops would come out of metropolitan units, which were neither experienced nor acclimatized to campaign conditions outside of Europe.
Knowledgeable marine officers criticized this plan almost immediately. Leaving aside for the moment the heavy dependence upon white troops, something that flew straight in the teeth of conventional colonial military wisdom, their objections were essentially two: The first was that Majunga was simply too far from Tananarive, and possibly on the wrong side of the island. The use of other ports on the east coast of Madagascar as a staging area, in particular the commercial port of Vatomandry through which most of the supplies including weapons arrived for Tananarive, would have cut the distance the invaders marched by two-thirds. A combination of Vatomandry and, say, Tamatave or Andevoranto might also have allowed the French to use separate columns to converge from different directions upon their objectives. The second objection was that the invasion force, which, excluding reinforcements sent later, eventually numbered almost fifteen thousand fighting men and seven thousand porters, was simply too large and too heavy. While the Hova army appeared formidable on paper, there were probably no more than ten thousand of them able to bear arms. A light force of five thousand fresh colonial troops, shorn of the impediments of heavy artillery and wagons included by the committee, and backed by eight thousand mules and porters, could make quick work of the Hovas.
As will be seen, many of these criticisms were proved correct by events. However, the committee invasion plan was not foreordained to near disaster. Many of the problems were not so much with the plan as with the plan’s execution. After all, Madagascar was a far larger country than Dahomey and its government was far more sophisticated than that of Behanzin, although the veneer of civilization reflected in the quasi-European aspect of its capital and the Christianity of its Hova inhabitants was a very thin one. Its army, even if not up to European standards, possessed some modern weapons and counted foreign mercenaries, particularly English officers, in its command. Furthermore, the mountainous interior offered many opportunities for even a half-trained force to make heroic stands in high passes and narrow defiles. It was perhaps acceptable for headstrong colonial officers to count upon a large degree of incompetence in their enemy when making campaign plans. But this was a more dangerous assumption to make from the vantage point of Paris, even if the Hovas proved feckless and incompetent beyond the wildest imaginations even of marine officers. The greatest mistake of the committee, however, was not in their choice of Majunga or in the numbers they dispatched to Madagascar. Rather, it lay with their failure to factor the problems of execution, what German military philosopher Carl von Clausewitz called “friction,” into their calculations.
From the sea, the Madagascan landfall at Majunga appeared inviting in April 1895, especially after days spent on transports that now smelled strongly of manure and human negligence, and where even the food had taken on the flavor of soot and coal dust. A spit of sand that separated the Indian Ocean from the Bay of Majunga, about five miles wide at its entrance, led from the water to the European quarter, a tiny grid of streets bearing names far grander than the modest architecture of the wooden houses shaded by large mango trees that bordered them appeared to merit. Behind this small concession to order, a modern chaos of native huts spread to the foot of a beautifully wooded hill, whose low summit was crowned by a small fort, crumbled and blackened by the naval bombardment it had suffered on January 15, 1895.
When Captain Emile Reibell invoked officer’s privilege to flee his stinking ship ahead of the general disembarkation, he quickly discovered two of the reasons why Madagascar would prove to be a tough nut to crack. His first night on shore, spent in a deck chair that he unfolded on the beach, was a torment of swarming insects.
The fact that some of these insects were the cause of “fevers,” and therefore constituted enemies far more lethal than most colonial opponents (especially the Hovas) was at that very moment in the process of being discovered. Nevertheless, even the dullest commander abroad had known for years that low swampy terrain was unhealthy and should be crossed as rapidly as possible.
The reason why the expedition’s commanding general, Charles Duchesne, had not moved quickly into the relatively healthy highlands became Reibell’s second discovery—all was confusion at Majunga. Engineers building a wharf into the ocean soon struck a coral reef that made the approach of large ships impossible, so that the process of unloading became lengthy and tedious. Worse, the Bay of Majunga, which stretched fifty miles to the mouth of the Betsiboka River and which appeared to be a haven of safety to mapbound Parisians, proved to have a swell that would swamp the rivercraft that had been unloaded in pieces and assembled on the Majunga beach. Therefore, the river support system upon which the French had counted to take them deep into the island was scratched from the beginning. Duchesne’s next option to move his expedition forward was to turn to porters. However, this proved impossible, essentially because the planners had skimped on this requirement to save money. Frantic efforts to recruit porters from as far away as Indochina collected only 1,400 men, mostly Somalis from the African mainland. Virtually the only successful recruitment of porters was carried out in Algeria by Major Francois Lamy, a tirailleur officer, who produced 3,500 volunteers, mainly highland Kabyles, by March. When news of the expedition’s logistical difficulties became known, and especially the failure of Indochina to deliver on its promise of 2,000 porters, a further recruiting drive in Algeria netted 2,000 men, mostly the flotsam of Algerian cities and therefore less handy than the Kabyles. These, together with a few Sakalave tribesmen recruited locally, produced 7,715 “auxiliaries” to support an invading army of 658 officers and 14,773 troops.
While small, this proportion of porters was not substantially different from the numbers projected by the planning committee. Rather than carry goods themselves, however, as was the practice in the remote areas of Indochina and Dahomey, the role of these auxiliaries, especially those from Algeria, was to conduct the Lefèbvre wagons pulled by mules. For experienced colonial soldiers, the decision to include two thousand of these wagons in the provisions of the expedition had been a curious one. Of course, the original purpose had been to employ them at the end of the river network to support the final push toward Tananarive. Perhaps the ministry denizens in Paris had envisioned the Lefèbvre wagons lumbering like Conestogas across the clean Madagascan highlands. But now they had to be used in the tropical lowlands. However, here it was quite obvious that they could not advance without a road. So General Duchesne ordered his troops to build one.
The initial reaction of the Hova government to the French invasion appeared to be a sort of agitated apathy, a fact that the Englishman E. F. Knight put down to the fact that the French had bribed many of the court into a state of, if not outright treason, at least one of cautious procrastination. An army was mobilized, which included veterans of the war with the French ten years earlier and younger conscripts, and installed in a large camp outside of Tananarive. “They were a ragged lot, and discipline there appeared to be none,” Knight wrote.
When paraded before the PM or some other great man, they used to raise cheers and brandish their arms, while their officers waved their swords with ridiculous gestures and simulated the slaughter of the foe. These were practically the sole manoeuvres, for the drill their European officers had taught them was now neglected as foreign trickery unworthy of Hova warriors.
Apart from a royal guard armed with Remingtons, about half the army carried rusty Snyders. However, only ten to fifteen bullets had been issued per soldier on the pretext that as each bullet would “kill a man,” stocks were then more than adequate to deal with the French. The rest of Her Majesty’s Forces were armed with muskets and even bows and arrows. “Soldiering in Madagascar for a native was a calling devoid of allurement,” wrote the Englishman Bennet Burleigh, who covered the war from Tananarive for the Daily Telegraph. “There was no commissariat, no pay, no outfit except a rifle, a few rounds of ball cartridge, and a bit of calico.” Indeed, robberies carried out by unpaid, starving soldiers became a real nuisance in the capital.
As the French forces disembarked at Majunga from January to April and advanced inland, the morale of the Hova army, already fragile, eroded further. The Legion contribution to the expedition consisted of a bataillon de marche of eight hundred men drawn equally from the two Legion regiments, which were organized into a régiment d’Algérie with two battalions of Algerian tirailleurs. On May 2, 1895, Marovoay, a town at the mouth of the Betsiboka, was attacked by a French force. Despite strong fortifications, and the fact that they outnumbered the French, the Hovas fled after a bombardment by French gunboats, but apparently not before killing their commander, whom they accused of treason. When on June 9 a battalion of the Legion participated in the capture of Maevatanana, a town on the Ikopa River, from an enemy that bolted so quickly that the legionnaires were unable to cut off their flight, morale plummeted in Tananarive. One Hova soldier told E. F. Knight that at Maevatanana he had been frightened by the “invisible death…. ‘There was no smoke,’ he said. ‘There was scarcely any noise, and yet our men fell in hundreds. We believe that there was magic in it.’ ”
Yet this gloom turned to the brightest optimism when it was learned that Duchesne had stopped his advance to build a road. Knight reported that even the Hovas could not understand why the French did not advance beyond
sickly Maevatanana. We heard that their men were dying like rotten sheep, that their transport had completely broken down, and that they were altogether in terrible straits. We were at a loss to understand how any sane general could keep his force so long in the deadly lowlands instead of pushing on, without delay, at any cost, to the healthy highlands, which were but a few day’s march distant.
How, indeed! Each morning at six o’clock the soldiers took up their shovels, which they had christened “rifle model 1895,” and began leveling the hard red clay into a roadway three meters wide that ran along the right banks of the Betsiboka and Ikopa Rivers. At ten o’clock work was suspended during the heat of the day, and then resumed until five-thirty in the evening. Lieutenant Langlois found that his legionnaires, often working up to their waists in the swamps of the Betsiboka, quickly became depressed: “Nothing is more enervating, discouraging than these long days, these interminable evenings spent in the rot of the swamps,” he wrote in June, “in the middle of the slime, beneath a terrible sun, always surrounded by the same yellow, rocky countryside, always sent to sleep by the plaintive cries of stray dogs.” Furthermore, the logistical system had virtually broken down, so that legionnaires lived virtually by slaughtering stray cattle left behind by the inhabitants. Still, this was insufficient to sustain men engaged in heavy roadwork. Malnutrition, when combined with low morale, hastened death: “Many men are dying at Tsarasotra,” Langlois reported a month later. “All day our men make crosses, dig graves, and every evening it is always the same funereal and lugubrious alignment of white cadavers. Nevertheless, we have grown used to these macabre spectacles, and we live almost indifferently next to that which, two months ago, would have had us all crying.” “Our men are terribly fatigued,” Lieutenant Colonel Jean-Louis Lentonnet of the régiment d’Algérie recorded in his diary on June 11. “The fevers strike down more each day. If this goes on much longer, no one will be spared.” This was no exaggeration. Brigade aid stations and field hospitals allocated to the expedition, calculated on the basis of 1,500 casualties, proved to be desperately inadequate, in part because the budget director believed that medical services were an area in which he could make savings. They soon filled to overflowing with soldiers racked with malaria and dysentery. “Still the roadworks,” Lentonnet catalogued on June 15. “We can no longer count the victims. And for what? To drag behind us the Lefebvre wagons. Whomever decided to send them to Madagascar is a real murderer. The cemeteries are filling up. When are we going to march forward?”
But Duchesne persisted, and persisted. The obvious question to ask is, why did he not look at other options? For instance, from the beginning, when it became obvious that his river support system would not work, he might have shifted his force by transport to the east coast, where as early as December 1894 the French navy had established a base at Tamatave. A second option would have been to hold his troops at Majunga or even at the substantial French naval base of Diégo-Suarez at the northern tip of the island until the problems of river supply could be sorted out, which they were, at least in part, by July through a combination of requisition and improvisation. “How many lives would have been saved if this system had functioned from the 15th of April, as it was supposed to!” wrote Reibell. A third option would simply have been to organize a light force sooner, using the Algerian porters and their mules as logistical support.
Yet he selected none of these options. Perhaps he felt constrained to follow the ministry plan, or felt the river route too insecure or uncertain to supply his army. As a veteran of the Second Empire who had fought in the costly conquests of Tonkin and Formosa, Duchesne simply accepted lack of food and high losses to disease as a normal part of campaigning. Nor was he alone: Silbermann was as horrified as his colonel when, during the 1900 China expedition, a group of young marines came to complain about the lack of food. “We had no right to complain,” he believed, “for we were all volunteers and when you go on campaign, you don’t go to a banquet.” It may also have been the case that tenacity and willpower, considered essential military virtues, in Duchesne’s case, as well as that of his chief-of-staff, General de Torcy, suffocated imagination and flexibility of mind. Langlois described Duchesne as “a little taller than average. His body, a trifle stooped, testifies to the fatigues endured in Tonkin and Formosa. What is immediately striking is his profoundly energetic and tenacious expression. One immediately feels in the presence of an iron will.” In any case, Duchesne’s conduct of the Madagascar campaign did not hurt his career, for he became a member of the Conseil supérieur de la guerre, reserved for the most senior generals.
Unfortunately, Duchesne’s iron will proved more durable than the bodies of his soldiers. By August, the work camps along the Betsiboka and the Ikopa were fringed with forests of crosses and peopled with men emaciated by dysentery and malaria, who “no longer have the strength to go to the Latrine,” wrote Legion Lieutenant Langlois. “They are living putrescence, real moribunds.” Officers who once attempted to encourage faltering soldiers with a word or a drink from their canteen now did neither: “The heart grows hard, [and] in the end one passes by with indifference, head low, before the victims of fatality,” wrote Reibell. By the end of the month, Langlois complained that Legion companies, normally two hundred strong, had been reduced to “seventy or seventy-five men. Entire units have disappeared.” In fact, the Legion, although badly off, still had 450 of its original 800 men able to answer roll call on September 1.
