Well there it is; it won’t work, but you must bloody well make it.
—General Sir Alan Brooke, Chief of the Imperial General Staff
In April, as the battle at sea was beginning to turn in the Allies’ favor, the joint organization established at the January 1943 Casablanca Conference to prepare for the Allies’ return to northwestern Europe, finally took form in London. Although the directive to the new organization was still evolving through transatlantic drafting, British Lt. Gen. Sir Frederick Morgan was assigned to lead this team as the Chief of Staff to the Supreme Commander (Designate). Morgan would begin with an American deputy, but with the supreme Allied commander still to be chosen. To name the new organization, Morgan simply drew from his new job title to create an acronym. “COSSAC” would be to the point for those with the “need to know” and, in its élan, potentially misleading to outsiders and enemies.
Morgan’s suitability to lead the task had been plumbed by the British Chiefs of Staff (COS) and Winston Churchill. Lt. Gen. Sir Hastings “Pug” Ismay, the prime minister’s chief military assistant, gave Morgan a file of previous planning work on assaulting Germany through northwestern Europe. Ismay asked Morgan for a reply within twenty-four hours to “elaborate the means by which the expedition was to be organized and undertaken.” Morgan responded with a memorandum on cross-Channel attack that he discussed with the COS on 24 March 1943.1 Churchill then interviewed Morgan on April 4. In a note back to Ismay, in which the prime minister approved the layout and draft directive to COSSAC as proposed by the COS, Churchill also passed judgment that Morgan “seems a capable and sensible officer.”2 Perhaps as relevant, given the toxic environment into which Morgan was being thrust, the 1928 Staff College report that Ismay shared with Churchill had seen in Morgan “a keen but kindly sense of humor which should prove a great asset to him.”3
Morgan queried Lt. Gen. Jacob Devers, recently arrived in London to take up command of the European Theater of Operations, U.S. Army (ETOUSA), for his recommendation for an American deputy for COSSAC. Among the Americans in London, Devers replied, “Well, the logical man is Ray Barker. He’s been in this racket all along. He’s more familiar with it and is more deeply immersed in it than anyone else.”4
Pairing Morgan with Barker for Anglo-American leadership of COSSAC proved to be inspired. They shared similar military backgrounds as artillery officers. Each came to the task with a measure of understanding and affinity for the other’s national culture. Barker had visited Britain and enjoyed studying its history. Morgan simply liked Americans. From time spent in contact with Canadian units in France in World War I and in 1940, Morgan had gained an appreciation for North American idioms and mannerisms that served him well with his allies.5 The Americans in turn quickly came to accept Morgan. Their assessment was, “He’s a straight shooter.”6
The two were well aware that the friction between the senior British and American officers above them sometimes approached open hostility. However, neither allowed suspicion and questioning of his ally’s motives or abilities to constrain candor or cooperation with each other. Morgan and Barker turned to advantage circumstances of security, urgency, shortage, and the unknown to build COSSAC into a community of purpose that would evolve to set a standard for integrated Allied effort.
At their first meeting as chief of staff and deputy in April 1943, Morgan set the tone with a small but telling act. He snipped a Royal Artillery brass button from his tunic and exchanged it for a brass button from Barker’s U.S. Army uniform. From then until the Quadrant Conference in August, each general went around the COSSAC offices and to external meetings, British, American, and joint, wearing a token from his ally’s army.7 They each ignored jibes about “going “Yank” or “going native.” Theirs was a clear statement of trust and joint endeavor to comrades and often fractious superiors.
COSSAC was allotted office space in Norfolk House in St. James Square, an easy walk from the Admiralty, War Office, Cabinet War Rooms, and other London centers of authority that were key to COSSAC’s mission. Morgan, however, did not consider this an auspicious location. The building had been the site for an earlier series of planning efforts that, with the exception of the Torch invasion, came to nothing good. Morgan rued Norfolk House’s “reputation as a home of lost causes.” The first of the new team to arrive found little to begin with beyond a few desks, chairs, and a dropped pencil.8 Yet the separate British and American planners of cross-Channel attack rallied on COSSAC from across London, and the organization made its start.9 Their plan for a “full scale assault against the Continent” was due to be submitted through the British COS in ninety days.
Setting the purpose of COSSAC in the first meeting of his staff, April 17, Morgan broke from a general assumption in stating that “although the primary object of COSSAC is to make plans, I am certain that it is wrong to refer to it in any way as a ‘planning staff.’ The term ‘Planning Staff’ has come to have a most sinister meaning: it implies the production of nothing but paper. What we must contrive to do somehow is to produce not only paper but ACTION.”10 Acknowledging that neither he nor the COSSAC staff began their task with executive authority, Morgan concluded with an irresistible challenge to his young officers:
My idea is that we shall regard ourselves in the first instance as primarily a coordinating body. We plan mainly by the co-ordination of effort already being exerted in a hundred and one directions. We differ from the ordinary planning staff in that we are, as you perceive, in effect the embryo of the future Supreme Headquarters Staff. I do not think I can put the matter any better to you than by quoting to you the last words of the CIGS, who said: “Well there it is: it won’t work, but you must bloody well make it.11
The organizational form of COSSAC initially followed the British staff model on the assumption that the supreme Allied commander would be a British officer with an American deputy. Over time in the absence of precedent, except for the command staff established by French Marshal Ferdinand Foch in World War I, the merging of effort led COSSAC’s organizational form to develop following function, not nationality. COSSAC came to be staffed accordingly, as Morgan had intended from the outset.
Barker later said of COSSAC, “Here was a team, and whether a man wore a U.S. or a Crown [uniform insignia] made not a bit of difference. . . . No one thought about whether a chap was British or American or Canadian [or] South African.”12 By the time General Eisenhower was appointed Supreme Allied Commander and the Supreme Headquarters, Allied Expeditionary Force (SHAEF) emerged, absorbing COSSAC, the initial, deliberate practice of assigning a British and an American deputy to each section had been replaced entirely. Throughout SHAEF’s organization, a carryover from COSSAC, a single section leader was assigned on the basis of matching ability to need with little regard to nationality.
The activating directive from the Combined Chiefs of Staff to Morgan, defining the task at the initiation of COSSAC’s work, was developed through March from alternative U.S. and British-recommended language and completed in April 1943. The directive blended both national views of the way ahead. At the first COSSAC meeting, Morgan read the COS version of the directive to his staff on the assumption that this draft would be approved by the Americans. Obvious between its lines was an expectation that further discussion at the highest Allied level, or events in Europe not under Allied control, would determine the strategy ultimately to be selected. The British draft of the task stated:
Our objective is to defeat the German fighting forces in North-West Europe. To this end the Combined Chiefs of Staff will endeavor to assemble the strongest possible forces (subject to prior commitments in other theatres) in constant readiness to reenter the Continent if German resistance is weakened to the required extent in 1943. In the meantime the Combined Chiefs of Staff must be prepared to order such limited operations as may be practicable with the forces and materiel available.13
The directive further ordered COSSAC to develop plans for three operations. First listed, and relevant to either the British or American preferred strategy for return to the Continent, was to prepare a series of operations over the summer of 1943. Termed Cockade, these operations were intended, through feints and deception, to test German strength in northwestern Europe and draw the Luftwaffe into air battles in which its capacity to oppose an Allied invasion, be that reactive or proactive, could be degraded by attrition. The task listed second was to plan “for a return to the Continent in the event of German disintegration at any time from now onwards with whatever forces may be available.” This would become known as Operation Rankin. The third, and last, task was to plan “for a full scale assault against the Continent in 1944 as early as possible.” Not yet named, this would become Operation Overlord.
Thus, from the outset, the competing British and U.S. views of strategy comingled in COSSAC’s governing directive: Rankin’s contingent reaction to opportunity, if offered by an internal collapse of the Reich, and Overlord’s proactive assault against Germany’s main force in the West as it stood at a time and place of attack unilaterally determined by the Allies. The U.S. Joint Chiefs of Staff–recommended draft differed from the British by setting the objective to “destroy” German forces in northwestern Europe and narrowed the date for full-scale assault to “the spring of 1944.”14 This became the agreed directive for cross-Channel assault when issued by the CCS on April 26.15
While actively engaged in planning and coordinating resources for deception operations under its first task, Cockade, COSSAC would also endeavor to develop plans required by its second and third tasks, Rankin and Overlord. In 1944, upon implementation of Overlord, plans to initiate Rankin continued to be maintained and updated regularly for many months beyond D-Day as a contingency response to a potential internal collapse of Germany.
The activating directive to COSSAC designated the British Chiefs of Staff to be the source for COSSAC’s further guidance as required and as COSSAC’s reporting channel to the Anglo-American CCS. The latter instruction would become challenging at a critical moment.
The directive also told Morgan to maintain close contact with the headquarters of ETOUSA. COSSAC was forbidden to interact with Allied military staffs other than British and American until told to do so by the COS, a stricture that was not easily relaxed. For example, early thinking had Canadian units at the center of the force to be landed. However, the “lukewarm” COS only “tacitly conceded” in April 1943 the benefit of appointing a Canadian liaison officer to COSSAC, Maj. Gen. G. R. Turner.16
To initiate the development of its tasks, COSSAC had access to the considerable body of work and the corporate memory of London groups—British, American, and joint—which had been assessing and planning for an Allied return to northwestern Europe, in some cases, since the Dunkirk evacuation in 1940. This included the lessons to be learned—many at a high cost in casualties—from raids on the Continent planned and conducted by the British Combined Operations Headquarters led by Vice Adm. Lord Louis Mountbatten. Inherited work also contained the output of a British committee, the Combined Commanders, with whom COSSAC’s new American deputy, Maj. Gen. Ray Barker, had participated as the liaison officer from the U.S. Army in Europe. Among the Combined Commanders’ products was a February 1943 outline plan that had not been formally approved. Called Operation Skyscraper, this plan laid out an assault on France’s Cotentin Peninsula very similar in location and size to what Overlord eventually became.17
However, at COSSAC’s initiation in April 1943, a viable plan to act on the generalized and debated intent to reenter the Continent through northwestern Europe was anything but clear. Many challenges and uncertainties revolved in kaleidoscopic combinations, each demanding consideration both separately and in relation to the others. Darkening all of COSSAC’s planning was the shadow of interwar and current war professional skepticism: was a large-scale opposed amphibious assault viable at all?
Hitler’s assertions about the strength of his defenses at the water’s edge, the Atlantic Wall, were obvious. Equally obvious was the British aversion to incurring again anything like the casualties of World War I. To be added to this was extensive, objective analysis of the costly, unsuccessful Anglo-French landing at Gallipoli in 1915. Professional military opinion, developed in the interwar years and by no means limited to the British, tended to the view that amphibious success was very doubtful against an adversary who easily could concentrate and reinforce from the interior of a continent.18 Freshly underscoring this conclusion in Europe was Operation Jubilee, the failed large-scale Anglo-Canadian raid on Dieppe in August 1942. Of Jubilee’s principally Canadian raiding force of 6,000, more than 3,300 had been killed or captured. Sobering too was admission by the operational commanders and service chiefs that the much larger, better coordinated, and supported American landings in Morocco in November 1942 against weak, brief French resistance succeeded in large measure by luck.19
Almost alone in the interwar years, the U.S. Marine Corps had continued to develop amphibious doctrine and tactics in anticipation of their mission to assault Japanese-occupied islands in the Pacific. However, the Marines’ island-centric task description did not address the assumption of a defending adversary’s reinforcement from the interior of a continent. Even the Marines entered World War II with no purpose-designed and built amphibious assault equipment.20
The Allies were starting on parallel tasks near the bottom of a steep challenge. Against the experts’ odds, they had to learn how to succeed in amphibious assault while creating the wherewithal to do so.
Potential landing beaches and the terrain behind them, from the Dutch coast to Brittany as far as Spanish border, had to be considered. However, from early days attention had tended to focus on two areas in France, the Pas de Calais and Normandy.21 Would there be one landing or multiple landings? Concurrently or sequentially?
On the fundamental material questions, the Combined Chiefs of Staff directive to COSSAC was silent. How many Allied divisions of what type could COSSAC anticipate being available at which phases for a “spring of 1944” assault? Silence from the CCS. In direct, crucial relation to the number of divisions, what was the capacity of the shipping that would be available for the amphibious assault and follow-on buildup and sustainment? Silence.
Definition of shipping requirements was impossible absent knowledge of how many divisions would be in the assault. The most that Morgan could offer in May 1943 was to warn the British COS of the obvious: that the required number of landing craft would be “large enough . . . to present a very serious problem, which has no precedent.”22
COSSAC knew with certainty only one thing. With its own champion, the Supreme Allied Commander, still to be designated, cross-Channel attack was last in line for assignment of scarce amphibious lift behind named commanders already leading scheduled operations around the world.
What would be the quality of the Allied troops the ships would land on the coast of northwestern Europe? On May 19, as the CCS were struggling with their compromise agreement at the Trident Conference, Lieutenant General Morgan, in London and unaware that the very issue was on the table in Washington, minuted his staff on this question. Combat-experienced troops, he told them, would be vital to success.23 Where would these troops come from, and at the expense of which other operations?
There were crucial, operations-shaping, but unresolved geographic, geologic, hydrographic, meteorological, and technical questions and challenges. Which beaches and their inland terrain, in combination with which ports capable of supporting the landed forces, offered viable prospects for a successful military assault and follow-on offensive? On what axes of advance? Without intervening rivers and their estuaries impeding the just-landed national armies’ capability to support each other in battle? Which combination of beaches and terrain also offered the best potential for rapidly establishing advanced, tactical airfields? Were the assault and follow-on forces really dependent on seizing French Channel ports for sustainment, or were nascent ideas for over-the-beach logistical support on an enormous scale a viable alternative to capturing ports? In that case, for which beaches specifically? For how long at what capacity? In spite of the Channel’s fearsome weather?
Taking all of this into consideration, where lay the relative advantage for Allied fighter planes flying from England in support? For the Luftwaffe opposing them in the air and attacking the Allied landing force? How were the assault troops to receive sufficient fire support in the critical moments before they hit the beach? What constituted sufficient fire support, from the sea and from the air, against a strongly fortified and defended coast? That was a type of assault with which, by the spring of 1943, the Allies had had no experience with the disastrous exception of Dieppe and the luck-retrieved confusion of Torch. Would the answer for fire support promise conditions allowing a landing in daylight? Or would a need for compensating surprise drive the assault toward a predawn landing, which would be an offshore organizing nightmare for the navies?
Lacking guidance from the CCS on the critical factors, COSSAC could begin its work in April 1943 only by making initial assumptions about strategic objectives and the forces required to be available to achieve them. Specific to its third task, cross-Channel attack into Europe, Morgan directed his staff to apply a broad assumption about resources and the timing of their availability. Their task, he said, was to reconquer Europe with an Allied army that would grow to number 100 divisions, 15 of which would be British (or more precisely, Anglo-Canadian) and 85 American, complemented by powerful air forces and naval forces of similar proportion.
Most of this force was still in North America. As COSSAC started its work, Morgan told his staff that only the advanced guard of this huge force was in the United Kingdom and subsequent to securing ports on the Continent, the bulk of these forces would enter the theater directly from across the Atlantic. Therefore, COSSAC must address its tasks with “a map which starts at one end in San Francisco and the other end in Berlin.”24 Morgan lost no time in hanging precisely such a map in his office. Morgan, who had been left to define Overlord’s military purpose himself, cast the operation’s strategic objective with equally broad scope:
Our ultimate object is to wage successful war on land in the heart of Europe against the main body of the German strategic reserve. It is true that we have to cross the enemy’s beaches, but that to us must be merely an episode. True, it is a vital episode and, if it is not successful, the whole expedition will fail. We must plan for crossing the beaches, but let us make sure that we get that part of the plan in its right perspective as a passing phase.25
COSSAC was going for Berlin.
COSSAC measured the operation’s ultimate objective against three factors. These were (1) the strength of enemy opposition to be encountered; (2) the enemy’s disposition and the quality of Allied forces to be applied, plus (3) the requirement to focus assets then available in the United Kingdom (bombers, for example) on preparation of the battlefield. The approach enlightened planning as it broadened the scope of the task.
Morgan concluded that the cross-Channel attack into northwestern Europe must be considered to have begun already in April 1943. He pressed the COSSAC staff to embrace their work with this assumption as a given and recommended this view to the Chiefs of Staff.
Infused as COSSAC was from the outset with energy and purpose, doubts about the strength of higher command support for their enterprise also tugged at Morgan, Barker, and certainly their perceptive principal staff officers. From direct experience and participation, they knew of the “corrosive and potentially explosive mixture” of strategy discussions in London and Washington into which COSSAC was introduced. The new organization proceeded with its work while Morgan pondered alternative, conflicting interpretations of COSSAC’s purpose. As he later described the alternatives, “Was it to be a form of chemical agent to induce a sea, land, and air change so that there might be brought about unity of intent, unity of method, and so unity of action? Or was it perhaps to be as a scapegoat upon which could be heaped responsibility for what seemed in the eyes of each to be the sins of the other?”26
Morgan did not want to presume upon the still-to-be-named Supreme Allied Commander’s authority by developing for him only a single option. Also, the driving question for initiation of Overlord had to be resolved: which of the most favored potential entry points to the Continent best met Allied needs?
To make a start, Morgan focused COSSAC’s work on a competitive assessment of the two alternative entry points that consistently came out of previous groups’ attention to the problem of cross-Channel assault, the Pas de Calais and Normandy. The British Army contingent at COSSAC was assigned to develop the Pas de Calais option, and the U.S. Army group was given the task of developing an assault into Normandy. COSSAC’s Air and Naval sections would help each team as needed. The possibility of multiple assaults, one in each area, was left open, as was the possibility that the odds against success would be compounded to such a degree as to force a conclusion that any cross-Channel assault was unfeasible.
Acting as close but neutral observers to the competition would be Morgan, Barker, and a very carefully selected few: Gen. Sir Bernard Paget, commander of the British Home Forces from which many of the divisions to cross the Channel would be drawn, ETOUSA commander Lt. Gen. Jacob Devers, and the chief of Combined Operations, Vice Adm. Lord Louis Mountbatten.27
To start the two teams on common ground, on May 25 Morgan issued to them his own outline plan for what he called Operation Rudge. Into Rudge Morgan incorporated all of the operational and logistical criteria common to an assault in either area, specifying that COSSAC had determined that the landings should be on the coast of France, but taking care in the wording of Rudge not to suggest a preferred choice of where. On the teams’ response to these criteria, the options of landings in the Pas de Calais or Normandy would be assessed.
COSSAC’s U.S. Army and British Army planners approached their task from different perspectives. In addition to knowing General Marshall’s strong preference for direct action to conclude a short war in Europe and then turn to the Pacific, the American planners had an appreciation for the enormous surge of troops and materiel building in North America and soon to be projected across the Atlantic. The British planners, conditioned by almost four years of thin reserves and shortages, were influenced by the long game strategy of their military chiefs and their prime minister to wear down the enemy through attrition.
Morgan used his Rudge paper to eliminate that difference in perception at the outset, at least within COSSAC. He affirmed to his teams that only one of these strategies could motivate them by stating, “The decision to invade Europe in 1944 will imply at once the abandonment of long term policies of attrition in favor of the short term policy of direct assault.” Just as unequivocal was Morgan’s anchoring of the magnitude of the challenge in “the inescapable fact that while the armies of the assault must be for the most part inexperienced and untried in war, those of the defense may well contain strong cadres of battle-hardened soldiers. This inequality is further emphasized by the fact that the attack must consist of that most intricate of all military operations, the seaborne assault.”28
The already complex interrelationship of beach and landing ground characteristics and enemy strength was further compounded to frustrate the teams’ assessments by the resource questions left unanswered in the April 26 CCS directive to form COSSAC. Morgan’s projection of the size of the ultimate force to liberate Europe was reasonable. But specific to the cross-Channel assault into France itself: how many Allied divisions, how many ships, how many planes? The teams’ response was to begin by developing their assessment of how many resources of what type they thought would be required for the assault without any clear knowledge of how many resources they actually would have.
At that time, even the name of the cross-Channel operation had been rendered ambiguous by past conceptions of its size. The original plan for a forty-eight-division assault presented by the U.S. Army to the British in 1942, and so far in excess of what the Allies could field in 1942 as to be considered naïve, had been code-named Roundup. The subsequently proposed contingency plan for a cross-Channel attack with whatever divisions were at hand in the United Kingdom, either as an emergency response to an imminent collapse of the Soviet armies or in response to an internal collapse of the Third Reich, was code-named Sledgehammer.29 Subsequent Anglo-American discussions of cross-Channel attack evolved toward an operation of a size somewhere between these two options that tentatively was called Roundhammer.
As COSSAC in London wrestled with the awesome, multifaceted challenge of an Allied return to Europe, at least some answers to their most basic questions were emerging from contentious Anglo-American discussions in Washington. Included would be the name for the operation that would stick, Overlord.