The End of the Panzer Divisions I

During the autumn of 1944, in the aftermath of the failed attempt on Hitler’s life on July 20, and in the aftermath of the Red Army’s colossal breakthroughs in the East, the Nazi regime and the German people mobilized their last reserves of ferocity and fanaticism. The propaganda vision of a people’s community at arms and the free rein given to violence on both foreign and home fronts enhanced a pattern of exploitation and dehumanization already permeating German society from the factories to the countryside. Rationality gave way to passion and to fear as retribution loomed for a continent’s worth of crimes.

The Wehrmacht went out fighting and it went down hard. Like the German people, it neither saw nor sought an alternative. The prospective fate implied in the Allies’ demand for unconditional surrender could assume terrifying form to men who had seen—and participated in—the things done “in the name of the Third Reich and the German people.” That meant reconstructing shattered divisions by placing officers at road junctions and impressing every man without a clear destination, even if cooks became tankers and sailors found themselves in the Waffen SS. It meant filling out ranks with teenage draftees and men combed out of the increasingly moribund navy and air force. It meant reequipment by an industrial system that continued to defy the best efforts of the Combined Bomber Offensive. It meant morale enforced by laws making a soldier’s family liable for any derelictions of duty. It meant field courts-martial that seemed to impose only one sentence: death.

Combine Eisenhower’s commitment to a continuous front with the relative weakness of Allied ground forces, and weak spots must inevitably emerge. The most obvious one was in the American sector: the Ardennes Forest, a static sector manned by a mix of green divisions and veteran outfits that had been burned out elsewhere. Hitler’s intention, shared and underwritten by High Command West, was to replicate the success of 1940 by striking through the Ardennes for Antwerp. The port’s capture would both create a logistical crisis for the Allies and divide the British from the Americans, opening the way to their defeat in detail and—just possibly—to a decisive falling-out between partners whose squabbling, egalitarian relationship was never really understood by German strategic planners who believed in client systems rather than alliances.

That the Allies still had absolute control of the air over the front, and that German fuel supplies were about enough to get their tanks halfway to Antwerp, did not concern the Führer. Nor were his generals excessively disturbed. The planners of High Command West preferred in principle a more limited operation: a double envelopment aimed at Liege. They were, however, never able to convince even themselves why Germany’s last reserves should be used that way. What was to be gained, except a drawn-out endgame?

At least the West was geographically small enough to offer something like a legitimate strategic objective. The Eastern Front presented only the prospect of a second Kursk, with the last of the panzers feeding themselves into a Russian meat grinder somewhere east of the existing front line. Panzer Lehr’s Fritz Bayerlein echoed many of his counterparts when he said he persuaded himself that the attack would succeed in order to give his orders credibility and sustain the aggressive spirit of his subordinates. If Operation Watch on the Rhine proved a Twilight of the Gods, then it would be a virtuoso performance as far as the army’s professionals and the zealots of the SS could make it.

By mid-December a buildup overlooked or discounted by confident Allied commanders gave the Germans a three-to-one advantage in men and a two-to-one advantage in armored vehicles in their chosen sector of attack. A new 6th SS Panzer Army had been organized in September under Sepp Dietrich. By this time in the Western theater the distinctions and antagonisms between army and Waffen SS had diminished, especially in the panzer formations, where the consistently desperate situation and the relatively even numbers of divisions made close mutual support a necessary norm. In the projected offensive, 5th Panzer and 6th SS Panzer Armies would fight side by side with few questions asked.

Part of the army panzers’ reconstruction involved reorganization. Both in Russia and in the West, the events of 1944 had resulted in serious losses of trained specialists and no less serious discrepancies between the numbers actually available in the combat units and those in the divisions’ rear echelons. One response was pairing panzer divisions by twos in permanent corps that would assume service and training responsibilities. Five were organized and saw action, against the Russians in the final campaign. More significant was the introduction on August 11 of the Panzer Division Type 1944. This gave each panzer grenadier regiment an organic pioneer company and each tank battalion organic maintenance and supply companies. Both changes acknowledged the decentralization that had become the panzers’ tactical and operational norm. Battalions consistently shifting rapidly from place to place and battle group to battle group would now be more self-sufficient. Divisions would now be better able to concentrate on planning and fighting—at least in principle.

The new panzer divisions were still authorized two tank battalions, each of as many as 88 tanks. Paper may be infinitely patient; reality is less forgiving. In the autumn of 1944, Allied heavy bomber strikes repeatedly hit most of the big tank manufacturing complexes: Daimler-Benz, MAN in Nuremberg, and the Henschel Tiger II plant in Kassel. Speer was able to sustain production, but only around half of the 700 Panthers and Panzer IVs scheduled for delivery in December reached the intended users.

The shortages also reflected decisions made in the Armaments Ministry. Speer had kept up tank production by transferring resources from the manufacture of other vehicles and by cutting back on spare parts. The latter dropped from over a quarter of tank-related contracts in 1943 to less than 10 percent in December 1944. Critical resources, like the molybdenum that made armor tough as opposed to brittle, were in critically short supply. Quality control slipped badly in everything from optics to transmissions to welding. The continued willingness of Germans to report for work despite the bombing is often cited. The on-the-job efficiency of men and women deprived of everything from their homes to a night’s sleep has been less investigated.

The increasing use of slave labor in war plants had consequences as well. Distracted, tired foremen and overseers were easier to evade. Risks that seemed foolhardy in 1943 took on a different dimension as the Reich seemed on the edge of implosion. Deliberate sabotage was probably less significant than hostile carelessness. But increasing numbers of panzers were coming on line with screws poorly tightened, hoses poorly connected—and an occasional handful of shop grit or steel filings deposited where it might do some damage. That was no small matter in contexts of frequently inexperienced crews and frequently nonexistent maintenance vehicles.

The immediate response was to reduce the number of tanks in a company to 14, and where necessary to replace those with assault guns of varying types. Even with these makeshifts, 15 panzer divisions still had only one tank battalion. Sometimes an independent battalion would be attached—Leibstandarte, for example, benefited by receiving the Tigers of the 501st SS as its de facto second battalion. Other divisions found themselves with new battalions equipped with Jagdpanthers or Jagdpanzer IVs, trained for antitank missions rather than tank tactics, or in the close cooperation with panzer grenadiers that remained the assault guns’ mission in an offensive.

Training and equipment were general problems in divisions preparing for the Ardennes offensive. Panzer Lehr, the army’s show horse, had its full complement of men, a third tank battalion equipped with assault guns, and one of the supplementary heavy antitank battalions. Das Reich, however, reported a large number of inexperienced recruits, and reported individual and unit training as at low standards. Leibstandarte described morale as excellent, but combat readiness above company level as inadequate. The 116th Panzer Division was short of armor, motor vehicles, and junior officers and NCOs. Second Panzer Division lacked a third of its vehicles: on December 14, one panzer grenadier battalion was riding bicycles. It was all a far cry from the spring of 1940.

In its final form, Watch on the Rhine6 incorporated three armies deployed on a 100-mile front under Model, commanding Army Group B since Rundstedt had been restored, at least nominally, to his former position in September. The balance of forces at the cutting edge, and their missions, demonstrated the army’s decline relative to the Waffen SS. Dietrich’s 6th SS Panzer was the spearhead, with Leibstandarte, Das Reich, Hohenstaufen, and Hitler Jugend as its backbone, and five army infantry divisions as spear-carriers and mop-up troops. Fifth Panzer Army would cover Dietrich’s left, and Manteuffel had the army’s armored contribution: Panzer Lehr, 2nd and 116th Panzer Divisions, plus four infantry divisions. Protecting his left flank in turn was the responsibility of 7th Army, with four infantry divisions and no armor to speak of.

Watch on the Rhine’s order of battle incorporated 200,000 men, 600 armored vehicles, almost 2,500 supporting aircraft—that number itself a triumph of concentration involving stripping the Reich’s air defenses. Radio silence was draconically enforced. Camouflage was up to Eastern Front standards. Parachute drops and sabotage units were expected to confuse surprised defenders even further. The offensive seemed structured to maximize what the Germans—the panzer troops in particular—considered their main strength: sophisticated tactical and operational expertise.

Model could in principle call on another ten divisions, but only two were panzers; the offensive would rise or fall with its starting lineup. The operational plan was Sichelschnitt recycled. Dietrich, at the Schwerpunkt, was to break through around Monschau, cross the Meuse around Liege, and strike full tilt for Antwerp. Manteuffel would cross the Meuse at Dinant and aim for Brussels. The panzers were expected to be across the Meuse before the Allies could move armor sufficient to counter them.

As so often before, however, German focus devolved into tunnel vision. None of the specific plans addressed the subject of Allied air power. The responsible parties similarly avoided addressing directly the fuel question. By comparison to the Western campaign’s early months, fuel supplies were impressive, but the Panthers and Tigers were always thirsty. Were the Americans likely to be so confused, so feckless, and so obliging as to leave their fuel dumps intact as refilling points? In the climate of December 1944, asking such a question suggested dangerous weakness of will and character.

Sepp Dietrich might be an unrefined, unimaginative, hard-core Nazi, but he did not lack common sense. All the Waffen SS had to do, he later said sarcastically, was “cross a river, capture Brussels, and then go on to take Antwerp . . . through the Ardennes when the snow is waist-deep and there isn’t room to deploy four tanks abreast let alone armored divisions. When it doesn’t get light until eight and it’s dark again at four—and at Christmas.” Sixth SS Panzer Army had four days to reach the Meuse. Dietrich and his staff set the infantry divisions to breach US defenses on day one, December 16. When the hastily reconstituted army units faltered in the face of determined American resistance, the word was “panzers forward.” But Kharkov and Kursk were a long way back. The Waffen SS had made its recent reputation in defensive fighting. Experience in offensive operations had been diluted by expansion. Officer casualties had been heavy. From battalion to division, 6th SS Panzer Army correspondingly eschewed finesse in favor of head-down frontal attacks.

Tactical maneuver was further restricted by rain periodically freezing into snow as temperatures hovered in the low thirties. Fields already saturated by the heavy rains of early autumn turned into glutinous paste when the heavy German tanks tried to cross them. The alternative was straight down the roads and straight down the middle from village to village. Each attack was expected to put the finishing touches on an enemy that seemed on the ragged edge of breaking. Yet the “Amis” held on—and without the fighter-bombers grounded by the same weather that slowed the panzers.

For the sake of speed the SS neither used their reconnaissance battalions to probe for weak spots, nor their pioneers to assist the tanks. The tanks repeatedly pushed ahead and just as repeatedly lost contact with their infantry, only to run afoul of ambushed Shermans and M-10s, or bazooka teams taking advantage of relatively weak side armor. Even the 57mm popguns of the infantry’s antitank units scored a few kills. The panzer grenadiers, many of them half-trained recruits or converted sailors and airmen, were at a surprising disadvantage against American regiments, some of which had been in action since Normandy.

The 12th SS Panzer Division, on the German right, lost most of its Panthers in the first two days and made no significant progress thereafter. Its neighbor, Leibstandarte, similarly held up, responded by sending forward an armor- heavy battle group. It included most of the division’s striking power: a battalion each of Panthers, Panzer IVs, and Tiger Bs: together around 100 tanks, a mechanized panzer grenadier battalion, pioneers, and some self-propelled artillery. Its commander was Lieutenant-Colonel Jochen Peiper. He had been Himmler’s adjutant from 1938 to 1941 and had enjoyed the Reichsführer’s mentoring and patronage. He had developed into a hard-driving, charismatic risk-taker whose men followed him in good part because of his reputation as someone who led from the front and was the hardest man in any unit he led. Peiper was, in other words, an archetype of the kind of officer the Waffen SS nurtured in Russia and now turned loose in the West.

Peiper was also just the man to throw the dice for an entire panzer army. His mission was to reach the Meuse, capture the bridges before they were demolished, and hold until relieved. It was 100 miles over back-country roads that were little more than trails. Orders were to avoid combat when possible, and to tolerate no delays. The battle group’s movements were a model tactical exercise—at first. The panzers bypassed confused American rear-echelon troops, slipped between elements of American convoys, and overran American supply dumps. With his fuel low, Peiper refilled his tanks with 50,000 gallons of US gas captured without firing a shot.

The panzers captured a key bridge at Stavelot on December 18, and pushed through to the Amblève River. All they had to do was cross, and the way to the Meuse would be open. But an American engineer battalion blew the bridges Peiper was expecting to rush—in one case literally under the gun of a Tiger VIB. American tanks and infantry, moving faster than expected, retook Stavelot and cut the panzers’ line of communication. The sky was clearing, and the fighter-bombers returned to hammer Peiper’s columns relentlessly. The battle group was down to three dozen tanks, as much from breakdowns as from combat loss. Its infantry, exposed to the weather day and night in their open-topped half-tracks, on cold rations and broken sleep, were numb with cold and fatigue. Peiper requested permission to withdraw. It was refused. Relief attempts were stopped almost in their tracks. As the Americans closed in, Peiper made his stand at a village called La Gleize. After two breakout efforts failed, with his tanks out of fuel and his ammunition exhausted, on Christmas he led out on foot the men he had left. Moving by night, 800 survivors of the 6,000 who began the strike a week earlier made it back to Leibstandarte’s forward positions.

Peiper left behind about 100 of his own wounded and another 150 American POWs. The senior US officer later reported that the Germans had appropriately observed the rules of war. But from the operation’s beginning, Kampfgruppe Peiper and the rest of Leibstandarte had left a trail of bodies in its wake: as many as 350 Americans and well over 100 Belgian civilians. The consequences were epitomized by the GIs bringing in some prisoners from Peiper’s battle group who asked an officer if he wanted to bother with them. He said yes. Not everyone did. For the rest of the war it was not exactly open season on Waffen SS prisoners, but they surrendered at a higher degree of immediate risk than their army counterparts.

Pieper was not necessarily a liar or a hypocrite when he not only insisted at La Gleize that he did not shoot prisoners, but seemed surprised by the allegation. He is best understood as resolving a specific form of the cognitive dissonance that increasingly possessed the Wehrmacht in particular and the Reich as a whole. The question of whether someone, Peiper or a superior, somehow either gave orders to take no prisoners or made it clear that “no delays” was a euphemism for “no prisoners” is misleading. Since Normandy, a pattern had developed in which both sides processed refusing quarter, shooting prisoners, and similar frontline atrocities, as mistakes, misunderstandings, or part of “the filth of war”: fear, frustration, vengeance, the semi-erotic thrill of having an enemy completely at one’s mercy.

Inexperienced troops are more prone to be trigger-happy, and there was ample inexperience on both sides of the line in June 1944. Even a thoroughly ideologized German was likely to see a difference between more or less Aryan “Anglo-Saxons” and despised, despicable Slavs. Nor was there much to gain by making things worse than they had to be. Within the same few days in Normandy, elements of the Hitler Jugend murdered Canadian prisoners in cold blood, and other troops of the division negotiated a local truce with a British battalion enabling both sides to bring their wounded to safety. Such agreements were not everyday occurrences, but they did happen. An officer of 9th Panzer Division describes one of his men bringing a wounded American back to his own lines and returning laden with chocolate and cigarettes as tokens of appreciation. A story improved in the telling? Perhaps. But nothing similar was plausible even as a rumor in the East. And one Russian Front was bad enough.

What did transplant from the East was a frontline culture that since 1941 had developed into something combining convenience and indifference, embedded in a matrix of hardness. Hardness was neither cruelty nor fanaticism. It is best understood in emotional and moral contexts, as will focused by intelligence for the purpose of accomplishing a mission. It was—and is—a mind-set particularly enabling the brutal expediency that is an enduring aspect of war.

In commenting on Kurt Meyer’s trial and death sentence, a Canadian general asserted he did not know of a single general or colonel on the Allied side who had not said “this time we don’t want any prisoners.” In fact, there is a generally understood distinction, fine but significant, between not taking prisoners and killing them once they have surrendered. Recent general-audience works on the Canadians and Australians in World War I, for example, are remarkably open in acknowledging relatively frequent orders at battalion and company levels of “no prisoners” before an attack. Shooting or bayoneting unarmed men is another matter entirely. It might be called the difference between war and meanness.

James Weingartner highlights the discrepancy between the US Army’s judging of war crimes by Americans and its response to comparable offenses involving Germans. That was not a simple double standard. For the Americans, as for the British and Canadians, expedience and necessity remained situational rather than normative, on the margin of legal and moral systems but not beyond them. On the German side of the line, hardness transmuted expediency into a norm and redefined it as a virtue. Impersonalization and depersonalization went hand in hand. Interfering civilians or inconvenient POWs might not be condignly and routinely disposed of—but they could be, with fewer and fewer questions asked externally or internally. The French government was shocked and embarrassed to find Alsatians represented among the perpetrators of Oradour. Defended in their home province as “forced volunteers,” they were tried and convicted, but pardoned by Charles de Gaulle in 1953 for the sake of national unity.

The inability of the Waffen SS to break through on the north shoulder removed any possibility of success Watch on the Rhine might have had. Instead of exploiting victory, Das Reich and Hohenstaufen found themselves stymied by roads blocked for miles by abandoned vehicles out of fuel or broken down. While the SS ran in place, however, Manteuffel had used his infantry skillfully to infiltrate, surround, and capture most of the green 106th Infantry Division’s two forward regiments before sending his armor forward. The 116th Panzer made for Houffalize. Second Panzer and Panzer Lehr pushed through and over the 28th Infantry Division toward Bastogne, destroying an American armored combat command in the process.

The 101st Airborne Division got there first, dug in, and has been celebrated in story, if not song, ever since. The Germans originally hoped to take the town by a coup de main. When that proved impossible, Bayerlein argued that Bastogne was too important as a transportation center to be bypassed. Manteuffel was already concerned that his forward elements were too weak to sustain their progress; attacking Bastogne in force would only make that situation worse. He was also too old a panzer hand to risk tanks in a house-to-house fight against good troops. The Panzer Baron had begun his career in the horse cavalry, understood the importance of time for Watch on the Rhine, and decided to mask the town and continue his drive toward the Meuse.

That the choice had to be made highlighted the growing difficulty the panzers faced in being all things in every situation. In 1940, motorized divisions had been available for this kind of secondary collateral mission. In 1941, the marching infantry could be counted on to come up in time to free the panzers for their next spring forward. In 1944, 15th Panzer Grenadier Division did not arrive at Bastogne from army group reserve until December 24.

Fifth Panzer Army benefited from a cold front that set in on the night of December 22, freezing the ground enough for the Panthers to move cross-country. Soft ground, however, was the least of Manteuffel’s problems. His spearheads knifed 60 miles deep into the American positions, along a 30-mile front. Second Panzer Division, generally regarded by the Americans as the best they faced, got to within five miles of the Meuse on December 24—ironically near Dinant, where 7th Panzer had staged its epic crossing in 1940. But its fuel was almost exhausted. Model responded by ordering the division to advance on foot. Brigadier General Meinrad von Lauchert had been commanding the division only since December 13. He had been a panzer officer since 1935, led everything from a company to a regiment in combat, and recognized bombast. But he was not a sorcerer, and could not conjure fuel where none existed. Second Panzer was at the far end of its operational tether.

By platoons and companies, Americans fought bitter defensive actions throughout the sector—in one case holding out in a castle. Thirty-two men would be awarded the Medal of Honor from first to last during the Battle of the Bulge, and determination increased as word spread that the Germans were not taking prisoners. To the north, what remained of the 106th, a regiment of the 28th, and combat commands of 7th and 9th Armored Divisions, held another key road junction—St. Vith—for five vital days against first infantry, then the elite panzers of the Führer Escort Brigade from Model’s reserve. Not until 2nd SS Panzer Division advanced far enough to threaten the town from the north did the hard-hammered garrison withdraw.

In the process of working their way forward, elements of Dietrich’s panzers seeking to evade the clogged roads in their sector began edging onto 5th Panzer Army’s supply routes. Manteuffel ordered them kept off; the corps commander responded by establishing roadblocks whose men were authorized to use force to regulate traffic. There are no records of shots being fired, but army and SS columns remained entangled as tempers flared and cooperation eroded.

Model released some of his characteristic nervous energy by briefly directing traffic himself, while reassuring Hitler that the chances of victory remained great. But the overall supply situation was rapidly deteriorating. The clearing skies accompanying the cold front meant the return of allied planes en masse: an average of 3,000 sorties a day, disrupting operations and turning the movement of troops and vehicles to nighttime—including the vital fuel trucks.

Model had from the beginning recommended eschewing a drive to the Meuse in favor of a quick turn north to isolate and then encircle the dozen or so American divisions concentrated around Aachen. Dietrich’s staff had been clandestinely working on a similar backup plan since December 8. Manteuffel underwrote their thinking on December 24, when he phoned the High Command and declared Antwerp was beyond his reach.

Any lingering optimism was dispelled on Christmas when 2nd Panzer was attacked by its literal counterpart, the 2nd US Armored Division. The Americans encircled the panzers’ leading battle group, destroying it as artillery and RAF Typhoons frustrated relief efforts by the rest of the division supported by elements of the 9th Panzer, newly arrived from High Command reserve. Six hundred men escaped—walking and carrying no more than their personal weapons. It was getting to be a habit for the panzers. Two thousand more were dead or prisoners. Over 80 AFVs remained on the field, knocked out or with empty fuel tanks. The rest of the division fought on around the village of Humain, so fiercely that it took one of the new flame-throwing Shermans to burn out the last die-hards. On December 27, 2nd Panzer was withdrawn. On January 1, 1945, it reported exactly five serviceable tanks.

Panzer Lehr, on 2nd Panzer’s left, had moved more slowly and less effectively—due in part perhaps to a bit of self-inflicted fog and friction. Bayerlein missed a possible chance to reach the Meuse when, on December 22, he halted to rest his men and allow them to celebrate Christmas with the extra rations sent forward for the occasion. According to some reliable accounts, Panzer Lehr’s commander had also sacrificed a good part of December 19 flirting with a “young, blond, and beautiful” nurse in a captured American hospital.

The story invites comparison with the “yellow rose of Texas,” whose dalliance with Santa Anna allegedly distracted the Mexican general in the crucial hours before the battle of San Jacinto. But a harem of nurses would have made no difference as Allied reinforcements continued to arrive in the northern sector and Patton’s Third Army conducted a remarkable 90-degree turn north that took it into Bastogne on December 26.

Hitler was confronted with two choices: evacuate the salient and withdraw the panzers for future employment, or continue fighting to keep the Allies pinned down and draw them away from the industrial centers of the Ruhr and the Saar. Being Hitler, he decided on both. The infantry was left to hold the line, supported by what remained of the army’s panzers and, temporarily, the Waffen SS, for whom Hitler had other plans.

The operational result was two weeks of head-down fighting as American tanks and infantry hammered into the same kind of farm- and-village strong points that had so hampered Watch on the Rhine. Now it was German antitank guns ambushing Shermans whose relatively narrow treads restricted their off-road mobility in the deepening snow. Not until January 16 did Patton’s 11th Armored Division connect at Houffalize with elements of the 1st Army advancing from the north, forcing back Panzer Lehr despite its orders to hold the town “at all costs.” For the next two weeks the Americans pushed eastward as the defense eroded under constant artillery fire and air strikes. It was not elegant but it was effective. The Bulge from first to last cost the Germans over 700 AFVs—almost half of the number committed. About half the Panthers still in German hands were downlined for repair.

1 thought on “The End of the Panzer Divisions I

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.