To Arms

by Hew Strachan

  The success of the 6th army enabled Frank’s 5th army to renew its efforts to cross the Drina on 13 September. Heavy rain continued to hamper bridging operations, and once on the far bank both armies found their mountain guns outmoded when confronted with the Serbs’ quick-firing field artillery. Potiorek’s struggles to break out of the bridgeheads which had been established by the 15th peaked in the mountainous areas south of Loznica on 21 and 22 September: control of Jagodnja changed four times on the 22nd alone. By then the 5th and 6th armies were held, but their efforts had been enough to neutralize the Serb offensives to their north and south. The Serbs, having penetrated sufficiently deeply into Hungary to force the evacuation of Semlin, broke off the attack in response to the advance of Frank’s 5th army, and were back across the Save by the morning of 14 September. On 23 September the Uzice group and the Montengrins broke into south-east Bosnia, followed a few days later by the first of the snow. Rallying fortress troops to the support of his field formations, Potiorek took the Serbs in their northern flank and by 30 October had driven them back across the upper Drina.

  Krauss had once again urged Potiorek to follow up the retreating 1st Serb army with a direct attack on Belgrade. But Potiorek had too few men to sustain operations simultaneously on the northern and western fronts. His complaints, that Conrad was starving him of troops and, above all, of munitions to feed the Galician front, at first found a sympathetic hearing. The invasion of Hungary, even if checked, had alarmed Tisza, and Berchtold remained dependent on smashing Serbia to carry through his Balkan diplomacy. By October, however, Potiorek’s refrain reflected the consequences of the positional warfare which now prevailed on the Drina front, as on others. Potiorek’s operational solutions were not adapted to the new circumstances, or to the resources available to him, but continued to be derived from textbooks. In military, if not political, circles his efforts to undermine Conrad began to lose him face. Whereas in August he had yet to secure his independence of AOK, by October he had, and the fact that he failed to deliver according to the grandiose expectations he generated reflected badly on him rather than on AOK. He had a ready-made scapegoat in Frank, characterized by one officer as senile, pedantic, and obstinate. But Potiorek protected the 5th army’s commander because he was also malleable. Caught in the crossfire between Potiorek and Conrad, Arthur von Bolfras, the chief of the emperor’s military chancellery, became sufficiently frustrated by the end of September to consider installing Franz Joseph as commander-in-chief, with himself as chief of staff.119

  Strategically, if not tactically, Serbia had won a major victory. Austria-Hungary’s bid to use military might to re-establish its Balkan pre-eminence had been thwarted. And yet its need for prestige prevented the dual monarchy from simply abandoning operations on its southern front. Conrad advocated the adoption of a defensive position against Serbia. He reckoned that a victory against Russia in Galicia, not the conquest of Serbia, would settle the Balkan aspirations of the empire. Potiorek disagreed. The tug of war between the two for troops kept the Austrian forces on the Drina stronger than they needed to be for true defence, but still too weak for a major offensive. The same applied in Galicia. Thus, in neither theatre was a decisive concentration achieved.

  GALICIA

  The embroilment of the Austro-Hungarian army in offensive operations in Serbia, and its inability to bring them to a speedy conclusion, was the third in a succession of blows knocking away the props of Conrad’s plans for operations in Galicia. The first, evident by January 1914, had been his realization that Austria-Hungary could not rely on the support of Romania, a covert adherent to the Triple Alliance. The second was the German decision to concentrate on the defence of East Prussia rather than on the drive into Poland.

  The manpower implications of these losses demanded a fundamental change in operational planning. Romania’s support would have secured Conrad’s southern flank and contributed up to 600,000 men; German operations in Poland would have reinforced the north. Of the four corps in the 2nd army, only two were released on 18 August, a third was retained until the 24th, and the fourth never left the Serb front. The first two corps did not reach Galicia until 28 August, and the third until 4 September. The total Austro-Hungarian strength in Galicia was thirty-seven divisions, and until 28 August only thirty-one divisions. The slowness of this concentration, in part the product of the mobilization against Serbia, but much more the result of Conrad’s decision to delay the mobilization against Russia until 4 August, meant that Austria-Hungary abandoned its attempt to gain any compensating advantage in time.

  Thus, from an anticipation of relative superiority, Conrad had to adjust his expectations to one of inferiority. At first it seemed that he would do so. The decision to deploy further west, on the San and the Dniester, implied a defensive intention. But Conrad was never good at modifying his grandiose strategic objectives in the light of reality. Franz Joseph was later to say of him: ‘We cannot find any suitable sphere of activity for a chief of staff with such soaring plans. We would be far better off with a man who doesn’t want to bridge the ocean.’120 The great envelopment operation, the convergence of the Austrians and the Germans on Siedlitz, east of Warsaw, increasingly played on Conrad’s imagination. The thrust northwards into Galicia began to lose its preliminary, precautionary intention; conceived as a move to clear the line of the Vistula before turning east, it had by 15 August become an operation in its own right, the southern pincer of a joint envelopment of the Russians in Poland. Having deployed to defend, the Austro-Hungarian army advanced to the attack. The long marches with which the Austrians had intended to weaken the Russians were instead imposed on themselves.

  When Conrad’s plans miscarried he blamed anybody but himself. His principal targets were his allies, for their failure to support the invasion of Poland. He argued that OHL had only told him of their abandonment of the Siedlitz manoeuvre when it was already too late to change his intentions. This was untrue. Moltke told him that the Germans would fight defensively in East Prussia on 3 August; on 20 August Conrad responded that he would advance on Lublin and Cholm nonetheless.121 He blamed his intelligence services. The conduct of long-range cavalry reconnaissance by both sides in Galicia, particularly given the suitability of the terrain, was of extraordinary incompetence. On 15 August the full strength of the Austro-Hungarian cavalry crossed the frontier to establish the whereabouts of the enemy. But by then, because of Conrad’s decision to deploy back from the frontier, many regiments had already been on the move for a week. They now had only four days in which to gather information over an area 144 kilometres deep on a 400-kilometre front. They had outstripped their divisional infantry and artillery, and were themselves trained for dismounted action. They therefore lacked the firepower to penetrate the Russian screen. As in East Prussia, the Russian cavalry filled the gaps between corps rather than pushing forward to scout, and so the two bodies failed to encounter each other. The Austro-Hungarian cavalry came back with little information of value. The combination of these demands with a new but ill-fitting saddle halved the horse strength of Conrad’s mounted formations virtually from the outset of the campaign.122

  Conrad expected about fifty Russian divisions to be concentrated on Galicia. Despite the defects in information collection, this was a good enough calculation. His difficulty was in the accurate plotting of their deployment. Early indications suggested that their main force was to the north, in the area Brest-Litovsk-Ivangorod-Cholm, and that any forces on the eastern Galician frontier were falling back rather than coming forward. Aerial reconnaissance reported that there were no major Russian formations on the roads between the line Proskurov-Tarnopol to the north and the River Dniester to the south. This was what Conrad wanted to hear; it compensated for the absence of the 2nd army in Serbia and it confirmed his own wish to strike north. Contradictory information was suppressed. From 18 August Hermann Kövess’s group, containing elements of two corps and responsible for screening eastern Galicia until the arrival of the 2nd
army, reported encounters with Russians. On 23 August a captured Russian officer revealed that the whole of the Russian 8th army was deployed around Proskurov; Conrad still refused to believe it. In fact, not one but two Russian armies were advancing from the east, marching by night but protected by the woods from overhead observation by day.123

  MAP 12. GALICIA

  Thus, neither the actions of his German allies nor those of his Russian opponents could deflect Conrad from his set intentions. As long before as 1878 he had stressed the need to seize the initiative in any war with Russia in Galicia. This emphasis was reinforced rather than moderated by the prospect of numerical inferiority. The possibility that weakness would increase, that the enemy’s strength would grow, were arguments against delaying the attack. Once into the Galician plains the Russians could not be easily stopped.124 Even after the war, Conrad remained adamant that the decision for the offensive had been the right one. He argued that the abandonment of eastern Galicia and the occupation of the Carpathians would have allowed the Russians to turn on Berlin. Furthermore, he had to attack between the Bug and the Vistula in order to protect both Silesia and the Austro-Hungarian communications. A move to the east would not only leave the centre of the eastern front exposed, it would also pull the Germans and the Austrians apart.125

  Although Conrad’s espousal of the offensive was clothed in rational argument, its basis lay in wishful thinking, in the hope of the great envelopment. The Germans, for their part, anxious to have the Austrians relieve the pressure on East Prussia, were not disposed to dampen Conrad’s optimism. Indeed, they themselves could be as fanciful in their expectations as the Austrian chief of staff. On 1 August 1914 the Bavarian military attaché at German headquarters managed to reduce the likely Russian forces facing the Austrians to two armies of five corps each: he calculated that three armies would face the Germans, and that major forces would be kept back to deal with the threats posed by Finland, Sweden, and Japan, and to counter the danger of domestic disorder. He also assumed that the Russians would fall back, trading space for time as they had done in 1812. Thus, an Austrian attack against the South-West Front would be victorious, and Conrad would be free to press on to strike the North-West Front facing the Germans.126 This sort of scenario, which gave the Austro-Hungarian army so decisive a role on the eastern front, was exactly the sort of thinking to which Conrad’s weaker side could so easily fall victim.

  Conrad’s main concentration in Galicia, the 3rd, 4th, and 1st armies, was completed between 19 and 23 August. The 3rd, on the right, was drawn up round Lemberg, facing east. Its task, together with that of the 2nd when eventually it arrived, was to guard the flank of the 4th and 1st armies. These latter two, the 4th round Przemysl and the 1st at the confluence of the San and the Vistula, were to strike north-east. Dankl, commanding the 1st army on the left, was ordered on 22 August to advance in the direction of Lublin, and Auffenberg with the 4th was told to follow a day later in the direction of Cholm. If the Russians, against expectation, attacked further south, moving from east to west, they would be held by the 3rd army while the 4th and 1st described a great arc to the north and fell on their rear. The front covered by these three armies was enormous, about 280 kilometres. It would be extended as the attack to the north developed. Furthermore, the left wing did not enjoy the preponderance of strength over the right envisaged by Conrad in his March 1914 plan. The 1st army was intended to be twelve divisions strong, and the 4th was to begin with eight and then be increased to ten or twelve by elements of the 2nd army. In fact, both the 1st and 4th armies contained nine divisions each. Thus, Conrad’s operational bias in favour of the thrust to the north was not reflected in his deployment. Nor was the eventual arrival of the 2nd army calculated to improve the situation. To fulfil his initial intentions, the 2nd army should have moved to the left, to western Galicia; the railways were available for it to do so. In fact it went to the right, to eastern Galicia, a victim of poor liaison between the operations department and the railway department. Conrad’s response to the situation was increasingly divorced from reality. Rather than reflect concern about his poor force-to-space ratio, he made a virtue of the fact that the 2nd army would arrive both late and in the wrong place, giving the right wing a more offensive role and allowing Brudermann’s 3rd army to push eastwards. Thus, Conrad’s convex line was pushed further outwards, its component parts pursuing divergent objectives, not massing for a decisive attack.127

  Whereas Austro-Hungarian planning lacked coherence despite unity of command, Russian objectives differed precisely because direction was not centralized. Ultimately, however, Russia’s numbers meant that its problems proved less grievous. By the thirtieth day of mobilization the Russians had forty-five infantry divisions and over eighteen cavalry divisions in the Galician theatre. The 9th army forming in Poland added another eight-and-a-half infantry divisions. Against thirty-seven Austro-Hungarian divisions, and two ill-equipped German Landwehr divisions under von Woyrsch screening Silesia, the Russians had a considerable margin for error.

  As in East Prussia, Russian intentions represented two sets of plans superimposed on each other—one drawn up by the South-West Front and one by Stavka. Both assumed that the Austrians would concentrate close to their eastern frontier, round Lemberg. The Front chief of staff was Alekseev. His plan, reflecting his pre-war emphasis on the priority and relative independence of the war against Austria-Hungary, saw Cracow as the key to Austrian control of eastern Galicia. The railway lines running west to east through Cracow were superior in capacity to those crossing the Carpathians. Thus, a Russian attack, launched from the line Ivangorod-Lublin-Cholm towards Cracow, would have the effect of cutting the main Austro-Hungarian line of communications, pushing the Austrians south, and pinning them against the wall of the Carpathians. Stavka, and specifically Danilov, envisaged an operation that was much more cautious in its intentions. Rather than descend from the north, they proposed to advance from the east. However, Danilov recognized the value of Alekseev’s proposal to the development of the second phase of operations, the attack from Poland. The attack on Cracow would drive the forces of the two Central Powers apart and would expose Silesia. Therefore both attacks, that from the north and that from the east, were approved, and were clothed as a masterful scheme for double envelopment.128

  The differing conceptions underpinning this apparent unity were evident as soon as the concentration and deployment of the South-West Front began. The Russian armies, the 4th, 5th, 3rd, and 8th, were drawn up along the Galician frontier from Ivangorod to Romania. Alekseev’s plan suggested that the 4th army, on the right, should be strongest; Stavka’s favoured the newly formed 8th army on the left. In reality the two central armies claimed twenty-two of the then-available thirty-eight divisions. The strength in the centre conformed with the Russians’ continuing expectation, based on their acquisition of pre-war Austrian plans, that the enemy would concentrate round Lemberg. They proved very reluctant to modify this supposition, despite intelligence reports to the contrary. The 4th and 5th armies on the right were in conjunction stronger than the 3rd and 8th on the left. The latter, according to Alekseev’s plan, should have advanced first, to pin the Austrians in eastern Galicia. In practice, the Grand Duke’s anxiety to respond to the urgent promptings of his French allies now—as in East Prussia—led him to push his armies forward before their concentrations were complete. Still convinced that the Austrians were around Lemberg and about to advance to the east, the Grand Duke cut across the intentions of the Front command and prematurely pressed the 4th army across the San and west of Przemysl. The poverty of communications on the southern wing soon slowed the advance of the 3rd and 8th armies. Thus, the 4th and 5th armies, moving from north to south, encountered the main weight of the Austro-Hungarian army, the 4th and 1st armies, advancing from south to north and not as yet distracted by any major threat to the east.

  On 23 August the 4th Russian army and the 1st Austro-Hungarian army ran into each other east of the San, and o
n the 24th Dankl was pushing the Russians back towards Krasnik. The Russians still refused to modify their expectations of Conrad, and, despite its superior numbers, saw Dankl’s 1st army as independent from the 4th, which was construed in its turn as the flank guard for the anticipated Lemberg concentration. Therefore they interpreted the Krasnik operations, not as the major Austro-Hungarian thrust but as an opportunity for isolating and enveloping Dankl. The commander of the 4th Russian army was blamed for the tactical setback at Krasnik—once again, troops on the advance had attacked semi-fortified positions with insufficient preparation129—and was replaced by Alexei Evert. The 9th army, forming in Poland, detached a corps to reinforce Evert’s right, and the 5th army (commanded by Plehve) moved to its right to threaten Dankl’s rear. Dankl had, for the time being, relieved any Russian pressure on Silesia by way of Poland.

  Conrad still underestimated the threat to the east, and therefore he too pivoted round Dankl. But the moment for an Austro-Hungarian victory on the left was already passing. The apparent success of the 1st army had to be weighed against the slow arrival of Auffenberg’s 4th army, whose concentration was still not complete when its advance began. Dankl was pulled in two directions, first to his outer flank and the possibilities of lapping round the Russians, and second to closing the gap to his right. The latter prevailed. Thus, the 1st army was drawn into a frontal battle, and its movements made to conform to those of the 4th army, not the 4th’s to the 1st’s. Furthermore, the 4th was due to be reinforced on its right with three divisions drawn from Brudermann’s army. By the evening of 25 August, when Auffenberg’s 4th army was coming up on Dankl’s right, Conrad’s focus was not on the 1st army and its thrust towards Lublin, but further east, on the 4th and Krasnostav. Auffenberg’s army therefore hit Plehve’s left flank as Plehve hit Dankl’s right. Auffenberg, his audacity fed by his gross underestimate of the strength of Plehve’s force (he imagined that only three divisions confronted him), pushed his flank forward hoping to envelop the Russians. His prospects of doing so were facilitated by Plehve’s attention to Dankl and by the imminence of the three divisions from Brudermann. By 29 August, after three days of confused and desperate combat, Auffenberg’s forces had not only pushed in Plehve on his left but had also separated him from Evert on his right. The three corps of the 5th Russian army were in immediate danger of envelopment. On the 30th Plehve decided to fall back on Krasnostav. By doing so he would expose Evert’s left flank. Simultaneously, the news of Samsonov’s defeat and the apparent imminence of German victory in the west suggested the possibility of a German attack into Poland, and thus danger to the rears of the 5th and 4th armies. Conrad’s northern thrust between the Bug and the Vistula seemed on the verge of success.