The Allied invasion of southern France in the late summer of 1944, an operation first code-named ANVIL and later DRAGOON, marked the beginning of one of the most successful but controversial campaigns of World War II. However, because it fell both geographically and chronologically between two much larger Allied efforts in northern France and Italy, both its conduct and its contributions have been largely ignored. Planned originally as a simultaneous complement to OVERLORD, the cross-Channel attack on Normandy, ANVIL actually took place over two months later, on 15 August 1944, making it appear almost an afterthought to the main Allied offensive in northern Europe. Yet the success of ANVIL and the ensuing capture of the great southern French ports of Toulon and Marseilles, together with the subsequent drive north up the Rhone River valley to Lyon and Dijon, were ultimately to provide critical support to the Normandy-based armies finally moving east toward the German border.
The controversy that swirled around ANVIL one that has continued to the present, concerned not its timing or success, but its very existence. Opponents of ANVIL, including British Prime Minister Winston Churchill, have long argued that the invasion of southern France did little more than sap the strength of the main Allied campaign in the Mediterranean, the drive north up the Italian peninsula toward Austria and Hungary. This direct thrust through the so-called soft underbelly of German-dominated Europe might also, in retrospect, have altered the East-West balance of postwar Europe. In contrast, defenders of ANVIL, mainly Americans, have steadfastly maintained that even if the rugged Italian campaign could have been accelerated, the operational and logistical difficulties of rapidly crossing the Julian Alps would have been impossible to overcome. Far more significant to the Allied cause in Europe was the capture of Marseille, France's largest port, and the rapid rehabilitation of the Rhone valley rail and road network. Until the opening of Antwerp in December 1944, this supply route was to satisfy over one-third of the Allied logistical needs in northern France. In addition, the Southern France Campaign resulted in the arrival of the third Allied army group opposite the German border, without which General Dwight D. Eisenhower's army groups would have been stretched thinner and pressed harder during the German Ardennes offensive in the winter of that year. And a more grievous Allied setback in December might also have had dire consequences on postwar Europe for the Western Alliance.
Following the entry of the United States into the war in late 1941, American and British leaders incessantly debated the future direction of their joint war effort. Both agreed that defeating Germany had priority over Japan, but their strategic approaches toward the war against Germany were fundamentally different. The American military chiefs, led by Army Chief of Staff General George C. Marshall, championed a direct approach—an immediate strike into northern France and then east to Germany with all available resources. The British preferred a "peripheral" approach, using Allied maritime mobility to attack the European Axis at its extremities, where it was weakest, sapping its strength and regarding the cross-Channel attack as the final blow. The two outlooks, in part the product of national temperaments, had their analogies in Britain's early emphasis on the economic blockade of the Continent and in the U.S. Army Air Force's later commitment to a direct strategic bombing campaign against the Nazi heartland. In the end, the lack of Allied resources in 1942 and 1943 dictated that the British course of action be followed. Thus TORCH, the Anglo-American landings against French North Africa in November 1942, was followed by HUSKY, the assault against Sicily in early July 1943, and the invasion of southern Italy in September.
During a series of Allied strategic planning conferences in 1943, the invasion of southern France, ANVIL emerged as a possible complement to the cross-Channel attack against northern France, now code-named OVERLORD and finally projected for 1944. Taking place either just before or during OVERLORD, ANVIL would weaken the overall German defenses in France or prevent the Germans in the south from reinforcing those in the north. Throughout the fall and winter of 1943 the U.S. Seventh Army headquarters based on Sicily thus drew up plans for a one-, two-, or three-division assault on the French Mediterranean coast, using what amphibious lift remained after all OVERLORD needs had been met. Meanwhile, the campaign in Italy continued, turning into a grueling uphill fight in mountainous terrain that heavily favored the defender. Neither the Allied amphibious attack behind the German lines at Anzio in early 1944 nor the subsequent capture of Rome in June promised to end what had become a painfully slow war of attrition. To many American leaders and soldiers, the Italian campaign looked increasingly like a cul-de-sac, or dead end. During the winter of 1943-44, Eisenhower, commanding the Allied forces in the Mediterranean, had left to take charge of the Allied expeditionary armies assembling in England for OVERLORD. Shortly thereafter, Lt. Gen. Jacob L. Devers, who had headed the American effort in Britain, moved to the Mediterranean to become deputy theater commander under its new British chief, General Sir Henry Maitland Wilson. There Devers, another Marshall protege like Eisenhower, pushed preparations for ANVIL. Although Wilson and the commander of the American Fifth Army in Italy, Lt. Gen. Mark W. Clark, had their eyes set on Rome and northwards, it was Eisenhower who put a temporary halt to ANVIL. In view of the strengthened German defenses in Normandy, he judged that OVERLORD would need all of the amphibious lift available in the European area to ensure its success. His concerns, together with the general shortage of amphibious vessels, especially LSTs (landing ship, tanks), and the demands of the Pacific theater for such shipping, finally led Allied leaders in April 1944 to cancel ANVIL.
ANVIL, however, proved difficult to bury. General Devers, as commander of the U.S. Services of Supply in the Mediterranean, refused to reallocate the supplies and equipment that had already been gathered for the landing. At the same time he instructed Lt. Gen. Alexander M. Patch, a recently arrived veteran of the Guadalcanal Campaign in the Pacific, to continue planning for the southern France assault as the new Seventh Army chief. Thus, in the summer of 1944, after General Wilson's latest Italian offensive had run its course and more critical, after Eisenhower's OVERLORD forces had bogged down in the hedgerows of Normandy, the possibility and even the need for ANVIL again became evident. With additional shipping from the Pacific and northern European theaters available, the Allied command officially resurrected it on 24 June. But it was not until 11 August, four days before the landing was scheduled that the Allied high command gave Wilson final approval for the assault, and even this was only over the strenuous objections of Winston Churchill.
As drawn up by General Patch's Seventh Army staff, the nucleus of ANVIL would consist of the U.S. VI Corps under Maj. Gen. Lucian Truscott with the U.S. 3d, 36th, and 45th Infantry Divisions, commanded respectively by Maj. Gens. John W. "Iron Mike" O'Daniel, John E. Dahlquist, and William W. Eagles. As shipping schedules and the situation ashore allowed they were to be followed by seven French divisions under the overall command of General Jean de Lattre de Tassigny. As the American divisions had significantly more combat and amphibious experience than their French counterparts, many of which were colonial units only recently organized in French North Africa, it seemed logical for Truscott's forces to make the initial assault. In fact, the officers and men of both the American corps and its three divisions probably constituted one of the most experienced teams in the Allied camp, in contrast to the many green American divisions that went ashore at Normandy. Most were veterans of the North African, Sicilian, and Italian campaigns who had long become accustomed to working with one another. Their teamwork would prove vital to the success of the ensuing campaign.
The relationship of the Seventh Army with the French command and the higher Allied theater headquarters was also critical. Here de Lattre agreed that, for the duration of the campaign in the south, his forces, which included two corps and one provisional army-level headquarters, would remain subordinate to Patch's Seventh Army and Wilson's Mediterranean command. It was understood however, that once the ANVIL forces joined Eisenhower's Normandy-based armies, these arrangements would change. At that time de Lattre would establish an independent command for all French combat units, while Devers would head another new headquarters, the Sixth Army Group, to control de Lattre's First French Army and Patch's Seventh, all under the overall command of Eisenhower. Both Eisenhower and Wilson approved of the agreement in July.
Other elements of the ANVIL order of battle included an ad hoc airborne division, the Anglo-American 1st Airborne Task Force, under Maj. Gen. Robert T. Frederick; the Canadian-American 1st Special Service Force, an experienced regiment-size commando force; and various French special assault detachments. Air support would generally stage out of Corsica, about 100 miles away, supplemented by naval aviation from several escort carriers operating offshore. The latter vessels, together with supporting warships and the entire amphibious assault fleet, were under the control of Vice Adm. Henry K. Hewitt, U.S. Navy, the veteran commander of the Western Naval Task Force.
German
armor passing through Toulouse. (DA photograph) |
Logistical considerations were critical for both ANVIL and the ensuing campaign. Patch had three principal objectives: establish a suitable beachhead; capture the ports of Toulon and Marseille; and drive north to join Eisenhower's forces. To assist the VI Corps in the initial tasks, he had given Truscott a brigade-size force, Combat Command (CC) Sudre, from one of de Lattre's two armored divisions. But since the American infantry divisions, with their attached tank and tank-destroyer battalions and organic vehicles, were highly mobile— each was roughly the equivalent of a full-strength German panzer grenadier division—this was probably unnecessary. More important was the fact that Seventh Army planners had heavily weighted the ANVIL force in favor of munitions and artillery rather than fuel and vehicles. The scarcity of shipping allowed them little flexibility in this regard. Since the Germans had bitterly contested all previous Allied assaults on the Continent, Patch and Truscott were prepared for a major battle at the beachhead with the campaign becoming mobile only after the Germans had exhausted all of their defensive capabilities. These logistical limitations, however, would later become a major factor in the campaign.
The selection of the actual landing site was less complex. After rejecting a direct assault on the major ports whose seaward fortifications appeared formidable, Allied planners finally chose the St. Tropez area some 30 miles east of Toulon. The proposed beachhead included about 30 miles of coastline (and 50 miles of shoreline) and swung inland to a depth of 20 miles, encompassing the Maures and Esterel hill masses. The region contained several excellent strands for landing operations and good exits into the interior and an early seizure of the hills would provide adequate security for the ensuing troop and supply buildup.
Possession of this terrain would allow ANVIL forces to move along the coastal roads northeast to Cannes and Nice or, more likely, to push west for Toulon, traveling either along the coast or through the Argens River valley, an east-west corridor that began just north of the Massif des Maures. Farther north and northeast the rough Maritime Alps were less appealing, but promised to impede any Axis interference with the landing from northern Italy.
Both air attacks and French partisan activities were to focus on interdicting the general beachhead area, preventing German reinforcements from arriving. Meanwhile, the 1st Special Service Force would assault German positions on the islands of Port Cros and Levant, which overlooked the southernmost invasion beaches, while French commandos sealed off the coastal road on both flanks. Finally, Frederick's paratroopers would jump in early, concentrating on the small French town of Le Muy located generally between the two massifs. Possession of the Le Muy area would both protect the initial landings along the coast and provide an entrance into the Argens valley corridor. Paratroopers, special service forces, and commandos would attack during the night of 14-15 August, but the actual landings would begin at 0800 on the 15th, providing ample time for the final air attacks and Hewitt's naval gunfire to do their work by daylight.
On the German side the attack would come as no surprise. Luftwaffe air reconnaissance had chronicled the Allied naval buildup for General Friedrich Wiese, commanding the defending Nineteenth Army in southern France. But neither he nor his superior, General Johannes Blaskowitz, heading Army Group G. could determine the precise landing area, nor had they the forces to defend the entire coastline adequately. Blaskowitz also had to maintain strong forces in western France, to defend the Atlantic coast, and since the Normandy invasion in early June, he had seen many of his best reserve units transferred north. Finally, German naval and air power in southern France was extremely weak, and steadily increasing guerrilla attacks by the French Resistance continued to hamper German lines of communication.
Nevertheless, the German leaders did what they could to prepare for the assault. Wiese, the Nineteenth Army commander, heavily fortified the areas around Marseille and Toulon and put his troops on almost constant alert. His three corps headquarters and seven infantry divisions, although understrength and underequipped, were rested and led by veteran officers and NCOs. In army group reserve were the 11th Panzer Division, at almost full strength, and a mountain division, the former still west of the Rhone near Toulouse and the latter based at Grenoble.
By 14 August, anticipating that an Allied assault was imminent and that the blow might well fall in the Marseille-Toulon region, the German commanders had begun to move both the 11th Panzer and two infantry divisions east across the Rhone. If Wiese's infantry divisions could delay any amphibious invaders along the coast, long enough forthe heavy panzers and other reinforcements to arrive, they might still give the Allies a sharp lesson in the art of war.
Allied intelligence organizations had accurately tracked German redeployments north from the Mediterranean and the repositioning of their existing forces in southern France. In this effort ULTRA, code-name for the highly secret Anglo-American program for intercepting, deciphering, and disseminating German radio traffic, was closely complemented by information supplied by the French Resistance. Patch's real problem lay in the late decision for ANVIL and the need to assemble all participating forces as rapidly as possible after he had received at least a tentative green light on 24 June.
Wilson had the majority of the American units earmarked for ANVIL pulled out of the front lines in Italy by the end of the month, along with four of the French divisions, but all were exhausted and in need of rest and recuperation. Also, because of its late arrival, Truscott's VI Corps, which was to execute ANVIL, had little input into the planning process, a risky but necessary fact of life. In sum, the time available for detailed logistical preparation, training and rehearsals, and final loading was severely limited and tied to rigid schedules.
In July all ANVIL preparations intensified. While the participating infantry, commando, and naval units undertook some quick amphibious refresher training off the coast of southern Italy, Frederick hastily assembled his division-size airborne and glider force from a variety of bits and pieces in the theater, and de Lattre attempted to put the finishing touches on his two brand-new armored divisions. Although equipping the French almost uniformly from American sources greatly simplified ANVIL logistics, de Lattre's dependence on primarily non-French-speaking soldiers from France's overseas empire greatly limited the capabilities of his forces. The ranks of his French officers and NCOs who could handle colonial combat troops were stretched exceedingly thin and his staff and technical capabilities remained weak. Once ashore, more troops could be recruited over time, but for now he would have to make do with what manpower was available.
With the clock running, the Allied land and naval staffs supervised the massive loading requirements of the D-day convoys, their departure from a variety of ports, and their subsequent rendezvous off Corsica during the night of 14-15 August. Together they comprised approximately 885 ships and landing vessels sailing under their own power and carrying nearly 1,375 smaller landing craft, about 151,000 troops (the bulk of the French were in follow-on convoys), and some 21,400 trucks, tanks, tank destroyers, prime movers, bulldozers, tractors, and other assorted vehicles. The campaign for southern France was about to begin.
Map: Seventh Army Assault 15-16 August 1944
Even as these forces assembled, Allied air attacks against the southern French coastline and the immediate interior, begun on 5 August, continued and intensified. So as not to reveal the precise landing area, targets all along the coast were struck, including many in the Genoa area to the extreme east. Also attacked were the Rhone River bridges, whose destruction would severely hamper German movements throughout the campaign. At the same time, French Resistance forays against lesser water crossings and rail and communication sites further paralyzed German movement behind the battle area and seriously degraded internal communications capabilities. Deception efforts on the night before the ANVIL landings included dummy paratrooper drops and visits by small fleets of patrol craft, one led by cinema star, Lt. Comdr. Douglas Fairbanks, Jr. (USNR), to other potential landing sites to simulate an invasion force.
The 1st Special Service Force successfully assaulted the islands of Levant and Port Cros shortly after midnight, surprising the German garrisons but finding only dummy artillery positions. Simultaneously, the French commandos struck along the coast, with the southern group establishing blocking positions around Cape Negre, but the northern group suffering severe casualties while trapped in a defended German minefield until dawn.
Inland the echeloned paratrooper and glider landings were characterized by confusion. Always a difficult proposition by night, the airborne attack was bedeviled by a low cloud cover that obscured drop zones for pathfinder teams and pilots alike. Although most landed within a ten-mile radius of Le Muy, daylight found some as far south as the Gulf of St. Tropez while others were located as far east as the Cannes region. But, as in the Normandy invasion, the confusion that the scattered landings caused within the German interior lines may have more than made up for the almost inevitable pilot errors.
Troops
of the 45th Division wade ashore near Ste. Maxime. (DA photograph) |
Daybreak on 15 August revealed a clear Mediterranean morning with the autumn storms, the French mistral, still weeks away. As (planned) the main landings began promptly at 0800, after the haze and smoke from the final air and naval bombardments had dissipated—the experienced Allied commanders considered the visibility a worthwhile trade-off when assaulting an unfamiliar shore en mass. Striking the crease, or boundary between the German LXII Corps' 242d and 148th Infantry Divisions, General O'Daniel's 3d Division put ashore on the target area's southernmost beaches on the St. Tropez peninsula; Dahlquist's 36th Division headed for those in the Frejus Gulf on the right, or eastern, portion of the landing area; and Eagles' 45th Division employed a series of small strands in between, near the town of Ste. Maxime. Initial resistance proved light, with the two defending grenadier (infantry) regiments from two different divisions unable to coordinate their actions and with naval gunnery silencing most of the German artillery positions.
The only exception to the desultory defense occurred at the head of the Frejus Gulf, the primary landing zone of the 36th Division. With the German fire there seemingly unaffected by the Allied bombardment and with an impressive array of beach obstacles in full view, the alert naval task group commander, Rear Adm. Spencer S. Lewis, ordered the bulk of the division to land on an adjacent beach, slightly to the north, an action that further minimized Allied casualties on D-day.
During the morning and afternoon of the 1 5th, the armor-supported American infantry slowly eliminated almost all resistance along the shoreline and began pushing east and west along the coastal road and north into the interior. By the following day they had secured the two hill masses overlooking the beaches, while tank destroyers from the 45th Division had penetrated due north to assist the paratroopers in a final assault against Le Muy. Only in the immediate vicinity of Frejus did the Germans put up a spirited but futile defense, while the Luftwaffe limited itself to a few radio-controlled missile attacks against Allied shipping. Thus, by the afternoon of 16 May Truscott found his forces in full possession of the planned beachhead with little evidence of any coordinated German response.
Confusion reigned at the various German headquarters. The LXII Corps at Draguignan, a few miles northwest of Le Muy, found itself isolated by roving bands of paratroopers. Reports of the landings arriving at the headquarters of the Nineteenth Army and Army Group G were fragmentary and confused, with most of Blaskowitz's information coming from naval sources and relayed to his command post at Toulouse through Paris. From Avignon, Wiese did what he could. He gave General Richard von Schwerin, whose 189th Infantry Division was currently attempting to cross east over the Rhone, a few units
from two other divisions. He sent von Schwerin down the Argens valley the morning of the 15th to clear the paratroopers from Le Muy and relieve the LXII Corps headquarters. But von Schwerin's "counterattack" that afternoon and a similar one from Cannes by elements of the 148th Division were small ad hoc affairs based on only sketchy intelligence. Both were easily dispersed by the Americans, who hardly noticed them, and swallowed up by the rapid ANVIL advance. More would be necessary if the Germans were to mount an effective defense, but Blaskowitz and Wiese could do little until more of their combat forces crossed the Rhone, especially the 11th Panzers.
Surveying the situation on 16 August, Truscott had no desire to wait for an effective German response. Earlier in the year, as commander of the 3d Infantry Division at Anzio, he had watched another corps commander delay movement inland with disastrous results. Thus Truscott, whose aggressive temperament would match that of de Lattre and complement Patch's more reserved, paternal style of command, recommended moving immediately west before the Germans could organize a coherent defense. Patch agreed. He ordered the VI Corps to strike west with the 3d and 45th Divisions, spearheaded by CC Sudre, while he accelerated the arrival of the first French divisions. The U.S. 36th Division was to secure the eastern flank of the beachhead until relieved by the paratrooper and 1st Special Service Force units. Here both Patch and Truscott remained wary, concerned that despite intelligence reports to the contrary, Field Marshal Albert Kesselring, the able German commander in the Italian theater, might spoil all their plans with a surprise thrust from northern Italy across the German-controlled passes of the Maritime Alps.
Between 17 and 19 August, the two American divisions pushed west, their projected route of advance passing north of Toulon and Marseille and leading directly to the Rhone and Avignon, the location of Wiese's command post. At the same time, the Nineteenth Army commander tried desperately to establish a north-south defensive line, using first the 242d Division centered around Toulon and then the 244th Division guarding Marseille, together with various bits and pieces of the 189th and 198th Divisions as they ferried across the Rhone. His efforts were to no avail against the fast-moving Americans.
Meanwhile, de Lattre's infantry had been coming ashore and heading west along the coast, relieving the southernmost 3d Division units for a French-directed attack on Toulon. Until their heavy equipment arrived, the French would have to depend on VI Corps artillery and support units, and for the same reason, CC Sudre had to be returned at once to de Lattre's control. The acceleration of the campaign also began to affect Allied logistical capabilities, not in terms of the amount of supplies available but in the ability to transport them inland. Nevertheless, all ANVIL commanders agreed that keeping the initiative was paramount.
Thunderbirds
on 645th tank destroyers move inland. (DA photograph) |
At this critical juncture in the Southern France Campaign, the German High Command began to reevaluate its entire position in the west. By 16 August the failure of the German counterattack at Mortain in northern France had confirmed the Allies' breakout at St. Lo from their Normandy beachhead. And with many of their divisions in danger of annihilation in the Falaise Pocket, the German leaders finally elected to order a general withdrawal from France. Instructions went out from Berlin that day ordering Blaskowitz to begin moving his Atlantic coast forces to the east, out of western France, and his Mediterranean forces to the north, up the Rhone River valley. The only exceptions were the two easternmost German divisions, Wiese's 148th Division in the Cannes-Nice area and Blaskowitz's reserve mountain division at Grenoble, both of which were to withdraw east into Italy where they were to pass over to Kesselring's control. Finally, in accordance with long-standing plans, Blaskowitz was to leave strong garrisons at Toulon, Marseille, and several key Atlantic ports. At Hitler's personal order, these vital coastal enclaves were to be defended to the last man.
With the disorganization of German communications in France, Blaskowitz did not receive the transmitted orders until 17 August. Meanwhile, the Allied ULTRA program had picked up the original radio message, quickly deciphered and translated it, and rapidly dispatched it to the Seventh Army headquarters. Thus, on the afternoon of 17 August, Patch, Blaskowitz, and Wiese were all digesting the withdrawal order at about the same time.
A second factor, unknown to the German commanders at the time, but ultimately also greatly affecting their situation, was the actions of Task Force Butler. Back in July Truscott had decided to form a small, mobile striking force under his assistant corps commander, Brig. Gen. Frederick B. Butler, soon after the initial landings. This force, drawn primarily from the 36th Division, included a motorized battalion from the 141st Infantry, two medium tank companies, a tank destroyer company, a light cavalry squadron, and a self-propelled artillery battalion.
Assembling the force on the night of 17-18 August, Butler had first struck out west in the wake of the 45th Division and then north into the mountains heading in the general direction of Grenoble. By the 19th he had reached the Digne-Sisteron region about fifty miles north of the 45th Division and one-third of the way to Grenoble. Although the mechanized unit had met no organized German resistance, it was still without a specific mission except for carrying out a general reconnaissance-in-force.
On the American side, news of the German retreat soon split the campaign into three parts. First was the conquest of Toulon and Marseille, a Seventh Army priority long assigned to the follow-on French divisions. Second was Truscott's general drive west, securing the northern flank of de Lattre's forces along the coast, but also having objectives of its own as it bulled its way toward Wiese's headquarters. Finally, there was Butler's unopposed foray to the north. In Truscott's mind, this opportunity was hard to resist. If Butler could be directed behind the German line of withdrawal and adequately reinforced' the Nineteenth Army might well be destroyed and all of the tedious fighting that had typified the Italian campaign avoided.
For Truscott's superior, General Patch, the situation was more complex. Logistics had now become his most pressing problem. The Seventh Army simply lacked the transport and gasoline to support two, let alone three major advances. For sustained operations inland, Toulon and Marseille would have to be taken and rehabilitated. Eisenhower's forces in northern France were already suffering from a lack of operational ports even as they raced toward Antwerp and Brest; Patch, with even less over-the-beach supply capability, could not afford similar difficulties. Thus, on 19 August he made Aix-en-Provence, some twenty miles north of Marseille, the western limit of Truscott's VI Corps advance. From there the American divisions could best protect what had always been the Seventh Army's second main objective, seizing Toulon and Marseille.
Prior to the execution of ANVIL, de Lattre had intended to attack the two ports in succession. The accelerated landings of his French forces, however, and the general situation allowed him to envision almost concurrent operations against both. He ordered Lt. Gen. Edgar de Larminat to move west against Toulon along the coast, with two infantry divisions supported by tanks and commandos. Simultaneously, a second force, under Maj. Gen. Goislard de Monsabert and consisting of one infantry division and similar supporting forces, would advance in a more northwesterly direction, encircling the naval port from the north and west and probing toward Marseille. De Lattre knew that the German garrisons at the ports were substantial: some 18,000 troops of all types at Toulon and another 13,000, mostly army, at Marseille. However, Resistance sources also told him that the defenders had not yet put much effort into protecting the landward approaches to the ports, and he was convinced that a quick strike by experienced combat troops might well crack their defenses before they had a chance to coalesce. Speed was essential.
On the morning of 20 August, with the German command in Toulon still in a state of confusion and the Nineteenth Army more concerned with Truscott's westward progress well north of the port, de Larminat attacked from the east while Monsabert circled around to the north, quickly outflanking Toulon's hasty defenses along the coast. By the 21st Monsabert had cut the Toulon-Marseille road, and several of his units had entered Toulon from the west, penetrating to within two miles of the main waterfront. Between 21 and 23 August the French slowly squeezed the Germans back into the inner city in a series of almost continuous street fights. As the German defense lost coherence, isolated groups began to surrender, with the last organized resistance ending on the 26th and the formal German surrender occurring on 28 August. The battle cost de Lattre about 2,700 casualties, but the French claimed 17,000 prisoners, indicating that few Germans had followed the Fuehrer's "stand and die" order.
Even as French forces occupied Toulon, Monsabert began the attack on Marseille, generally screening German defenses along the coast and striking from the northeastern and northern approaches. Early gains on the 22d put French troops within five to eight miles of the city's center, while a major Resistance uprising within the port encouraged French soldiers to strike deeper.
Although de Lattre urged caution, concerned over the dispersion of his forces and the shortage of fuel for his tanks and trucks, Monsabert's infantry plunged into the heart of Marseille in the early hours of 23 August. Their initiative decided the issue, and the fighting soon became a matter of battling from skeet to street and from house to house, as in Toulon. On the evening of the 27th the German commander parlayed with Monsabert to arrange terms and a formal surrender became effective on the 28th, the same day as the capitulation of Toulon. At Marseille the French took over 1,800 casualties and acquired roughly 11,000 more prisoners. Equally important, both ports, although badly damaged by German demolitions, were in Allied hands many weeks ahead of schedule.
While de Lattre's forces besieged the ports, Truscott refused to remain idle. On the evening of 20 August and with Patch's approval, he ordered Task Force Butler west toward Montelimar, a small French city on the east bank of the Rhone directly astride the German evacuation route. There Butler was to establish blocking positions and await the arrival of the 36th Division, elements of which were already headed toward Digne and Sisteron.
At the time the orders arrived, Butler's forces and the leading units of Dahlquist's 36th Division were scattered between Aspres and Gap, twenty to thirty miles above Sisteron, all the way back to Le Muy, and oriented more for an advance north to Grenoble than for a dash west to the Rhone. The shift was further delayed by incessant communications problems in the rough Maritime Alps, where towns and roads were generally located in the deep valleys, and above all by the general Allied shortage of fuel and vehicles. Captured German fuel dumps at Draguignan, Le Muy, and Digne helped put Butler's force and one additional battalion of the 36th Division in the Montelimar region by the evening of the 22d, but most of the American infantry had to move north in a complex and time-consuming series of foot marches and truck shuttles.
Complicating matters was the inexperience of the 36th Division commander, General Dahlquist, one of the few new senior officers in Truscott's corps, as well as the indecision of his superiors. On the afternoon of the 21st, for example, Truscott and Patch had received an ULTRA intercept informing them that elements of the 11th Panzer Division had crossed the Rhone and were headed directly for Aix-en-Provence. The news apparently made Truscott hesitate to send specific instructions to Dahlquist and led him to temporarily cancel the northward movement of one 45th Division regiment as well. In addition, the limited distribution of ULTRA information, a security precaution, meant that neither Butler nor Dahlquist had any clear idea of German withdrawal plans up the Rhone valley or east from Grenoble into Italy. The net result was that American units arrived at Montelimar in dribs and drabs, leading to several angry communications between Truscott and Dahlquist. Without greater logistical capabilities, however, American flexibility at the extremities of the campaign area was limited.
For Wiese, Truscott's pause westward was an obvious relief. The respite enabled him to withdraw his remaining infantry divisions and corps—the LXII Corps at Draguignan had been eliminated—across the Rhone and Durance Rivers by 23 August and to send them on their way up the Rhone toward Lyon. Accompanying them were hordes of naval, Luftwaffe, and German support and administrative personnel, civilian and military alike—the baggage of the long German occupation, many of whom would not survive the long trek north to the German border.
If Wiese's situation below Avignon had been greatly eased by Truscott's inaction, his problems to the north were only beginning. Sometime on the 21st the German army commander had received word that American combat forces, including armor and artillery, had begun interdicting traffic a few miles above Montelimar, between the Drome and Roubion Rivers. Local troops had repulsed an attack against Montelimar itself, but the Americans had occupied the hill masses to the north and northeast in strength. Wiese's immediate response was to urge Maj. Gen. Wend von Wietersheim, commander of the 11th Panzer Division, to hasten efforts to ferry his heavy vehicles across the Rhone and move north as quickly as possible to secure the German route of withdrawal. In fact, the small armored team which Wiese had earlier sent to Aix-en-Provence—which had been duly reported to Truscott via ULTRA channels—had been merely a ruse, something to keep the Americans guessing. But von Wietersheim would have a difficult time moving his armor north, and the Germans would now begin to pay for their ill-thought-out withdrawal plans and the absence of any security elements west of the Rhone above Avignon.
The ensuing struggle around Montelimar ultimately pitted the U.S. 36th Division against von Wietersheim's panzers and, as they arrived northward, the surviving infantry divisions of the Nineteenth Army. In brief, the battle, which lasted for about nine days, from 21 to 29 August, saw both sides commit increasingly larger forces against the other with indecisive results. The retreating German units ultimately forced their way to Lyon, but suffered horrendous casualties in the process.
Butler's actions on the very first day typified the American dilemma. Shooting up whatever attempted to move north of Montelimar during the afternoon of the 20th, the task force lacked the infantry to physically occupy the road, especially at night, or the munitions to interdict the highway by fire alone, despite the arrival of two corps artillery battalions dispatched by Truscott. Instead, Butler focused his strength on Hill 300, a sharp north-south ridgeline about four miles north of Montelimar overlooking the main road near the river village of La Coucourde. By night, American armor generally pulled back into the Condillac Pass area, immediately north of Hill 300, where Butler established his command post.
Fortunately for the Americans, von Wietersheim's transportation problems were equal to their own, with the number of ferries capable of carrying his 45-ton Mark V Panther tanks limited, the roads crowded, fuel at a premium, and incessant Allied strafing attacks forcing him and his fellow commanders to make most of their movements by night. But as it slowly arrived in the critical region, the German armored division proved equally aggressive. On 22 August von Wietersheim's armored reconnaissance battalion, the first element of the 11th Panzer Division to arrive at Montelimar, launched an immediate attack, moving east along the southern bank of the Roubion and then striking north behind Butler's positions. The dangerous thrust was thrown back by some late armored arrivals of Task Force Butler from Sisteron.
On the 23d and 24th, von Wietersheim repeated these flanking attacks with greater strength but less success as Dahlquist, assuming command of the battle from Butler, positioned more of his newly arriving forces on the Roubion front. Both sides also launched attacks and counterattacks against one another in the immediate vicinity of Montelimar with equally indecisive results. But to the north German infantry was finally able to clear the western slopes of Hill 300 of Americans, allowing the German withdrawal to resume.
Still unhappy with the situation, Wiese ordered von Wietersheim to clear the Americans from the entire area on 25 August. In addition to his armored division, the army commander put the bulk of the 198th Infantry Division, which had now arrived, together with two Luftwaffe air defense regiments and a medley of other units, including several railway guns, at his disposal. Complying, von Wietersheim launched five separate attacks on 25 August in an effort to keep the American center occupied while his armor struck deep into both flanks along the Roubion and Drome Rivers, surrounding the American position. But lack of coordination hampered the complicated series of attacks, which met strengthened American resistance.
Dahlquist was able to assemble his entire division as well as additional munitions supplies in the area. By that evening the American commander had not only been able to avoid encirclement, keeping secure his supply lines east, he had also managed to block the main highway just below La Coucourde with an infantry-tank team after the Germans had inadvertently left the area undefended. Only an impromptu midnight cavalry charge by German heavy tanks, led personally by a disgusted von Wietersheim, restored German control of the immediate roadway, knocking out ten of the lighter American tanks and tank destroyers in the process.
Despite von Wietersheim's success in keeping the road open— save for harassing American artillery fire—the situation of the Nineteenth Army was becoming increasingly desperate. In the south the two ports had been invested, allowing Patch to begin directing more supplies to Truscott's VI Corps and in turn allowing Truscott to push O'Daniel's 3d Division west and then north in pursuit of the withdrawing Germans. At the same time the VI Corps commander also sent Eagles' 45th Division north, backstopping Dahlquist's positions in the Montelimar region with one regiment and sending the bulk of the division toward Grenoble.
In the north, far above Lyon, Blaskowitz's forces from western France—a corps headquarters, two infantry divisions, and an assortment of other odds and ends—were still desperately fleeing east, as was most of the German civil-military establishment that remained in France. Harassed by Allied air attacks and the increasingly bolder French Resistance, it was only a matter of time before some of these columns would be overrun by Patton, or another aggressive Allied commander. For all these reasons, on 26 August Wiese ordered von Wietersheim to begin moving the bulk of the 11th Panzer Division to Lyon, leaving the Montelimar region in the hands of General Baptist Kniess, commander of the LXXXV Corps which had just arrived.
Between 26 and 28 August, Kniess had his withdrawing infantry divisions keep up the attacks against the 36th Division in the Roubion and Drome areas and in the hill masses in between. But his actions were primarily defensive, keeping Dahlquist too occupied with his flanks to launch a determined attack on the road while German forces moved north, many traveling on the western bank of the Rhone. Meanwhile, a rear guard engineer unit tried to keep the 3d Division at bay to the south. O'Daniel's forces had entered an undefended Avignon on 24 August and were pushing north in pursuit of Wiese's columns, their progress delayed primarily by shortages of fuel and vehicles.
By 27 August the bulk of the 11th Panzer Division had crossed north of the Drome together with almost all of the retreating infantry divisions. Only General Otto Richter's 198th Infantry Division remained at Montelimar, with the rear guard engineer detachment to the immediate south. On the night of 27-28 August, Richter led his remaining two regiments, together with a miscellany of other Germans who still hoped to elude capture, in an impromptu scramble north. In the process, one group ran straight into a major 36th Division offensive against Montelimar itself, leading to the capture of General Richter and about 700 of his troops, with the Americans suffering some 100 casualties.
As units of the 36th and 3d Divisions converged on Montelimar the following morning, they took approximately 500 prisoners, while a more thorough sweep of the battle area in the days that followed netted approximately 2,500 more. The Nineteenth Army had made good its escape, but had suffered terrible losses in the gauntlet through which the 36th Division had forced it to run.
With the fighting at Montelimar over and the southern ports secured about the same time, supplies once again began flowing to the VI Corps. Truscott was eager to begin the pursuit north. The 3d Division flowed through Montelimar almost without pause heading north toward Lyon, while the 45th Division took a slightly easterly route, moving through Grenoble and then north along the Swiss border. There was no appreciable resistance. Joining them was the 36th Division, now moving behind the 3d; lead elements of a French Algerian infantry division following in the wake of the 45th; and west across the Rhone the 1st French Armored Division, which had recently landed on the coast, moving rapidly up the river's opposite bank.
If Truscott and Patch refused to pause, neither could Wiese or Blaskowitz. With the Americans and French in hot pursuit, the Nineteenth Army commander instructed von Wietersheim, whose armored division was now just about the only effective combat unit left in his command, to cover the withdrawal of his forces farther north to Dijon.
At Dijon, Blaskowitz hoped that Wiese could form a loose cordon for a few days to allow for the arrival of those German forces streaming in from western France. That accomplished, Blaskowitz intended to withdraw what was left of his army group directly east into the Vosges Mountain-Belfort Gap area, establishing a juncture with the retreating northern army groups along the trace of the Franco-German border. Wanting no repeat of the Montelimar affair, von Wietersheim's armor was to secure the eastern flank of the withdrawal, gradually pulling back in a northeasterly direction to the Belfort Gap. While Wiese's infantry divisions plodded through Lyon and farther north during the first days of September, von Wietersheim thus prepared to fight a delaying action against the pursuing Americans.
German concerns proved well founded. Leading elements of the U.S. 3d and 45th Infantry Divisions had pulled up nearly abreast of Lyon by 31 August with the French armored division to the west keeping pace. Unwilling to become bogged down in street fighting, Truscott instructed O'Daniel to bypass the city to the east. He hoped once again to use his more mobile forces to outpace the Germans and either strike their flank or cut them off from the German border. But this time von Wietersheim, whose panzer units were now falling back on their own supply lines, was ready. On 1 September he launched a limited attack against one of the 45th Division's regimental command posts, while elsewhere the advancing Americans found bridges blown and strong German infantry-tank teams defending key towns, road junctions, and water crossings.
The
Champagne Campaign comes to a close. (DA photograph) |
Frustrated, Truscott ordered the 117th Cavalry Squadron, the corps reconnaissance unit that had formerly been a part of Task Force Butler, to have its patrols probe north and northeast, searching for gaps in the panzer division's extended front. On the evening of 2 September one unit, Troop B. managed to wind its way through undefended back roads and trails to the small town of Montrevel, about thirty miles north of the 45th Division and less than twenty miles east of the main German withdrawal route. That night Truscott ordered the squadron commander, Lt. Col. Charles J. Hodge, to concentrate the rest of his force at Montrevel as quickly as possible and to hold the area until reinforcements could arrive. Here was possibly the base for another attack on Wiese's columns.
Hodge not unexpectedly had a difficult time reuniting his widely dispersed forces. Moreover, the cavalry unit was neither organized nor equipped for sustained combat. Its three scout troops had only armored cars and jeeps, while Company F and Troop E were equipped with light tanks and 75-mm. self-propelled howitzers on similar chassis. By daybreak, 3 September, the cavalry commander had been able to reinforce Montrevel only with another scout troop, Troop A; and 45th Division units were still many miles to the south. Wietersheim, whose headquarters was near Bourg-en-Bresse only about ten miles south, was quicker. Immediately upon learning that American elements were on his own line of withdrawal, he dispatched to the scene his reconnaissance battalion, reinforced with tanks, engineers, and self-propelled guns.
The battle began about 1100 and continued late into the afternoon. From the start, the two American cavalry troops were at an extreme disadvantage, their light cannon and machine guns having little effect on the German armor. The arrival of Hodge's equally poorly armed light tanks and self-propelled guns sometime in the afternoon had little effect on the bitter fight. One officer, 2d Lt. Daniel W. Lee, used captured German panzerfaust rocket launchers, one of the few effective weapons available, to keep the German armor at bay. But by 1630 the American situation inside the town had become hopeless, with the number of wounded mounting and ammunition about exhausted. Shortly thereafter the survivors, those who had failed to escape through gaps in the German lines, surrendered. Additional American units that began to reach the battle area that evening could do little. When Hodge later took count, he found that while Troop A had lost only twelve men, only eight remained from Troop B. Most of the engaged vehicles together with several of F Company's light tanks had been destroyed.
By the night of 3-4 September, with Wiese's final units well north of Lyon and the Americans becoming increasingly aggressive, von Wietersheim pulled his panzers out of the Bourg-Montrevel region and headed northeast. Almost at once Truscott's forces took up the pursuit, three divisions abreast, with the French 3d Algerian Infantry Division coming up on the far eastern flank along the Swiss border. Meanwhile, Wiese had paused again, this time to string his divisions out in defensive positions along the Doubs River, which flowed from the Belfort Gap area to the region just south of Dijon. But his combat forces were now little more than improvised battle groups, remnants of his infantry divisions and regiments, filled in with whatever manpower and equipment could be scavenged by local commanders from the withdrawing German columns. Blaskowitz positioned a new corps he had received, whose forces were in no better shape, north of Dijon, thus creating the "pocket" the German commanders had planned to form until their Atlantic forces could arrive.
By now the problems of Truscott, Patch, and de Lattre were almost totally logistical, with the VI Corps living on about one day's worth of fuel and the French bringing supplies forward with borrowed farm transport and any other means available. Every move northward placed them that much farther from their supply depots along the coast, and many weeks would be needed to put the Rhone railway back into operation.
Nevertheless, on 6 September, with Blaskowitz's Atlantic forces finally beginning to stream into the Dijon region from the west, Truscott resumed the attack. By the 8th his divisions had rolled through Wiese's flimsy Doubs River defenses, forcing the Germans into a pell-mel1 retreat to the northeast. Only along the eastern portions of the river line, where the 11th Panzer Division guarded the approaches to the Belfort Gap in fairly good defensive terrain, did the 45th and 3d Algerian Divisions have a tougher time forcing the Doubs. Meanwhile, west of the Lyon-Dijon area, the 1st French Armored Division raced directly north cutting off many of the German forces fleeing eastward including six trainloads of troops, vehicles, guns, and supplies and thousands of disorganized German soldiers.
From 9-14 September, Truscott's forces advanced northeast, heading for the Belfort Gap. But lack of fuel and supplies, rather than German resistance, steadily reduced the pace of their progress. By now Wiese could throw no more than an assortment of weak provisional units at the Americans, forces composed of survivors, stragglers, and administrative odds and ends, including many police units and other paramilitary detachments. Placed at key towns and road junctions, they had little staying power and were easily swept aside. But until Patch could establish an effective logistical system between the southern ports and the northern battlefields and bring up more forces to bolster the now fairly exhausted VI Corps team, the Seventh Army's race to the German border was over.
On 11 September, French armored units operating north of Dijon physically linked up with elements of Patton's U.S. Third Army driving east, thereby fulfilling Patch's final objective. Shortly thereafter, control of the Seventh Army passed from Wilson's Mediterranean command to Eisenhower's northern European one. What many Seventh Army soldiers called "the champagne campaign" was over, and the battle for Germany was about to begin.
Both Army Group G and the Nineteenth Army had escaped from France despite the best efforts of Truscott and others. But Blaskowitz's troops had paid a terrible price in the process. Although the general administrative confusion in the German command structure makes an exact count of German losses difficult, some facts are fairly clear. To some 31,000 German prisoners taken in the St. Tropez, Toulon, and Marseille areas must be added the thousands taken around Montelimar, with another 12,000 surrendering during the Allied drive north from Lyon and some 20,000 more cut off west of Dijon by the early collapse of Wiese's pocket. In addition, Army Group G prisoners collected by Patton's Third Army numbered about 10,000 while some 25,000 more were left isolated in their Atlantic coast garrisons (still others had retreated into Italy). Estimates of German combat casualties in southern France run as high as 7,000 killed and three times that number wounded, figures that might be accurate if losses to Allied air attacks and French Resistance actions could also be added in.
In all, Blaskowitz may have lost over half of the 250,000 troops assigned to his command at the beginning of ANVIL. Allied losses included about 4,500 American battle casualties (slightly over 2,000 killed, captured, or missing), with French losses running slightly higher and French Resistance casualties also hard to estimate.
The two critical determinants of the campaign were Allied aggressiveness and Allied logistical considerations. From the beginning it was Truscott's relentless pursuit that set the tenor of the Franco-American effort, a posture that Patch and de Lattre were quick to support and emulate. Wiese was thus reduced almost entirely to reacting to Allied moves, forcing the vaunted panzers to lead the German exodus rather than acting as a mobile rear guard.
Of course, the campaign might well not have taken place at all without the efforts of General Devers to continue preparations for ANVIL after its abrupt cancellation in April. In addition, de Lattre's rapid conquest of Toulon and Marseille, which together would soon be providing for over one-third of the Allied supply needs in northern France, allowed the ports to become operational significantly before the stormy mistral season began. Indeed, by 14 September, D plus 30, the Seventh Army had achieved objectives that ANVIL planners had not expected it to attain until about D plus 120. Therein, of course, lay Patch's problem.
The acceleration of the campaign was the root cause of Allied logistical difficulties. As early as D plus 1, fuel shortages had developed; by 19 August they had become critical. On 21 August, for example, VI Corps' three infantry divisions needed about 100,000 gallons of gas per day, while only 11,000 gallons were left in ANVIL beach dumps. During the last stages of the campaign, VI Corps captured another 183,000 gallons of high-octane gasoline and 36,500 gallons of diesel in a German fuel depot along the Doubs, enough to push Truscott's forces a bit closer to the Belfort Gap, but not enough to do much more. In this respect Wiese was fortunate to escape with anything at all, for only the American transport problems prevented Truscott from putting more combat power at Montelimar sooner.
The Allied commanders were clearly assisted by the ULTRA intercept program, which revealed the details of the German withdrawal, a rare intelligence coup. However, even Truscott hesitated to move decisively until German movements could be confirmed, especially when another intercept several days later led him momentarily to expect the arrival of the German panzer division at Aix. And despite ULTRA intelligence to the contrary, Patch remained concerned about his eastern flank opposite the Italian border, retaining his airborne and special service force troops there in blocking positions. If the intelligence was inaccurate or incomplete, as it was later regarding the German offensive in the Ardennes, then only Patch would be responsible.
In sum, the Southern France Campaign showed what experienced, well-led Allied troops could do against their German foes. Neither Truscott nor Patch wanted a repeat of the Italian campaign, the long slugging match on Guadalcanal, or the two-month stalemate in Normandy. De Lattre and his French officers were equally eager to get on with the liberation of their homeland. All were willing to take risks to shorten the campaign, and each was confident that his troops and commanders could carry out even the most difficult maneuvers. It was in this respect that the campaign for southern France, one which resulted in the presence on Eisenhower's southern flank of a strong Allied army group rather than a hostile German one, differed markedly from many other Allied efforts and deserves more study and attention than it has yet received.
Further Readings
The number of scholarly works treating the campaign in southern France is extremely limited. The official histories are Samuel Eliot Morison, The Invasion of France and Germany, 1944-1945, Volume II in the History of United States Naval Operations in World War II series (1959), and the official volumes in the United States Army in World War II, including Robert W. Coakley and Richard M. Leighton, Global Logistics and Strategy, 1943-1945 (1968); Maurice Matloff, Strategic Planning for Coalition Warfare, 1943-1944 (1959); Forrest C. Pogue, The Supreme Command (1954); Marcel Vigneras, Rearming the French (1958); and especially Jeffrey J. Clarke and Robert Ross Smith, Riviera to the Rhine (1993), the latter based on German and French records as well as on U.S. Army archival holdings of units participating in the campaign. Useful memoirs include Jean de Lattre de Tassigny, The History of the French First Army, trans. Malcolm Barnes (1952), and Lucian K. Truscott, Jr., Command Missions, A Personal Story (1954), while historian Arthur L. Funk examines the French Resistance and associated special operations in Hidden Ally (1992). Nevertheless, many gaps in the story remain.
CMH Pub 72-31
last revision