To Arms

by Hew Strachan

  At no stage, therefore, did the Germans mount a major attack across the Orange into South African territory. Pre-war instructions from the Colonial Office in Berlin, to remain on the defensive, were reiterated by Heydebreck on 4 August. The Germans were unable to exploit Botha’s moment of maximum weakness, in November 1914, for fear that offensive action on their part would undermine the bases of rebel support. When in late January they did plan an attack, it miscarried owing to the Boers’ failure at Upington and their subsequent surrender. By February 1915, when the rebellion was over, the Germans’ moment to launch limited attacks against weak and scattered South African forces had passed. The only real advantage which the rebellion brought the Germans was a stay of execution. Botha was forced to postpone his invasion of the German colony, and thus was boosted the German hope that victory in Europe would come in time to settle Germany’s position in South-West Africa.

  The delay was not, however, a period in which the Germans’ capacity for protracted defence waxed noticeably stronger. Seitz’s consciousness of his vulnerability was accentuated by the fact that the forces available to him were considerably less than they had been a decade earlier. In August 1905, during the Herero rebellion, the Germans had 21,000 troops in South-West Africa. In August 1914 they numbered 2,000. Of the German population of 15,000, about 3,000 were mobilizable reservists. Thus, the Germans’ total strength hovered around 5,000 men. The South Africans consistently exaggerated it, and even after the war put it at 7,000. Such a figure could not have been reached except by including auxiliaries of little military value. The major manpower resource, the native population of 80,000, was deliberately neglected, in the firm expectation that war with an external enemy would provoke at least the Hereros, and perhaps other tribes, to insurrection. The same argument kept the police force, in any case only 482 men, tied to its peacetime role except in the frontier areas. The Boer ‘free corps’ raised just over 100 men from a population of 1,600, and was disbanded after the fiasco before Upington at the end of January 1915.

  In addition to numerical weakness, the Germans suffered from a lack of tactical and operational cohesion. This was not, as it was in the Cameroons, the product of inadequate communications. The completion of the railway in 1910 had been used to justify the reduction of the colony’s garrison. So efficient were the internal wireless links that the Germans used them to excess, feeding the South Africans a flow of valuable intelligence in the form of intercepts. But the Schütztruppen had not, as a consequence, been grouped in larger formations. Instead, they were scattered in squads throughout the country, so as to provide local protection to the German settlers. As late as 10 February 1915, 132 separate units could still be counted. Thus, the senior officers had no experience of higher command. Moreover, in a war of low casualties it was ironic that those with staff training proved particularly vulnerable. Von Heydebreck fell victim to a premature explosion from a rifle grenade on 12 November 1914; his obvious successor had been killed at Sandfontein; and on 31 March 1915 the chief of staff to Viktor Franke, the new commander, died as a result of a fall from his horse. For the major stages of the campaign the Germans had as their chief of staff a reservist without staff training, and they had no officer to run the railway on military lines. They did organize three, and later four, battalions, each of three to four companies. They gave these the title of regiment in a bid to deceive the South Africans, not to reflect their actual strength, which at 450 men was equivalent to about half of a normal battalion.

  The German forces in South-West Africa were therefore small both in aggregate and in their component parts. But South-West Africa could not have sustained forces of any larger size on a war footing. The colony had about 7 million marks in circulation; Seitz reckoned a further 5 million were needed to cover the costs of mobilization and defensive preparations. On 8 August, disregarding the colonial office’s instructions to the contrary, Seitz printed his own note issue, and then introduced a savings scheme to keep gold in circulation.129 With this cover he was able to accumulate sufficient food stocks to provide for the Schütztruppen’s peacetime strength in men and horses for fourteen months. But Seitz reckoned that, for the population as a whole, there was food for five months, and in some areas, including Windhoek, barely enough for three. In October the private purchase of food was forbidden. Only the Ovambo in the north cultivated enough to produce a surplus, and that only in years of heavy rain. The 1915 harvest was bad, and the Ovambo themselves starved. The Herero and Hottentot to the south had been hunters until the arrival of the Germans, and had become dependent on imports of maize and rice. The German farmers concentrated on cattle farming rather than on arable. On one level, therefore, the postponement of the South Africans’ attack worked against the Germans. Oxen and mules were requisitioned to meet the Schütztruppen’s transport needs. Consequently, the livestock normally available for cultivation consumed existing stocks of fodder without contributing to its replacement.130

  The military impact of virtual famine in 1915 was considerable. Units could not remain either concentrated or stationary for long, as they had to disperse to forage and to water. Apart from a camel-mounted company for service in the Kalahari desert, the regular Schütztruppen were organized as mounted infantry with Cape ponies. The loads which they carried were heavier than those borne by the South African commandos; the latter, by riding lighter, put less strain on their mounts and proved far more mobile than their opponents. Moreover, the lack of fodder deprived the Germans of their ability to exploit their one area of real military strength. The Schütztruppen had forty-six guns, in addition to eleven machine cannons and nine light mountain guns; furthermore, they possessed, in dumps at Windhoek and Keetmanshoop, sufficient shells.131 But the guns went short of ammunition for want of food for the oxen to draw the wagons.

  The obvious route by which the Germans could relieve their economic plight lay to the north, through Portuguese Angola and its main southern port Mossamedes. German officials had already prospected across the frontier before the war. Ideas for linking the Portuguese and German railways had been adumbrated. But such talk was not congenial to the Portuguese. Their hold on Angola was incomplete, about a fifth of the colony enjoying effective independence in 1914, and was sustained only by continuous and brutal campaigning. Indeed, so notorious was Portuguese colonialism, so damaging to the cause of European civilization, that Britain and Germany had considered the partition of Portugal’s African colonies in 1913.132

  Anglo-German hostility in Africa both deepened and eased Portuguese fears for their colonies. German rhetoric about a central African empire gained credibility, and threatened Portugal’s two major possessions, Angola and Mozambique. On the other hand, worries about British designs were abated by virtue of the Anglo-Portuguese alliance. The alliance, which dated back to 1386, did not require Portugal to become a belligerent. Indeed, the disorganized state of the armed services, the volatile political position after the fall of the monarchy in 1910, and the lack of any immediate war aim determined the contrary. Nor was there pressure from Britain. Portugal seemed likely to be a liability, not an asset; on 3 August Sir Edward Grey asked it to be neither neutral nor belligerent. But such an undignified stand, with Portugal obliged to Britain but not equal with it, rankled. Little by little, the belief that Portuguese self-respect demanded active belligerence, and that Portugal’s African colonies would thereby be assured of British guarantees, gained credibility.133

  The combined effect of these responses was to move Angola onto a war footing. On 11 September 1,500 troops left Lisbon for Portugal’s West African colony, with a similar contingent bound for Mozambique. In Angola itself the governor-general ordered a state of siege on 8 September. His public intention was to check the banditry of the Ovambo in southern Angola; his true purpose was to stop the Germans’ traffic from Mossamedes, via Humbe, and across the frontier. The troops from Europe, which boosted the total Portuguese strength in southern Angola to between 6,000 and 7,000 men, were to mak
e this barrier effective. On 19 October 1914 a German patrol (according to the Portuguese) or mission (according to the Germans), fifteen strong, was arrested at the Portuguese border fort of Naulila. The Germans’ interpreter, a Dane, deepened the confusion rather than elucidated it. In the ensuing mêlée the German administrator from Outjo and two reservist lieutenants were killed, apparently while making their escape.

  When the news of Naulila reached Seitz he was uncertain whether or not Germany and Portugal were at war. The destruction of the Kamina wireless station precluded regular and direct contact with Germany; transmissions from Windhoek were interrupted by electric storms, and reception (Windhoek could listen to messages between Nauen and the United States) did not necessarily answer specific questions. In reality, as Seitz discovered in July 1915, Germany and Portugal were not at war. But the evidence on the ground— the build-up of Portuguese troops in southern Angola, the closing of the frontier to commerce, and now the murders of German officials—suggested the contrary. Seitz could not afford to have large bodies posted on his northern frontier. But the Boer rebellion gave him sufficient respite to organize punitive actions with a view to negating any Portuguese threat at the outset.

  On 31 October a Portuguese post at Cuangar, its garrison oblivious of the events at Naulila, was surprised and massacred by a small German detachment operating out of Grootfontein. Four adjacent posts were then abandoned by their men rather than face the Germans. Meanwhile a much larger force, about 500 Germans, aided (as the Grootfontein force had been) by local Africans, and commanded by Franke, temporarily quitted the south for an attack on Naulila itself. Franke’s advance beyond the railhead was slow, his column needed 2,000 oxen to move, and the Portuguese were alerted to his approach by mid-November. Franke attacked Naulila on 18 December. The two sides were approximately equal in strength, but the Naulila fortifications had been designed to deal with native insurrection, not the Germans’ six artillery pieces. A lucky shell detonated the Portuguese munitions dump. The Portuguese, poorly commanded and not acclimatized to African service, broke and fled; their losses totalled 182.

  The defeat, though severe, was local. But Alves Roçadas, the Portuguese commander, fell victim to exaggerated notions of German military brilliance. Anticipating a German envelopment, he fell back to Humbe, abandoning all the Ovambo region between the Cunene river and the Rhodesian frontier. Equipped with the arms (including 1,000 rifles and four machine-guns) left by the Portuguese in their panic, the tribes of the entire area rose in revolt, spurred by their hatred of Roçadas, by the evident military weakness of Portugal, and by famine. The Portuguese, now commanded by Pereira d’Eça, confronted a long campaign, punctuated with major battles and conducted with fearful brutality. Pereira d’Eça was alleged to have ordered the killing of all natives aged over 10: some were hanged with barbed wire, others crucified. Franke, meanwhile, retired southwards. Throughout the rest of the South-West African campaign Germany’s northern frontier would be neutralized by a buffer of insurrectionary Ovambo.134

  Franke’s reputation as a fighting soldier, evidenced by his being awarded the pour le mérite for his services in the Herero rebellion, was confirmed by the Naulila attack. He returned to Windhoek to find himself appointed commander of all German forces in South-West Africa. But his tenure of that command suggested that courage and initiative on the battlefield were not allied to strategic or operational resourcefulness. The conduct of the German defence, which in 1914 had not been without its rewards, was in 1915 to be marked by an almost total lack of fighting spirit.135

  Heydebreck, Franke’s predecessor, had correctly identified the main routes by which the South Africans might advance. But the lack of German fortifications at Swakopmund and Lüderitz, and the problems for an invader of crossing the coastal desert strip, had decided him to concentrate his western defences inland at Usakos and Aus. In a plan drawn up in 1911 he had identified the major danger as lying in the south, and had proposed to conduct his principal operations on the Orange river. The course of events in 1914 reinforced his pre-war thinking.136 No landing had taken place at Swakopmund. That at Lüderitz was advancing on Aus, rebuilding the railway which the Germans had destroyed in their retreat, but its progress was slow and easily observed.

  Franke’s strategy followed Heydebreck’s—to fall back into the interior and to the north, forcing the enemy to expend both time and effort in coping with the inhospitable border regions. While Franke was at Windhoek with two companies, two were left at Swakopmund, four were positioned at Aus, and seven were distributed in the south. Franke’s intention to withdraw and Windhoek’s central position in relation to the colony’s railway and wireless communications made sense of his dispositions, provided he remained responsive to enemy movements. But the bulk of the Schütztruppen lay outside the orbit of his direct command, facing south, and not ready to guard the Germans’ line of retreat to the more productive areas of the north.

  On 25 December 1914 the South Africans landed at Walvis Bay. The destruction of Spee’s East Asiatic Squadron on 8 December had removed the major threat to British amphibious operations in the South Atlantic, as well as German hopes of naval success. The German response was extraordinarily lackadaisical. Major Ritter, temporarily commanding in Franke’s absence in Angola, and determined to mount an offensive in the south, reckoned that any advance from Walvis Bay could be pinched out round Windhoek by redeployment from Aus and by Franke’s troops returning from the north. Franke took over from Ritter on 20 January, but even he, though much less optimistic, averred that the operations in the north were no more than a demonstration. The South Africans occupied Swakopmund without opposition on 13 January. The Germans fell back to defensive positions between Riet and Jakalswater. Throughout January most of their efforts were put into reinforcing Aus, and planning the abortive thrusts across the Orange. On 25 February the German command finally acknowledged that the major South African advance might develop from Swakopmund. Offensive and counter-offensive operations south of Kalkfontein were abandoned. But only one company was diverted to the north. Aus remained the largest single concentration, and the troops at Kalkfontein were given the task of protecting it from the south and east.

  With hindsight, the Germans would have been better advised to abandon the south of the colony and to concentrate all their forces against Swakopmund. Such decisiveness, however, would have presumed a greater clarity and urgency in the movements of the South Africans.

  By early 1915 South Africa had at least 70,000 men under arms, of which 43,000 were employed in the campaign in South-West Africa. Such abundance of manpower apparently freed Pretoria from the compulsion to concentrate. Thrusts from Swakopmund, Lüderitz, the Cape, and even across the Kalahari were all possible, and all undertaken. Botha, who on 22 February landed at Walvis Bay to take over command of the northern force, was convinced that the advance from Swakopmund on Windhoek would be the blow that proved strategically decisive. It would sever the Germans’ main axis of communications at its centre and wrongfoot the German strategy of a fighting withdrawal from south to north. He was also persuaded, both by common sense and by the intercepts of German communications, that the Germans would recognize this and withdraw to the north of the colony. He wished, therefore, to co-ordinate the offensives from Lüderitz and across the Orange river in the light of this appreciation. But Botha was not able, at least at first, to give the Swakopmund landing the priority which his status suggested.

  Throughout 1914 the landing at Lüderitz had assumed a primacy which could not now be easily set aside. Progress there was slow: the railway had to be restored, and horses died of thirst or sank knee-deep in the soft sand. The force commander, Sir Duncan McKenzie, proceeded with excessive deliberation. But Botha feared that if he pushed him too hard he would resign, and that his Natal commandos would take offence, thus bringing English-Afrikaner tensions into play. To the south Smuts, although nominally defence minister in Pretoria, was planning his own campaign, combining three bri
gades in a push on Keetmanshoop from the south, and a further column coming across the Kalahari from the east. Botha told Smuts that his advance would be redundant if McKenzie took Aus, as all points to the south and east would then fall automatically. In April he prevailed on Smuts to go to Lüderitz, to direct the southern operations both there and on the Orange river. The penalty, however, was further confusion in Pretoria, where the defence ministry was robbed of its head and therefore could not issue orders on its own responsibility.

  If the manpower superiority available on paper had translated directly into fighting power these frustrations would not have mattered. But Botha’s concept of operations rested on the mobility and horsemanship of the Boer commandos. Manoeuvre, envelopment, and speed were the essentials with which he planned to dislodge the Germans. To do this, the mounted brigades need their own integrated transport. The only alternative or additional means of supply was the railway line from Swakopmund inland. But this had been destroyed by the Germans and had first to be reconstructed. Moreover, the decision was taken to convert it from narrow gauge to the South African standard gauge. Although the earthworks and embankments were intact, progress was sluggish: 42 kilometres in two months, up until the end of February, and 1–5 kilometres a day thereafter. At that rate Boer mobility would be forfeit, and the Germans would have ample opportunity to fall back on their own communications, fighting a series of defensive actions in prepared positions.