To Arms

by Hew Strachan

  Nonetheless, a strategic breakthrough was achieved between March and October 1915. It came on the front where neither Merlin nor Dobell had sought it. Its significance was therefore more evident to the Germans than it was to the Entente powers.

  In April 1915 the dominant concerns of both sides in the northern Cameroons were defensive. A German thrust to the Benue on 10 April caused the Emir of Yola and his native administration to flee, so threatening to undermine the British hold in northern Nigeria. Lugard, therefore, wanted Garua taken to restore British prestige in northern Nigeria, not to give Britain control of the Cameroons. The purpose, however, of the German attack was to ensure that Cunliffe’s forces did not simply mask Garua in an advance into the highland plateau. Zimmerman had told von Crailsheim in January that the task of the Garua garrison was to protect Ngaundere. Provided the Germans held Ngaundere, they could exploit their interior lines in order to concentrate sufficient men against individual enemy columns scattered over all fronts. If Garua fell then Ngaundere would follow, and, in Zimmerman’s words, ‘the whole war plan would collapse’.92

  Crailsheim’s faith in his chief’s analysis was confirmed by the capture of a letter from Morisson, saying that the French forces in the south could not advance any further until Cunliffe’s column in the north had captured Garua. But Zimmerman changed his mind. In fresh orders, dated 13 April 1915, Zimmerman now warned against the dangers of locking up all available forces in a fort. He feared that Cunliffe might mask Garua, and fall on Ngaundere and Banyo. In such circumstances the defence of Garua, however heroic and protracted, would be useless. Garua’s garrison was cut to one-and-a-half companies, a total of 250 rifles, so as to create a mobile force to cover the flanks of Ngaundere and Banyo. To add insult to injury, Zimmerman also took two officers and 60,000 rounds (although the Garua garrison was convinced that the total was twice that) to reinforce the Südabteilung. Crailsheim protested that in the rainy season the high water meant that Garua could not be bypassed; he argued in vain.93

  Cunliffe, with fourteen companies in all, plus supporting arms, now enjoyed an overwhelming superiority. But he continued to treat Garua with the respect its fortifications deserved. He supplemented his mountain guns with two heavy artillery pieces, a naval 12-pounder sent up the rivers Niger and Benue when the waters were high enough, and a French 95 mm from Chad. By 9 June his infantry had worked forward to within a kilometre of the German positions, the British to the south and south-east and the French under Brisset to the east. He hoped to cut off Crailsheim’s retreat, but the askaris were alarmed by their first experience of heavy artillery fire. Half the garrison escaped by swimming down the Benue, and thence to Banyo; the balance, 300 men, surrendered without an assault having taken place, on 10 June.

  Cunliffe was oblivious to the enormity of the blow which he had delivered the Germans. He was convinced that Jaunde, not Ngaundere, was the centre of their defensive effort, and that his role was, as the Duala conference had specified, to support Dobell. Dobell, however, was in retreat, and could not resume his advance until the rains ceased. Therefore, Cunliffe concluded, the sensible thing for him to do was not to push south but go north, and reduce the now-isolated Mora with his heavy artillery.

  Brisset, technically his subordinate, disagreed. Of all the allied commanders, both Brisset and his immediate superior, Largeau, the military governor of Chad, had been the most impressed by the intelligence pointing to the German reliance on Ngaundere. The fertility of the highland region, the support of neighbouring Chad, and the relative lack of rain made rapid movement easier in the north than in any other area of operations. On 28 June a British column took Ngaundere. Brisset then moved south-west on Tibati, and pushed a company south to Kunde to link with Morisson’s right. By late July Brisset’s thoughts and actions aimed at a total inversion of allied planning, placing the initiative on a general offensive with the North Cameroons force.

  For the British, Brisset’s independence was not a manifestation of a different strategic view but of insubordination and French bloody-mindedness. Cunliffe concluded that the rainy season would preclude immediate operations south of the line Kontscha-Banyo-Bamenda, and that those positions should therefore be consolidated, whereas the lack of rain in the north created the ideal opportunity to take Mora. Largeau’s ambivalence—recognizing the military need for obedience, while being in sympathy with Brisset’s objectives—gave Brisset continuing leeway. That leeway was consolidated by delays in communication with Merlin, Dobell, or Aymérich. The subordinate status of Brisset’s command, in reality clear enough since February, was not reaffirmed until the end of August. However, Brisset lacked the men to achieve the objectives which he had set himself without Cunliffe’s aid. Thus the Germans had time to regroup.94

  With the fall of Garua the Schwerpunkt of German defences did at last shift to Jaunde. Thus, for the first time in the war German strategy and allied objectives were brought into line. Zimmerman’s lifeline could no longer be the northern plateau; it had now to be the resources of neutral Spanish Muni, and therefore the axis of his operations became Jaunde and Ebolowa, both of them supply dumps and manufacturing centres. In the north he created a new Nordabteilung by drawing troops from Banyo and Dschang, and sending them east. Too late to hold Ngaundere, they centred their defences on Tibati and Banyo, forming an arc that ran west to Ossindinge through Dschang to the Sanaga, a front of 1,000 kilometres. Equipment was pulled back south of the Njong. Allied confusion gave Zimmerman sufficient respite to plug the gaps in the north, at least for the time being.95

  Dobell’s retreat and the coming of the rains brought a pause in the west which allowed Zimmerman to improvise an attack in the east against Morisson’s communications between Bertua and Dume, the aim being to drive the French once again beyond the Kadei. The Germans’ intention was to reinforce their left in order to lap round Morisson’s northern flank. But the command structure was confused; the troops drawn from the west and from Banyo were still formally part of the Nordabteilung; and Duisberg, the officer in charge of the attack, was pulled south by his concerns for the dangers from that quarter. Thus, by 13–16 September the German offensive, their last major effort of the campaign, had stalled. Even as it did so, Zimmerman drew off two companies to the south and one to the north, thus blocking any hopes of its renewal.

  Broadly speaking, therefore, the major German concentrations between July and September were pointed north and south, not east and west. The latter were, however, the points from which the British and the French proposed to press their advance. The fall of Ngaundere had convinced Dobell that Jaunde was now his proper objective in a way it had not been in March, and thus only its reduction, not the blockade of the coast, would force the Germans to surrender. Moreover, he was determined to create a force sufficiently strong to enable his columns to take Jaunde even if Aymérich could not give them effective support. The French were equally decided that Aymérich should be enabled to advance on Jaunde from the east even if British action in the west remained limited. On 25 and 26 August Merlin and Dobell, this time with Aymérich also present, conferred once again at Duala and a joint thrust on Jaunde was fixed to begin after the rains were over, on 15 October in the east and on 15 November in the west. The problems of communication with Cunliffe’s command reinforced the decision that, as before, opportunities in the north should be subsidiary and supportive.

  The second Duala conference also discussed an attack from the south. Merlin and Aymérich favoured a thrust from Campo on the coast, designed to combine with the French columns coming up from Gabon. They had been prodded in this direction by Doumergue, who, with one eye cocked towards French territorial ambitions and the other to the German line of communications to Spanish territory, wanted to give the French companies at Campo a role more significant than that of ‘frontier guards and customs officers’. But Doumergue was in Paris and had to defer to Dobell in Duala. The latter was the de facto allied commander in the Cameroons, and he preferred to concen
trate his forces. He was supported by Mayer, who was anxious that he, rather than the Campo contingent, should secure any available reinforcements. Britain’s material superiority meant that Mayer was struggling to keep France’s end up in the advance from the west; only two months previously Doumergue himself had reminded him that, although he was Dobell’s military subordinate, he was also ‘the representative of the Republic in the Cameroons’.96

  Such grandiloquent language might have surprised Aymérich, whose role in the campaign now became direct. With Morisson’s and Hutin’s columns adjacent to each other, but with the two column commanders at loggerheads, Aymérich could assume personal control of the eastern advance. His left, under Hutin, moved along the Njong, its task being to protect the major thrust by Morisson, whose right rested on the Sanaga, feeling north in the hope of linking with Brisset. In practice, the balance in Aymérich’s advance was reversed. Hutin’s column encountered the heavier German resistance, while Morisson’s progress on the Sanaga was comparably easier. Moreover, the link with Brisset proved elusive. Not until 6–8 December did news filter in of Brisset’s position, and he was then still 150 kilometres from the Sanaga. Of Dobell’s movements Aymérich knew nothing; in going forward to exercise command himself, he lost contact with the other columns. By late December his isolation and his lengthening communications forced him onwards, the choices confronting him being the rapid seizure of Jaunde or retreat. On 2 January 1916 his gamble began to pay off. He learnt that Brisset was just north of the Nachtigal rapids, above Jaunde on the Sanaga. Two days later he heard that Dobell’s British column had reached Jaunde. The co-ordination of the columns, even if not their communication, had been achieved, and thus their individual movements had had reciprocating effects.

  In the west Dobell’s British column had retaken Wum Biagas on 9 October and his French column Eseka on 30 October. Both places were prepared as forward bases, linked by motorized transport to Edea, so that the final assault on Jaunde could proceed without a halt. A force of 9,700 men was assembled. Their advance in late November avoided the tactical failings of the first attempt: a strong column on the forest road was flanked on either side by detachments on as wide a front as possible, so checking German sallies against the rear. By mid-December the British were out of the jungle and into open country. The French reached a comparable point a week later, at Mangeles. The British pressed on without waiting for the French, and entered Jaunde on New Year’s Day 1916.

  To the north, converging forces from Bare and Ossindinge had cleared Dschang by 6 November. Cunliffe, recalled from Mora without capturing it, directed Brisset on Tibati, and most of the rest of his force on Banyo. Banyo was a strong-point as formidable as Mora and Garua. The position rose to a height of 400 metres, and was formed of a network of 300 stone breastworks or sangars. The advance of Cunliffe’s troops up its steep slopes on 4 November was protected by mist; holding the ground gained the following day, they assaulted the summit on 6 November. On 2 December the Ossindinge column took Fumban, and by 18 December Cunliffe’s command was arrayed on the front Joko–Linte–Ditam. On 8 January 1916 his and Aymérich’s troops linked at the Nachtigal rapids.

  The allied advances from east, west, and north had encountered little sustained German resistance. In part, this was the product of superior numbers, and (at last) efficiently organized supply columns and better-protected lines of march. The Germans, as they were boxed in, should have profited from a shortening of their lines of communication to put up a more vigorous defence. In reality this was impossible. None of Zimmerman’s units had sufficient ammunition to allow itself to be drawn into a sustained firefight. Most of the rounds issued to askaris were the ersatz manufactures of Jaunde and Ebolowa. One company facing Morisson’s column fired eight rounds per man on 29 November; by the end of the following day each soldier had only twelve rounds, and then only four after further fighting on 1 December. An effort by Zimmerman to counter-attack against Dobell on 16 December lasted forty minutes, until his men ran out of ammunition.97 Retreat was therefore unavoidable, and as they fell back the Germans could neither plunder the ammunition of their enemies nor, increasingly, rescue their own spent cases for recycling.

  But the softness of the German defences had another cause. The sources of ammunition lay to the south of the Njong—ersatz production was based at Ebolowa, and the only route for communication from Europe lay via Spanish Muni. And yet the thrust of the allies’ advance on Jaunde was directed north of the Njong. Aymérich’s efforts had emphasized not his left but his right, Morisson’s column and the link with Brisset. Dobell was similarly resistant to being drawn south. He was told by the War Office on 2 August of the importance of the axes between Jaunde and Muni and Jaunde and Kribi, and of the road from Ebolowa to Ambam.98 But in his desire to take Jaunde without having to rely on Aymérich, Dobell concentrated his forces further north, on the direct route to Jaunde, and made clear to the French that he could not support Le Meillour.

  The task of cutting the Germans’ line of retreat, therefore, lay with the Gabon columns. Both the Brazzaville conference in February and the second Duala conference in August had paid lip-service to the need to block the Germans off from Muni. But in the allocation of manpower this objective never received sufficient priority. In July Le Meillour’s efforts to clear the eastern side of the Muni border were rewarded with the capture of Ojem and Bitam, but were halted by the Germans on the River Ntem. Merlin and Aymérich hoped to support Le Meillour by reinforcing a column landed on the coast at Campo, and intended to stop traffic on Muni’s northern border. But Dobell refused to favour the Campo column at the expense of the advance on Jaunde, and even withdrew the troops which he had provisionally allocated to it. Merlin appealed to Paris for six companies from French West Africa. On 20 October 1915 two companies landed at Campo from Dakar, but without maps, interpreters, or guides. On 15 December two further companies and a proper complement of supporting arms made the Campo column a going concern. But it was too late. Le Meillour’s efforts to cross the Ntem in late October and again in late November were checked, not least because for two months part of his command was diverted to Akoafim. Le Meillour finally got beyond the Ntem in mid-December. Even then his progress towards Ambam remained slow.99

  Thus, in December the Spanish frontier still remained open for 200 kilo-metres. Zimmerman determined that he would take his command through this gap and into neutral territory. The German decision to retreat was therefore not prompted by the fall of Jaunde. As at Duala, the Germans preferred not to sacrifice the investments and developments of peacetime colonialism in a short-term defensive action, but instead to evacuate the town before Dobell’s arrival. What proved decisive was Le Meillour’s crossing of the Ntem, which made the cutting of German communications only a matter of time.100 On 28 December, four days before the British arrival in Jaunde, Zimmerman issued orders for the retreat of the German forces to the south-west. The axis Jaunde-Ebolowa-Ambam now became the corridor whose collapsing walls must be kept open long enough to allow the evacuation of the Schütztruppen still in the north and west. The gap on the frontier closed to 50 kilometres, but by 15 February a great exodus had been accomplished. About 15,000 people—and possibly more—had crossed into Spanish territory, including 1,000 Germans, 6,000 askaris, and 7,000 family and followers. The loyalty of the Beti of Jaunde to the Germans was the major factor in ensuring that, even in defeat, the Schütztruppen retained their integrity and their cohesion as a fighting force. Thus was the German policy of integrating the wives and children of the askaris vindicated; thus too was evidence provided that many in the Cameroons saw the German defeat as no more than temporary.101

  The failure of the Campo and Gabon columns to cut off the German retreat was complemented by the delay in the allied pursuit from Jaunde. Dobell was slow to recognize the direction the Germans had followed. Some anticipated that the Schütztruppen would break into small units to continue guerrilla operations within the colony. Dobell himself reckone
d that major units still lay north and west of Jaunde. When he did finally realize that his efforts should be bent to the south, he was anxious not to press the pursuit too hard for fear of driving the Germans across the frontier before the Campo and Gabon columns had been able to form a cordon to block their path.

  Aymérich was put in command of the pursuit. But his eastern columns were exhausted, and their lines of communication extended. Not until 14 January was he ready to leave Jaunde. Hutin’s column moved south and south-east to Sangmelima, to mop up any pockets of German resistance in that direction, and to be ready to outflank Ebolowa. Two columns, one under Haywood and another under Faucon, moved directly against Ebolowa, Faucon entering the town on 19 January. The supply problem was now acute. Hutin’s column was left to live off the land. But the area between Jaunde and Ebolowa had been stripped of supplies by the retreating Germans, and the Beti had abandoned their villages in deference to German wishes. Dobell therefore argued that forces the size of Aymérich’s could not be sustained in the south-western Cameroons. Moreover, for him the fall of Ebolowa marked the end of German resistance. Haywood and the British forces in the pursuit were directed towards Kribi. Not until 28 January did Dobell reverse his decision, and by then Haywood’s resumption of the march to the south was too late.