CHAPTER 9

THE GERMANS

Major Albrecht Count von Stosch was worried about his machine guns. His regiment, the 11th Grenadiers, held the German line at the far left of its division, the 117th Infantry. The 11th Grenadiers’ territory ran northward up the Barrois ridge all the way to Montfaucon. Stosch, newly appointed as regimental commander, was well aware that a defensive line is most vulnerable at the “seams” between units. It was his responsibility to make sure that the seam between the 117th and its neighbor to the east, the 7th Reserve Division, held firm if attacked. Earlier in the afternoon of September 25 the 7th had borrowed three of his heavy machine guns to support a large raid they were to make that night on the American lines. Stosch had been counting on those guns to cover the seam. Now there was a dangerous gap on the eastern end of his own line, made the more ominous because the major had no reserves left to plug the hole.

To calm his anxiety, Stosch paid a visit to the headquarters of his neighbor, Major Kuhr, commander of the 66th Infantry Regiment, which had borrowed the guns. No problem, Kuhr reassured him, he would have his machine guns back before dawn. Anyway, the raid would be against the enemy troops directly opposite the gap, so they could not possibly mount a surprise attack. Unconvinced, Stosch ordered Lieutenant Baum, the commander of the Liaison Company, to seal off the gap as best he could. He positioned a light Minenwerfer (trench mortar) platoon there and ordered his 1st Battalion a bit eastward into a ravine below Septsarges, except for a machine gun company, which he installed in a small hollow north of the Bois de Montfaucon. Then, having done what he could, the major waited for events to unfold.¹*

* Throughout, translations from German primary sources are by Julie Allen and Neil Berkowitz; I have edited these for military terminology and usage.

One may reasonably wonder why Major Stosch was so concerned. On the map, the German position looked almost impregnable. The western end of the Meuse–Argonne sector was dominated by the high ridge of the Argonne Forest, crowned with dense, tangled woods, which overlooked the low ground to the south while providing excellent artillery cover against troops attacking up the Aire River valley to its east. On the extreme left (as seen by the Germans, who were facing south), the Meuse River protected their flank while, beyond, the Heights of the Meuse gave panoramic views of the American positions, which were laid bare to the heavy guns the Germans had hidden among the hills. In the center, the ridge leading south from Montfaucon, with its eroded spurs running east and west, was an ideal place for machine guns to fire on soldiers struggling up the slopes. Montfaucon itself, studded with concrete bunkers, offered superb views of the entire line.

The parallel spurs leading away from the Montfaucon ridge and the ravines that separated them were perhaps the most formidable natural defenses the Germans could hope for. Mostly unwooded, the slopes provided clear fields of fire for guns positioned on the ridges above them. As attackers advanced up a south-facing slope, they would be caught in the open by machine guns firing downward, parallel to the ground. Should the ridgeline be taken and the attacking infantry swarm over the top and down the opposite side, they would have no shelter from the guns on the next spur to the north, which could rain fire on them from above, seeking them out in the trenches and shell holes where they had taken cover. From the spine of the main ridge leading north up to Montfaucon, four such spurs ran to the east and four to the west.

Although much of the landscape had been cleared for farming, patches of woodland added to the defensive value of the terrain. The lower slopes of the sector held by the 117th Division, close to the American lines, were covered by a series of contiguous woods about four miles wide and two and a half miles deep in all. From west to east they were the Bois de Cheppy, the Bois de Béthincourt, the Bois de Montfaucon, and the Bois de Malancourt. These woods were not the patches of bare sticks that protruded from the ground at Ypres and the Somme; many of the original trees remained standing, and enough time had passed since the Verdun battles for dense brush and brambles to cover the ground. In peacetime, these had been communal woodlots for the towns whose names they bore; in war, they were ideal defensive positions.

The Bois de Montfaucon, in particular, was a formidable obstacle. Two or three miles down the slope from the town of Montfaucon itself, this large expanse of woodland was divided laterally by a narrow valley with the obscure name of Ravin de Lai Fuon. An enemy coming up through the lower half of the wood would emerge from the shelter of trees and scrub into a wide but shallow clearing; from the far side, only a few tens of yards away, German machine guns and infantry would be able to fire on them while remaining concealed. At the Ravin’s eastern end it opened to form the Golfe de Malancourt, a triangular field bordered on the south by the Bois de Malancourt and on the north by the Bois de Cuisy, an extension of the main Montfaucon wood. Guns positioned just inside the southern edge of the Bois de Cuisy would have an unobstructed field of fire one and a quarter miles wide and the same deep.

Other, smaller woodlots were scattered around the sector of the 117th. As one passed north and northwest from Montfaucon one came in succession upon the Bois de Beuge; a small wood known only by its elevation as Bois 268; another nameless wood designated Bois 250, to its east the Bois des Ogons; and north of them, and four miles north of the town, the twin woods of Bois de Cunel and Bois de Faye. None of these were larger than half a square mile. When the attacker was distant, they were good places in which to hide long-range artillery. As the enemy approached, they provided concealment for machine gun emplacements that commanded the open country around them.

Even the ground underfoot worked against the attacker. The 1916 battles for the Mort Homme and Hill 304 and the French counterattacks in 1917 had carved the landscape with trenches and pulverized it with shell fire. As an American intelligence officer described it, “The whole ground was pitted with shell craters; the half destroyed timber had sprouted up in a thick second growth; and across this tangled and broken ground ran elements of old trenches, wire entanglements and obstacles of every sort in almost inextricable confusion.”² Here and there were ruined villages, their remains now practically obscured by grass and vines. When it rained, which was often, the ground melted into the gluey mud that had sucked at the boots and wagon wheels of Count von Moltke’s army 58 years earlier. An editor of Stars and Stripes later wrote, “Whatever Prussian evolved the motto ‘Gott mit uns’ must have been thinking of the topography of the Meuse–Argonne.”³

The Germans, of course, had not left the security of their defense solely to nature or to the detritus of past battles. But although their line in the Meuse–Argonne had been essentially stable since September of 1914, the defenses constructed in that sector, with some exceptions, were sketchy. In early 1917 Ludendorff had withdrawn his armies from the front between Arras and the Aisne to rest and reorganize in preparation for the great Spring Offensives of 1918. To cover the new position he had built a strong defensive zone from Lille in the north to Champagne, just west of the Argonne Forest. Later in 1917, almost as an afterthought, the construction was continued to the southeast, atop the old 1914 lines, as far as Metz. The various segments of this defensive zone were named after mythical Germanic heroes; the one in the Meuse–Argonne was called after Kriemhilde, the wife of Siegfried. The Allies named the whole complex the “Hindenburg Line.”

* Ölberg, in the original. No such place appears on the German Army maps of the sector available to me, although they show the bend in the German line. Modern French maps show a roadside calvary just east of Malancourt; perhaps it was originally located to the west of the town, and was appropriately nicknamed by the German soldiers.

As with the other parts of the Hindenburg Line, the name Kriemhilde denoted not only a specific set of obstacles and strong points but also a deep defensive complex consisting of at least four mutually supporting positions, or Stellungen. The first position in this sector was the Wiesenschlenken-Stellung and its eastward continuation, the Haupt-Stellung. This defensive belt was about three miles deep, from No-Man’s Land to the northern edge of the Bois de Cuisy. It contained three lines of defensive positions. The first, only 30 or 40 yards from the American lines, consisted of little more than light machine gun nests in the swampy ground at the foot of the Bois de Montfaucon. About a hundred yards behind that, a parallel trench line supported the first line. The Haupt-Widerstands Linie, the “main line of resistance,” was just over a mile further up the slope, on the north side of the Ravin de Lai Fuon, for as long as it remained within the Bois de Montfaucon. Once the line emerged to the east, however, it took a sharp left turn at a point the Germans called the Mount of Olives,* west of Malancourt, and climbed the hill northward for one and a quarter miles, turning east again once it got to the level of the Bois de Cuisy.⁴ The western side of the sector was thus the most vulnerable, because the main line of resistance was relatively close to No-Man’s Land, while the woods shielded advancing soldiers from view until the moment they emerged into the Ravin de Lai Fuon.

GERMAN DEFENSIVE LINES (“HINDENBURG LINE”) IN THE MEUSE–ARGONNE

Behind the main line of resistance was the Hagen-Stellung, covering the southern slopes of Montfaucon; only partially completed, it contained a number of concrete shelters that were badly built and quickly became waterlogged when it rained. Otherwise, it was not much of a line at all. Still further north, running through and over Montfaucon itself, was the Etzel-Stellung. The barbed wire was in place here, but while the trench lines had been laid out and were well sited they had so far been dug only to a depth of a foot or two. The peak of Montfaucon, part of the Etzel-Stellung, was a daunting objective in itself, rising 360 feetabove the American lines. But although its crest was covered with concrete blockhouses (American engineers later counted 17 of them) and over a hundred dugouts, the hill itself was organized for observation and shelter, not defense.⁵ Still, in the words of an officer of the American 37th Division, “Deep trenches and seemingly endless belts of barbed wire encircled the slopes rising to Montfaucon and gave notice of the difficulties to be encountered in storming the stronghold.”⁶

Above Montfaucon, and extending northward from Nantillois and the Bois de Beuge, was the Giselher-Stellung. This was simply a string of natural positions that included various woods and ridgelines, but which offered many sites for machine gun nests and light artillery emplacements. Finally, the Kriemhilde-Stellung proper ran through the wooded heights of Cunel and Romagne roughly six miles north of the German front line. Begun in 1917 but never completed, it nonetheless sported a strong wire defense; its shallow trenches were made up for by the commanding positions on which they were sited. Behind that there were no more defensive lines; if the Germans had to retreat beyond the Kriemhilde-Stellung, they would be fighting in the open with their backs to the Meuse River. *

* Each of these Stellungen was in turn subdivided and there were many switches and connections among them, so that their exact number and the meaning of the term itself were matters of confusion even at the time. See, e.g., G.M. Russell, “Summaries of Intelligence, 5th Army Corps, St Mihiel–Meuse Argonne,” August–November 1919, C.E. Fogg Papers, World War I, USMA, entry for October 2, 1918, p. 1.

That the lines comprising the Kriemhilde-Stellung were only partially finished was, in itself, not much of a handicap. Since 1916 the Germans had abandoned a defense based on continuous trench lines, as had been used by both sides for much of the war. Instead, they adopted the defense in depth. Front lines were held thinly by small groups of men in fortified outposts. Artillery and machine guns were distributed throughout the defensive zone, placed so that their fields of fire overlapped and protected by emplacements such as pillboxes or blockhouses. This maximized the effect of defensive machine guns, which could sweep large areas. If one machine gun nest was attacked, the others could fire on its attackers. Concealed emplacements that were bypassed by the first line of infantry could chew up the second or third waves as they advanced, or fire into the flanks and rear of the formations that had passed them by. The attackers who were not killed or wounded would be forced to take cover in shell holes and old trenches, making them sitting targets for plunging shell fire from artillery situated behind the line. Scattering the machine gun and artillery positions over a broad area also diffused the effect of enemy artillery, which no longer had defined trenches to shell. It forced attacking formations to expose their flanks to machine guns located in the strong points and to lose cohesion. German infantry, held in reserve for the purpose, could then counterattack isolated groups of enemy soldiers. The counterattack came in two forms: the Gegenstoss, or immediate counterstroke, a quick strike on the attackers without extensive preparation, timed to catch the enemy when he was still disorganized from his recent advance; and the Gegenangriff, or deliberate assault, launched after a preparation lasting as long as several days against an enemy that had already consolidated itself in the captured positions.⁷ If the counterattack failed to dislodge the enemy and the Stellung had to be abandoned, the remaining guns and men could quickly be withdrawn a few miles to the next defensive zone, where the process would begin again. In short, the German defenses no longer aimed to stop the enemy like a brick wall. They would absorb him like a sponge, then squeeze.

No terrain on the Western Front, save the nearby Heights of the Meuse, was as perfectly suited to these tactics as the Meuse–Argonne sector. Every ridge, rise, and hillock could be fortified as a strong point. Every large shell hole, fold in the ground, ruined house, or patch of brambles could conceal a machine gun and its crew. The climbing hills gave long fields of fire in all directions. A well-manned, well-trained, and well-equipped army could hold this line indefinitely.

But the German Army of September 1918 was not the efficient machine of 1914, or even of Ludendorff’s huge offensives of the past spring. As long ago as the winter of 1917 the Germans had begun inducting undernourished youths, convalescents, and the elderly. Training was greatly cut back. Acts of rebellion began to appear. Men refused to board trains for the front. Almost 18,000 soldiers were under arrest for indiscipline; in the city of Cologne the Army counted nearly 30,000 deserters.⁸ By the beginning of 1918 it was clear that most of the Army—particularly the older men, who made up a majority of the troops—had neither the physical nor mental capacity to serve as the “attack divisions” (Angriffsdivisionen) that Ludendorff required. Only about one-quarter of the infantry divisions were designated for that role; they were assigned the highest-quality replacements, supplies, and weapons and were trained in the new, sophisticated assault tactics. The others were classified as “positional divisions” (Stellungsdivisionen) and were assigned to defense. While the latter still possessed assault units, these were a minority, and did not compensate for the poor overall quality of the rank and file.⁹

It was with the attack divisions that Ludendorff gambled and lost in his five assaults against the western Allies between March and July 1918. Having suffered heavy casualties during the offensives, these units practically evaporated in the French and British counterattacks. In June, July, and August, the Germans lost 1.2 million men, of whom roughly half were prisoners or “missing,” most of the latter being deserters. Since the end of May, divisional strength on the Western Front had fallen from about 200 to 125, of which only 47 were rated fit for combat.*¹⁰ Ludendorff had to break up some divisions to maintain the others at strength; to stretch manpower yet further, infantry battalions were reduced from four to three companies and some companies were reduced to 60 men. Additionally, soldiers had been taken from the front to work in industry and agriculture; the women, foreign workers, and prisoners of war who were sent to the farms and factories did not suffice to prevent severe shortages of industrial products and food. By the late summer of 1918, 2.4 million German soldiers had been diverted for this purpose.¹¹

* The number of available German divisions is often given as 185; see, e.g., Trask, AEF and Coalition Warmaking, p. 115. That figure is taken from the writings of British Major General Sir F. Maurice, who published only a year after the Armistice; see Sir F. Maurice, The Last Four Months: The End of the War in the West (Casswell and Co., 1919), p. 122. I have chosen to rely on the figure given by Herwig, who wrote in the 1990s and had superior access to German sources.

Morale had plummeted. Troops deserted or surrendered wholesale. Some formations had to be coaxed or bribed into an attack; others mutinied outright and killed their officers. Looting was a common occasion for rioting and violence against officers. Many new recruits jumped from troop trains. Few assault units remained, and those that did were rarely in shape to deliver the Gegenstoss. The infantry now consisted of static positional divisions and the tattered remnants of the assault divisions. The only dependable formations were the artillery and the machine gunners, who remained highly motivated and were able to conduct a mutually supporting defense in depth.†

† It was widely reported in the AEF that German machine gunners had been found chained to their weapons to prevent them from fleeing. Americans in Europe, a German newspaper published in English for American POWs, excoriated Stars and Stripes magazine for publishing what it called “that four-year-old English lie,” pointing out that cowardly soldiers do no better for having been chained to their guns than otherwise. An editor of Stars and Stripes later wrote, “I have forgotten (and neither tigers nor constitutional amendments would drag it from me if I had not) who was guilty of the chained-to-the-guns charge. Certainly we should have known better. The heavy German machine gun was equipped with chains so that its crew could haul it from position to position with readiest convenience, and some adroit intelligence [officer] in the early days of the war, perhaps discovering a dead crew with the chains in their hands, had jumped to the chained-to-the-guns conclusion.” (John T. Winterich, ed., Squads Write! A Selection of the Best Things in Prose, Verse and Cartoon from the Stars and Stripes (New York and London: Harper & Brothers, 1931), p.79.)

Nevertheless, the Germans were confident enough in the defenses provided by the terrain and their own preparations, and (as we shall see) convinced enough that the Meuse–Argonne would remain a relatively quiet sector, to entrust the line to only four weak divisions with several more in reserve. The commander of this part of the front was General Max von Gallwitz, a veteran of the Franco-Prussian War and one of the few high-ranking German officers whose appointment resulted solely from his ability rather than from noble descent. An artilleryman, he had served on the German General Staff and had commanded the troops on the western flank of the Verdun offensive, including the bloody conquests of Hill 304 and the Mort Homme. For much of 1916 he was assigned to command the Second Army on the Somme; but now he was back to face the Americans on familiar ground. Army Group Gallwitz occupied the front from Grandpré in the Argonne Forest past Verdun to Metz, a distance of about 75 miles. It was the southernmost of Gallwitz’s two armies, Composite Army C, that had failed to hold the line at St Mihiel, so he had seen the Americans fight. The other army, the Fifth, held the northern part of Gallwitz’s sector, from the Argonne Forest to the Heights of the Meuse, almost exactly the sector designated by Pershing for his Operation B.

The four divisions that Gallwitz ordered into the line were a mixed bag.* The right-most position, in the Argonne Forest, was held by the 2nd Landwehr Division, a static unit that had been in Lorraine sector more or less continuously since September 1914.† To bring Ludendorff’s assault divisions up to strength for the Spring Offensives, it had been stripped of most of its young men early in 1918. To its left, covering the valley of the Aire, the hill of Vauquois, and the Bois de Cheppy, was the 1st Guards Division. This was an elite shock formation, trained in open warfare, that had fought well in several of the spring battles both on the offensive and in retreat. By early September, however, it had been badly depleted. Gallwitz designated these two divisions the “Argonne Group.”

* Unless noted otherwise, information on German divisions is from US War Office, Histories of Two Hundred and Fifty-One Divisions of the German Army Which Participated in the War (1914–1918) (London: London Stamp Exchange, Ltd., 1920, republished 1989).

† American intelligence reports put this division further east, in the Bois de Cheppy and the Bois de Malancourt. I rely instead on the German situation map for September 26 that is reproduced in Rexmond C. Cochrane, “The 79th Division at Montfaucon,” in U.S. Army Chemical Corps Historical Studies: Gas Warfare in World War I (Washington, DC: US Army Chemical Corps Historical Office, 1960), p. 10.

Next in line to the east, covering the Montfaucon sector from the Bois de Malancourt to the ruined village of Haucourt (and almost exactly opposite the US 37th and 79th Divisions), was the German 117th Division. In May it had participated in Ludendorff’s attack at the Lys River and taken heavy losses. One of its regiments was disbanded and the 11th Grenadier Regiment, recently arrived from Macedonia, was assigned to replace it. By August, the 117th had been assigned to Crown Prince Rupprecht’s army group opposite Amiens. Just before Haig’s August 8 counterattack in that sector, the German army commander rated it as one of only two divisions out of 13 available to him that was fully fit for combat.¹² But it had been in position less than a day when Haig struck; taken by surprise, it collapsed like the rest of the German line. So badly was it hit that, once it had been withdrawn to rest and refit, a second of its regiments was dissolved, its surviving men sent to fill out the rosters of other divisions. The 450th Regiment, itself a battered survivor of the same British attack, was brought in to bring the division up to something resembling full strength. But “full strength” was a dubious concept; by September 26, the companies of the 117th were down to between 55 and 80 men, with only one as high as 100.¹³ Furthermore, with two of its three regiments new to the division, it was no longer a coordinated fighting organization.

Last in line in the Meuse–Argonne sector, holding the position from Malancourt clear to the Meuse, was the 7th Reserve Division. A static unit, the division had suffered heavily at Verdun in 1916. More recently, it had fought in several of Ludendorff’s spring battles and in August defended against the French at Soissons; in all of these engagements it took many casualties. Having been in more or less continuous combat since March, it was withdrawn for training between September 8 and 17, then put into the left of Gallwitz’s line. By this time its companies had fewer than 100 men. The 117th and 7th Reserve constituted Gallwitz’s “Maas Group West.” All told, the four German divisions between the Argonne Forest and the Meuse River could each muster about 4,300 men.*

* The normal complement of a German infantry regiment was about 3,200 officers and men. (General Staff [UK], German Army Handbook, April 1918 (London, Melbourne: Arms and Armour Press, 1977; reprint of 1918 edition).) A division had three infantry regiments and one artillery regiment as well as a heavy artillery battalion, a signal unit, pioneers, and trains, for a rough total of 13,000. By September 1918, however, most divisions were at one-third strength. (Trask, AEF and Coalition Warmaking, p. 120.) The divisions that faced the AEF therefore held on average about 4,300 men of all ranks.

The four front-line divisions were not the only ones Gallwitz had available. Four more were located within a few hours’ march of the front and another eight within a day or two. Of the first group, two are worth examining because they played an important role in the events around Montfaucon on September 26–30. The 37th Division (not to be confused with the American division of the same number) was a first-class assault unit, trained in infiltration tactics and open warfare. It did very well in Ludendorff’s spring battles at the Somme, the Aisne, and the Marne. Like all German divisions, however, it had lost many men in the course of 1918, and was badly depleted despite receiving large numbers of replacements. From August 13 until September 20 it was in Gallwitz’s line south of Montfaucon, but was relieved by the 117th and moved back to a rest camp at Longuyon, about 25 miles northeast of Verdun. A mere five days later it was trucked to Dun on the Meuse River, assigned to army reserve, and told to be ready for action.¹⁴

The other division close to the Montfaucon sector was the 5th Bavarian Reserve Division, not an elite formation but one that had the great advantage of being fresh, having seen little heavy fighting so far in 1918. Nevertheless, it averaged only 60 men per company as the Meuse–Argonne campaign began. As they marched west to reinforce Fifth Army, they crossed the Meuse at Brieulles. There they were caught by the opening American bombardment and badly shot up; their artillery and support trains were heavily damaged and regiments, battalions, and companies were scattered over the countryside.¹⁵

The Germans were not simply negligent when they assigned only a few weak divisions to hold the front line in the Meuse–Argonne. Their troop dispositions were based on the knowledge that they could not mount a strong defense everywhere, and on their best guess as to where the next blow would fall. By August 9 Ludendorff knew that the Central Powers could not win the war; his job now was to keep from losing it. A month earlier, Foch’s attack at Soissons, which had rolled up all the German gains since May and removed the threat to Paris, had come as a shock to OHL and to the government; never had the Army been forced into such a precipitous retreat. But Ludendorff, increasingly divorced from reality, refused to admit that the Army was morally beaten, blaming the setbacks on his commanders, his staff, and the poor quality of the reserve troops. Then, on August 8, Haig attacked at Amiens. The British tactics called for unprecedented coordination and mutual support among the infantry, artillery, tanks, and aircraft. They worked, as we have seen in Chapter 6; and the British advanced 12 miles before the assault was halted four days later. The Germans lost 228,000 men, half of them deserters. Ludendorff later called this the “black day of the German army.”¹⁶ At a meeting of the Crown Council at Spa on August 14, he joined the Austro-Hungarian Kaiser Charles in recommending immediate peace negotiations. Kaiser Wilhelm authorized Admiral Hintze, the Minister of Foreign Affairs, to begin negotiations through a neutral country; but before anything could come of this, Ludendorff calmed down and the initiative sputtered out.

The next blow came on September 12, when the Americans attacked the St Mihiel salient south of Verdun. As we saw in Chapter 7, the Germans were caught in the middle of a withdrawal; within two days almost all the salient was in American hands. The rapid overrunning of the salient accomplished two strategic objectives for the Americans, which they did not realize at the time. It further demoralized the German High Command, already rocked by the Allied successes at Soissons and Amiens. Gallwitz and Ludendorff were particularly disturbed that no meaningful resistance had been mounted and that large amounts of supplies were destroyed or captured. Ludendorff showed signs of coming unhinged: Admiral von Müller heard from another officer that the news of St Mihiel had rendered him “a completely broken man.”¹⁷ Another senior officer who saw Ludendorff on the first night of the attack considered him “so overcome by the events of the day as to be unable to carry on a clear and comprehensive discussion.”¹⁸ Hindenburg telegraphed Gallwitz, “The severe defeat of Composite Army C on September 12 has rendered the situation of the Groups of Armies critical.” Blaming the outcome on “faulty leadership,” he continued, “I can only hope that the Group of Armies employing the forces which I am allotting to it will hold the position. The Group of Armies will bear the complete responsibility for this ... Wherever commanders and troops have been determined to hold their position and the artillery has been well organized, even weak German divisions have repulsed the mass attacks of American divisions and inflicted especially heavy losses on the enemy.”¹⁹

On September 14 the Austro-Hungarian Kaiser Charles, following up his proposal (seconded by Ludendorff) of August 14, made a public appeal for the contending governments to “send delegates to a confidential and non-binding discussion on the basic principles for the conclusion of peace,” to be held “in a neutral country and at an early date.”²⁰ Over the next few days the offer was rejected in turn by the British, the Americans, and the French, who believed it was a tactic to divide the Allies. Puzzled German soldiers wondered why their side, which they had been told was doing so well, seemed to be offering an armistice; their officers explained that “it was not weakness but the desire to communicate that led to the offer. Every front-line soldier knows that an honorable peace can only then be achieved, when the Entente’s desire for destruction is confronted by our unshakeable determination to defend ourselves.”²¹

The second effect of the fall of the St Mihiel salient was to cement in the minds of the German High Command the idea that the Americans’ true objective, whatever appearances might indicate, was the fortress of Metz. At first, the Germans were startled that the Americans did not continue their successful attack in that direction. On September 16, Gallwitz assessed the situation thus: “The enemy’s delay is undoubtedly due to the fact that he has reconnoitered our outpost area, wishes to work his way forward over this terrain first and move his artillery up correspondingly. In addition, he must reconnoiter the possibilities of moving his numerous tanks before resuming his attack.”²² By that time, Pershing had already begun withdrawing his divisions and sending them on a 60-mile journey northwest to the Meuse–Argonne. Five days later Gallwitz’s intelligence staff wrote, “[T]he attack will be directed principally against Composite Army C, and will probably be extended to the east bank of the Moselle.” Composite Army C was, as we saw, the army now holding the line behind the former St Mihiel salient. Similar assessments emanated from various German headquarters every few days. On September 27, as the Meuse–Argonne offensive was in its second day, the intelligence office of the German Crown Prince was writing, “[T]he possibility of an imminent, large scale, Franco-American attack in Lorraine has not been decreased by the Franco-American offensives which opened yesterday in eastern Champagne and between the Argonne and the Meuse ... The front between Verdun and Lunéville [i.e., opposite Metz] is still occupied by large numbers of fresh, trained troops.”²³ As late as October 18 Gallwitz’s intelligence section was warning, “The probable entrance into line of Americans in place of the French 69th Inf. Div. may have some connection with preparations for a large-scale American attack planned against Metz.”²⁴

The reduction of St Mihiel had a third, somewhat paradoxical, consequence. The Germans, although defeated, concluded that the Americans weren’t very good fighters. An intelligence officer in Composite Army C headquarters wrote:

Great clumsiness was shown in the movement over the terrain of the waves of riflemen which followed each other closely. The shock troops hesitated when met by the least resistance, and gave the impression of awkwardness and helplessness. Neither officers nor men knew how to make use of the terrain. When met by resistance, they did not look for cover but went back erect ... The command throughout was bad and clumsy. The enemy had obviously very many officers, but these officers lack all qualities of leadership. Unmistakable was the perplexity after the initial objective was reached. The enemy was helpless when confronted by a new situation and unable to exploit the success ... The entire lack of military skill was also evident in the pursuit. No advantage was taken of favorable opportunities for attacks on flanks and envelopments…

…

Concluding Estimate: The American is too much of a dilettante, and therefore also in a major attack needs not to be feared. Until now, our men had a much higher opinion of Americans, due to the fact that in patrol undertakings they had shown themselves as dashing soldiers ... Our troops had expected much more of them in a major battle. In spite of some local reverses, their confidence of being able to deal with the Americans has been raised.²⁵

Most of these criticisms were accurate, and were known to the Americans themselves. The flaw in the assessment lay in what it omitted: all of the shortcomings noted were to be expected in a newly formed army in its first battle; intelligent, highly motivated soldiers (which even the Germans knew the Americans to be) would quickly learn their jobs.

For all of these reasons—lack of clear thinking at OHL, fixation on Metz as the target, and disdain for the offensive abilities of the AEF—the Germans failed to credit that a massive assault on the Meuse–Argonne sector was imminent. They were aware of the American buildup there, but they did not see it as the main thrust. The Americans had decreed strict measures to disguise their concentration: no movement by day, no fires, and so forth. But they could not hide half a million men and their equipment completely. The German divisions holding the line knew that something was up, but not what it might be. On September 23, a Summary of Intelligence by a German division (most likely the 1st Guards) noted the presence of new American units near the front and concluded:

All these identifications, particularly the circulation in hitherto quiet sectors, point to the possibility of an attack along the whole front from Reims to Verdun. During the day-time, only circulation far in the rear could be observed, but at night great activity reigned along our front. The noise of narrow gauge railways, motor trucks, the unloading of heavy material, loud cries, sirens and klaxons could be heard throughout the whole night.²⁶

On the same day the 1st Guards Division’s operations report stated, “Brown uniforms having been observed along our front on Sept. 23, it would appear, although not definitely, that the presence of American troops must be taken into consideration ... On the left of the neighboring group, in sectors previously held by the French, the presence of the 79th and 80th American infantry divisions has been confirmed.” The same report noted that interrogation of a French prisoner revealed that “A vigorous attack for the purpose of creating a diversion from the American offensive in front of Verdun can be expected.”²⁷ In other words, an attack was possible, but only as a diversion from a larger assault further east.

Only on September 25 did Gallwitz realize that the threat did not lie behind the old St Mihiel battlefield, it lay between the Argonne and the Meuse.²⁸ To meet it, he did what he could. The closest division to the front was the 5th Bavarian Reserve, camped at Fontaine only a few miles east of the Meuse; early on the morning of the 25th he ordered it to concentrate near Dun, preparatory to crossing the river westward. He also assigned to it a complement of aircraft for reconnaissance. Later in the day, as the American buildup continued, he released that division and the 602nd Field Artillery Regiment to Maas Group West, to be available for counterattacks. (The 5th Bavarians, as we have seen, were cut up by American shell fire at this time while crossing the Meuse to get to their assigned position.) Later in the day, Gallwitz ordered the 37th to move south by truck from its rest camp at Longuyon, up near the Luxembourg border, to Dun. To further bolster his line, he alerted the 28th Division, quartered six miles east of the river near Étain, to assemble and be ready to move.²⁹ Late in the evening Gallwitz had his corps commanders make their final dispositions.

At 2:00 a.m. on the 26th Gallwitz finally received the evidence he needed to confirm the imminence of a major attack. A raid yielded a prisoner from the American 4th Division, which he had previously seen at St Mihiel. The transfer of this veteran unit to west of the Meuse convinced Gallwitz that here was where the main American blow would fall. He did not have to wait long for further proof. At 2:30 a.m., as Pershing had ordered, all 2,700 American guns opened up at once, saturating the German defensive zone with shells. Locations that might harbor German artillery or observers were particular targets: the Heights of the Meuse, the eastern edge of the Argonne Forest, any woods that offered concealment, Montfaucon itself. Bivouac areas, headquarters, crossroads, and telephone exchanges received special attention.³⁰

The concentration of fire against the Germans’ strong points and communication centers left their front line largely untouched—for the moment. The 66th Infantry Regiment, the one that had borrowed Major Stosch’s machine guns, sent its patrol out into the lines of the American 4th Division. It returned by 4:30 a.m. in a disorganized state.³¹ The major would not get his guns back.

At exactly 5:30 a.m. American artillery, still following Pershing’s plan, concentrated all of its fire on the German front line. In the words of an officer of the 66th Infantry, “All communications to the front line are immediately cut—telephone wires shot to pieces, too badly to be repaired by the crews that are sent out. Signal lights no longer work because of the thick fog and the smoke and dust of the shelling. Runners are sent forward but not heard from again. Brigade HQ is unreachable because it has moved, and the new location is unknown.”³² Next door, in the 11th Grenadiers, “The detonations followed each other without interruption and the air was torn by a constant droning sound. In the villages, the remaining ruins tumbled down; in the forested areas, the chaos of trees and roots became even more confused, and even the laboriously erected defenses could not hold out for long. A thick, dark cloud lay over the entire region, while a mix of dust and smoke, artificial and natural fog, and gas reduced visibility to 1 meter.”³³

Gradually the intensity of the bombardment on the front lines lessened as the rolling barrage moved northward. German machine gunners who had taken cover hastened back to their weapons and peered through the smoke and fog, trying to see movement. Shadows began to appear in the murk. Back through the woods and up the slopes sped the cry, “Sie kommen! Sie kommen!”³⁴