Chapter X: 
QUADRANT - Shaping the Patterns: August 1943
As the QUADRANT Conference drew near, General Marshall and his staff were convinced of the need for a showdown with the British. Once before-in July 1942-Marshall had led a move for a showdown. Then he had had to yield on a cross-Channel operation and accept TORCH instead. A series of opportunistic moves had followed in the Mediterranean-moves the U.S. staff sought to parallel with limited offensive actions in the Pacific. Marshall had fought to keep the Mediterranean commitments limited while he struggled to keep the BOLERO idea alive and the war against Japan progressing. But there was always the danger that the two limited wars-one in the Pacific, the other in the Mediterranean-would become all-out wars or absorb so much that little would be left for a major offensive in northwest Europe. The Army planners now feared that the Mediterranean trend had already gone so far as to be well-nigh irreversible. There were also signs, as Marshall was aware, of increasing restlessness among Navy planners, anxious to get on with the Pacific war, over his European strategy. At hand was an acceptable plan for concentration in the United Kingdom for a cross-Channel operation-Plan OVERLORD. The time for a final decision on European strategy therefore appeared to Marshall to have arrived. But would the President now support him, and, if so, would they be able to convince the Prime Minister and his staff?
Staff Planning and the President's Position
Part of the answer was soon to come. General Marshall, on the eve of his departure for Quebec, met with the President at the White House to discuss the line of action to be followed at QUADRANT. At this meeting on 9 August, the President observed that the planners were "always conservative and saw all the difficulties," and that more could usually be accomplished than they would admit. Between OVERLORD and post-HUSKY operations in the Mediterranean (now called PRICELESS), he assured the Army Chief of Staff that he was insistent on OVERLORD. But he felt that more could be done in the Mediterranean than was currently proposed by the U. S. planners. Pointing to the scheduled departure of seasoned troops from the Mediterranean to the United Kingdom, he agreed that the seven battle-tested  

divisions be provided for OVERLORD. He proposed, however, that seven fresh divisions be dispatched from the United States for PRICELESS. At the same time the President assured Marshall that he did not wish to have anything to do with an operation into the Balkans nor did he even intend to agree to a British expedition in that area that would cost the United States vital resources such as ships and landing craft necessary for other operations. He was in favor of securing a position in Italy to the north of Rome and taking Sardinia and Corsica, thereby posing a serious threat to southern France.
General Marshall replied that the United States "had strained programmed resources well to the limit" in the agreements already reached regarding OVERLORD and PRICELESS. While the movement of three divisions from PRICELESS forces to OVERLORD could be undertaken without a loss in troop lift and with some advantage in equipping the French, beyond this point movements to OVERLORD of veteran units would cost the United States part of its troop lift. Marshall feared that the proposed movement from the United States to PRICELESS would result in a corresponding reduction for OVERLORD. He promised the President, however, that he would have a critical review made of the logistical factors involved. In a humorous vein, the President remarked that he did not like Marshall's use of the word "critical" since he wanted help in carrying out his idea rather than obstacles placed in the way.1
The Army planners that same day presented a report on the logistical implications of reinforcing PRICELESS.2 It indicated that, on the basis of optimistic predictions of available personnel lift in the Atlantic, it would be possible, by 1 May 1944, to build up in the United Kingdom the force of 1,300,000 U.S. troops provided for in TRIDENT estimates and, in addition, to lift approximately 100,000 more either to the United Kingdom or to some other theater.3 Recent troop lists prepared by the European Theater of Operations, however, called for over 1,400,000 U.S. troops by I May 1944 to make up balanced striking forces for the Combined Bomber Offensive (POINTBLANK) and OVERLORD. If the TRIDENT estimates of 1,300,000 were adhered to, and if the optimistic shipping estimates for the Atlantic proved correct, the Army planners admitted that seven U.S. divisions could be lifted to North Africa by the middle of 1944 without affecting the availability of divisions for OVERLORD as set at TRIDENT. But, the planners pointed out, the utilization of personnel shipping to lift troops to the Mediterranean before June 1944 would not contribute as much to striking a direct decisive blow at the European Axis as the employment of the same shipping to insure a well-balanced force in the United Kingdom. They observed, moreover, that it was the opinion of General Eisenhower and of the JCS that the force 

currently committed to the Mediterranean (less the seven divisions scheduled for transfer to the United Kingdom) would be adequate to achieve the desired objective of occupying Italy to a line north of Rome, seizing Sardinia and Corsica, and making a diversionary effort against France from the Mediterranean. The addition of seven divisions to General Eisenhower's forces in the Mediterranean would make a total of thirty-one divisions available in that area, as compared with twenty-nine for the main effort, OVERLORD. The Army planners therefore called for the full troop lift of 1,400,000 to be allocated to OVERLORD and POINTBLANK.4
At the session of the JCS at noon on 10 August, shortly before a scheduled conference of the JCS with the President, General Marshall reported the President's inclination to furnish seven new divisions from the United States to replace the seven veteran divisions scheduled for transfer to the United Kingdom. Arguing that this reinforcement of PRICELESS would occur at the expense of the build-up for OVERLORD, he emphasized that, even if the seven additional divisions were provided, they could not arrive in the Mediterranean before June 1944. The President, he went on, should be informed of General Eisenhower's report that he had sufficient resources to conduct the proposed operations in Italy. The President should also be apprised that an additional force of seven divisions would in reality constitute an expeditionary force available for use in the Balkans. General Marshall felt that the President was opposed to operations in the Balkans, and particularly to U.S. troop participation in them, on the ground that they represented an uneconomical use of shipping and also because of the political implications involved. In rallying the JCS against the Presidential proposal, General Marshall was supported by Admiral King, who was particularly fearful lest the provision of shipping for seven new divisions to the Mediterranean seriously curtail planned Pacific operations.5
Meanwhile-at one o'clock that same afternoon-Secretary of War Stimson, who had recently returned from the United Kingdom, conferred with the President at the White House. That very morning, he had decided to present his conclusions to the President in writing. So serious did he consider the action he was about to recommend decisions that would affect Marshall's position in the Washington high command-that he called Marshall in to let him read what he was going to say, in case Marshall had any vital objections. Stimson recorded the reaction of the Army Chief of Staff in his diary entry for that day: "He said he had none but he did not want to have it appear that I had consulted him about it. I told him that for that very reason I had signed the paper before I showed it to him or anyone else." Stimson also recorded that the conference that followed at the White House was "one of the most satisfactory" he had ever had with the President.
In the course of the conversation he produced his letter of conclusions. In it Stimson reasoned:

We cannot now rationally hope to be able to cross the Channel and come to grips with our German enemy under a British commander. His Prime Minister and his Chief of Imperial Staff are frankly at variance with such a proposal. The shadows of Passchendaele and Dunkerque still hang too heavily over the imagination of these leaders of his government. Though they have rendered lip service to the operation, their hearts are not in it and it will require more independence, more faith, and more vigor than it is reasonable to expect we can find in any British commander to overcome the natural difficulties of such an operation carried on in such an atmosphere of his government.
Stimson went on to point out that the difference between the Americans and the British was "a vital difference of faith." The U.S. staff believed that only by massing the great vigor and might of the two countries under overwhelming mastery of the air could Germany be defeated. The British theory was that Germany would be beaten by a "series of attritions" in the Mediterranean and the Balkans. The USSR, to which both the United States and Great Britain were pledged to open a second front, would not be fooled by "pinprick warfare"-a special danger in the light of postwar problems. Stimson concluded his letter:
I believe therefore that the time has come for you to decide that your government must assume the responsibility of leadership in this great final movement of the European war which is now confront- 'in us. We cannot afford to confer again g and close with a lip tribute to BOLERO which we have tried twice and failed to carry out . . . . Nearly two years ago the British offered us this command. I think that now it should be accepted-if necessary, insisted on.
Finally the time had come to put ". . . our most commanding soldier in charge of this critical operation at this critical time." Lincoln had had to fumble by trial and error until he discovered the right man. Wilson had to choose a relatively unknown. But Roosevelt was far more fortunate. He had General Marshall, who
. . . already has a towering eminence of reputation as a tried soldier and as a broadminded and skillful administrator. This was shown by the suggestion of him on the part of the British for this very post a year and a half ago. I believe that he is the man who most surely can now by his character and skill furnish the military leadership which is necessary to bring our two nations together in confident joint action in this great operation. No one knows better than I the loss in the problems of organization and world-wide strategy centered in Washington which such a solution would cause, but I see no other alternative to which we can turn in the great effort which confronts us.
The President, Stimson noted, "read it [the letter] through with very apparent interest,, approving each step after step and saying finally that I had announced the conclusions which he had just come to himself." 6
Later on 10 August the JCS joined the Secretary and the President at the White House to discuss the coming conference with the British at Quebec.7 At this meeting the President reported that, from his conversations with the Secretary of War, he had learned that the 

Prime Minister currently favored operations against the Balkans but was opposed to an operation against Sardinia. The Secretary of War qualified this statement, pointing out that Mr. Churchill had disclaimed any wish to land troops in the Balkans, but had indicated that the Allies could make notable gains in that area if the Balkan peoples were given more supplies. The Secretary of War affirmed that the British Foreign Secretary, Mr. Anthony Eden, wished the Allies to invade the Balkans. To this the President added that the British Foreign Office did not wish the Balkans to come under Soviet influence, and therefore the British wished "to get to the Balkans first." He himself did not follow the logic of the British thinking on the Balkans. He did not believe, he stated, that the USSR desired to take over the Balkan states but rather that the USSR wished to "establish kinship with other Slavic people." He assured the U.S. military leaders that he himself was opposed to Balkan operations. In arguing against a Balkan operation the President reasoned along the lines of the view that had been emphasized by General Marshall and General Handy on the undesirability of basing hopes for victory on political imponderables. He declared that it was "unwise to plan military strategy based on a gamble as to political results." 8
In recommending against the President's proposal for replacing the seven trained divisions to be taken from the Mediterranean with seven from the United States, General Marshall and Admiral King repeated the arguments they had advanced at the meeting of the JCS earlier in the day. On the basis of the War Department study he had made, General Marshall reported that the seven new divisions could be transported to North Africa by the end of June 1944 and the planned build-up for OVERLORD could still be executed. But he emphasized General Eisenhower's belief that even without the seven divisions to be sent to the United Kingdom he would still have a sufficient force to conduct the projected operations in Italy, capture Sardinia and Corsica, and have fourteen divisions available for an invasion of southern France in co-ordination with OVERLORD. Dispatching an extra seven divisions to the Mediterranean, Marshall argued, would meet the desires of Mr. Churchill and Mr. Eden, invite their use in an invasion of the Balkans, and so extend the Mediterranean operations as to have a harmful effect on the main effort from the United Kingdom. Following the presentation of these arguments, the President announced that he would advocate leaving General Eisenhower with his current build-up, less the seven divisions earmarked for transfer to the United Kingdom.
Turning to the basic question in grand strategy, Admiral King suggested to the President that, if the British insisted upon abandoning OVERLORD or postponing OVERLORD indefinitely, the United States should abandon the project. The President replied with the optimistic view that the United States itself could, if necessary, carry out the cross-Channel operation. He felt certain that the British would make the necessary bases available to the United States for the operation. General Marshall ob-  

jected to the President's suggestion on the ground that fifteen British divisions were already available in the United Kingdom. In no other place in the world, he maintained, could fifteen divisions be put into an operation without entailing great transportation and supply problems. The President affirmed his wish for the preponderance of U.S. forces in OVERLORD from the first day of the assault in order to be able to justify the choice of an American commander for the operation. In line with the emphasis that had been placed by Washington military planners on the wastage in the Allied war effort resulting from past divergences from the main plot, General Marshall cautioned the President against subsequent changes in basic decisions. He was prepared to accept only minor diversions from the main plan-and those only when absolutely necessary. It was especially important to avoid such dislocation of the American war effort as had resulted earlier from the change from BOLERO to TORCH. Marshall reminded the President that every such shift in plans resulted in changes in production, loading of convoys, and other phases of U.S. war mobilization, which "reached as far back as the Middle West in the United States." 9
The U.S. leaders left the meeting of 10 August agreed to insist on the continuation of the current build-up for the cross-Channel operation from the United Kingdom and on carrying Out OVERLORD as the main U.S.-U.K. effort. The JCS now had the President behind them in their plans for Europe.10 When they had touched on strategy and planning differences in the war against Japan, the JCS had urged the President to try to persuade the Prime Minister to put full British support behind the projected Burma operations-operations to which the President had already agreed on 26 July.11
Stimson has recorded the delight of the U.S. staff with the "clear and definite" stand of the President on the conduct of the war against Germany, marking the full acceptance by the Commander in Chief of the military policy for which Stimson and Marshall had been fighting.12 The ranks of the American high command appeared closed as never before. Cheered as the staff was, the question remained whether the President and his military advisers could see the agreed policy through in the conference with the British.

The Conferees Assemble
After the completion of their preparations, the U.S. military delegation left for Quebec. In preparing for the meetings with the British, American military planners as well as their chiefs had carefully studied British preparations, representation, and techniques in negotiations at past conferences and had taken steps to match them.13
At Quebec, amid the quaint 18th century charm of the French city, the military staffs were quartered and held their meetings in the impressive Château Frontenac overlooking the St. Lawrence River. The President and Prime Minister made their headquarters at the old fortress known as The Citadel, close by the historic Plains of Abraham, and currently the summer seat of the Governor General of Canada. Special ramps had been installed for the President's use, and the two plenary sessions of the conference were held here for his convenience. As the delegations assembled, the news from the war fronts-especially of the war against Germany-was definitely encouraging. Reports of Italian peace moves were persistent. In Sicily the campaign was in its final stages, and by 17 August-early in the conference-Sicily was entirely in Allied hands. On the Eastern Front the Russians had seized the initiative and had begun to drive the Germans back to their homeland. The Combined Bomber Offensive had finally gotten under way in earnest. In the war on the U-boats in the Atlantic, the tide that had turned in the spring of 1943 was running even more strongly in Allied favor. American intelligence estimates on the eve of the conference predicted that the German war against Allied shipping would continue, but with diminishing effect; that the Germans would try, during 1943, to improve their defensive position in the USSR, and to impair Soviet offensive capabilities by attrition; and that the Germans would stand on the strategic defensive on all fronts during 1944, yielding outlying territories only under compulsion. The estimates held that Germany would resist as long as there was any hope of a negotiated peace.14
In Alaska, U.S. troops had occupied Kiska, and the Japanese had finally withdrawn from the Aleutians. General MacArthur's Southwest Pacific forces were ready to advance on Salamaua in New Guinea and South Pacific forces were driving ahead on the island of New Georgia. Only in the CBI had the front remained more or less stationary. U.S. intelligence estimates on the Pacific and Far East situation held that Japan would probably remain on the strategic defensive unless convinced that an attack by the USSR was imminent or that


TOP MILITARY PLANNERS AT QUEBEC. From left: Mal'. Gen Thomas T. Handy, Brig. Gen. Albert C. Wedemeyer, Mai. Gen. Muir S. Fairchild, and Vice Adm. Russell Willson
TOP MILITARY PLANNERS AT QUEBEC. From left: Maj. Gen Thomas T. Handy, Brig. Gen. Albert C. Wedemeyer, Mai. Gen. Muir S. Fairchild, and Vice Adm. Russell Willson.
major operations were to be launched by the Allies in China. Serious reverses for the Allies in the Pacific or for the Soviet Union in Europe might also lead to a shift by Japan to the offensive.15
American and British planners arrived on the scene to lay the groundwork for the conference a few days before the principals began-their sessions.16  Then in the eleven-day period of the conference between 14 and 24 August 1943, the U.S. and British Chiefs of Staff met for a full-dress debate on Allied strategy in the war. Present among the American delegation to assist General Marshall were General Handy, the Assistant Chief of Staff, OPD, and General Wedemeyer, the Army planner, and a considerable number of other Washington Army planners delegated for duties on the planning and working staff level.17 

The Prime Minister brought to the conference a full staff complement, including such special assistants as General Morgan (COSSAC), Brigadier Wingate, and Maj. Gen. A. W. S. Mallaby, General Wavell's Deputy Chief of Staff for Operations, who had flown in from India.18 Churchill arrived at Quebec on to August, then journeyed to Hyde Park for a brief visit with the President, returning to Quebec on 15 August. The President did not arrive at Quebec until the 17th-three days after the Combined Chiefs had begun their sessions.
Debating the Issues in the War Against Germany
The Arguments
In the discussions at QUADRANT between the staffs on the war against Germany, the U.S. Joint Chiefs sought a final resolution of the question whether the main effort was to be made from the United Kingdom or in the Mediterranean.19 In the process, they sought agreement on the relationship between operations in the two areas. As usual in the conferences with the British in mid-war, General Marshall served as the principal American spokesman on European strategy. In part this was recognition of his strong convictions and his talents in advocacy and military diplomacy. In part it was acceptance of the view that the American concept of European strategy was essentially that of the U.S. Army and its defense in debate with the British should properly be conducted by the War Department spokesman. As already suggested, Marshall was convinced that a final choice between the basic alternatives of cross-Channel versus Mediterranean now had to be made. He was prepared and willing to risk a showdown with the British at this point-the consequences of which he fully realized. He made clear to his colleagues that
 . . . we must go into this argument in the spirit of winning. If, after fighting it out on that basis, the President and Prime Minister decided that the Mediterranean strategy should be adopted, he wished that the decision be made firm in order that definite plans could be made with reasonable expectation of their being carried out.20
As it had previously decided, the American delegation immediately presented its proposal that OVERLORD be given overriding priority over other operations in the European theater.21
Sir Alan Brooke replied for the British Chiefs of Staff that the British were in complete agreement with the U.S. Chiefs of Staff that OVERLORD should be the major U.S.-U.K. offensive for 1944. Nevertheless, he went on to stress forcefully the necessity of achieving the three main conditions on which the success of the OVERLORD plan was based: (1) the

reduction in German fighter strength; (2) the restriction of German strength in France and the Low Countries and of German ability to bring in reinforcements during the first two months; and (3) the solution of the problem of beach maintenance. To create a situation favorable to a successful OVERLORD was the main British aim of Allied operations in Italy. The desired Allied, vis-à-vis enemy, strength, Brooke emphasized, could be attained by operations in Italy to contain the maximum German forces and by air action from the most suitable Italian bases to reduce German fighter forces. In this connection Sir Charles Portal argued the advantages of gaining the northern Italian airfields.22 Not too surprisingly, the British soon turned the discussion to the much-debated question of the seven divisions. If the seven divisions were withdrawn from the Mediterranean, the British Chiefs argued, the Americans and British would run risks in the Mediterranean that might preclude or jeopardize success in OVERLORD. On the basis of this reasoning, Sir Alan Brooke concluded, therefore, that the decision sought by the U.S. Joint Chiefs between OVERLORD and operations in the Mediterranean would be "too binding." 23
In reply, General Marshall questioned whether the necessary conditions for OVERLORD could be brought about only by increasing Allied strength in the Mediterranean. If Italian resistance proved to be weak, he agreed, the Allies ought to seize as much of Italy as possible. While it would be better if the Allies held the northern airfields of Italy, he believed that almost as much could be accomplished from the Florence area. In his opinion, a successful OVERLORD could be insured only by giving it an overriding priority. Unless OVERLORD were given that priority, the operation might never be launched. Unless the seven divisions from the Mediterranean were dispatched and the necessary means were concentrated for OVERLORD, OVERLORD would at best become a "subsidiary operation." A delay in such decisions not only would hinder the OVERLORD build-up but also would have repercussions on Pacific operations. Marshall again emphasized, this time to the combined staffs, that any exchange of troops contrary to TRIDENT agreements "would absorb shipping" and upset supply arrangements "as far back as the Mississippi River." Unless OVERLORD were given an overriding priority, General Marshall went on, the entire U.S.-U.K. strategic concept would have to be revised. In that event, the United States and the United Kingdom would have to rely on air bombardment alone to defeat Germany, and only a reinforced U.S. Army corps for an "opportunistic" cross-Channel operation might well be left in the United Kingdom. Although the Combined Bomber Offensive had accomplished great results, the final outcome of that operation-and the very possibility of an opportunistic cross-Channel undertaking - remained "speculative." Such a recasting of strategy, he pointed out to the British, might lead to a possible reorientation of American offensive efforts toward the Pacific.24
The British position, as could be expected, was not inflexible. On 16 Au-

MEMBERS OF U.S. AND BRITISH STAFFS CONFERRING, Quebec, 23 August 1943. Seated around the table from left foreground: vice Adm. Lord Louis Mountbatten, Sir Dudley Pound, Sir Alan Brooke, Sir Charles Portal, Sir John Dill, Lt. Gen. Sir Hastings L. Ismay, Brigadier Harold Redman, Comdr. R. D. Coleridge, Brig. Gen. John R. Deane, General Arnold, General Marshall, Admiral William D. Leahy, Admiral King, and Capt. F. B. Royal.
MEMBERS OF U.S. AND BRITISH STAFFS CONFERRING, Quebec, 23 August 1943. Seated around the table from left foreground: vice Adm. Lord Louis Mountbatten, Sir Dudley Pound, Sir Alan Brooke, Sir Charles Portal, Sir John Dill, Lt. Gen. Sir Hastings L. Ismay, Brigadier Harold Redman, Comdr. R. D. Coleridge, Brig. Gen. John R. Deane, General Arnold, General Marshall, Admiral William D. Leahy, Admiral King, and Capt. F. B. Royal.
gust, General Marshall informed his American colleagues that Churchill had told him the previous evening "that he had changed his mind over OVERLORD and that we should use every opportunity to further that operation." Marshall had taken the opportunity to tell the Prime Minister that he could not agree to the logic of supporting the main effort by withdrawing strength from it to reinforce the effort in Italy. In Marshall's view, the British approach to OVERLORD was by "indirection." 25
To counter the British reservations and qualifications, the JCS on 16 August accepted for presentation to the CCS proposals submitted by General Handy. Handy called for the acceptance by the CCS of the TRIDENT decision for OVERLORD-including the definite allotment of forces for it-and of the American proposal of overriding priority for OVERLORD, without reservations or conditions. The JCS decided to withhold  

the second part of General Handy's proposals-alternative recommendations for a radical reversal in U.S. strategic policy -calling for the abandonment Of OVERLORD and placing the main effort in the Mediterranean, in the event the British Chiefs of Staff refused to back OVERLORD wholeheartedly. On this "Mediterranean alternative" scheme, foreshadowed in General Hull's analysis a month earlier, the JCS were noncommittal.26
At the same time, the JCS decided immediately to inform the President, who had not yet arrived at the conference, of the emerging divergences in British and American staff views and especially of their concern over apparent reservations of the British on OVERLORD. General Handy was delegated to fly to Washington at once.27 On 17 August the President arrived in Quebec to lend his support to OVERLORD. By that time-after three days of staff debate-it was already clear that a compromise was in the making and that the U.S. staff would have to accept something less than "overriding priority" for the operation.28
In arguing his case before the President and CCS in plenary session, Churchill declared that he had not favored SLEDGEHAMMER in 1942 Or ROUNDUP in 1943, but he "strongly favored" OVERLORD for 1944. 29 His objections to the earlier operations, he stated, had been removed. He wished all to understand, nevertheless, that the implementation of the OVERLORD plan depended on the fulfillment of certain conditions. One of these conditions was that no more than twelve mobile German divisions were to be in northern France at the time the operation was mounted. Another was that the United States and the United Kingdom had attained definite superiority over the German fighter forces at the time of the assault. He urged that the OVERLORD plan be subject to revision by the CCS in the event that the German strength exceeded the twelve mobile divisions. He also suggested that the Allies keep a "second string to their bow" in the guise of a prepared plan to undertake Operation JUPITER-the invasion of Norway, long a favorite project of his.30
Churchill and General Marshall agreed that an increase in the initial assault force would greatly strengthen the OVERLORD undertaking. The Prime 

Minister called for an addition of at least 25 percent strength. General Marshall pointed out that actually there would be four and one half divisions in the assault rather than the force of three divisions suggested at the TRIDENT Conference. The President seized the opportunity to express his desire, already stated to the JCS, to speed the shipment of U.S. troops to the United Kingdom. General Marshall repeated that the matter was being studied. At the same time, he emphasized to the conferees that the greatest limiting factor on all the prospective Anglo-American operations was the shortage of landing craft. Had landing craft been available, Marshall pointed out, the Anglo-American forces could have already made an entry into Italy. 31
Turning to Mediterranean operations, the British and U.S. military leaders sought to speed the elimination of Italy from the war and decide the course of action to be taken after the prospective landings in Italy. Keeping abreast of current plans of General Eisenhower's staff for two amphibious assaults to be launched early in September- BAYTOWN (across the Strait of Messina) and AVALANCHE (into Salerno Bay)-they took steps to expedite negotiations on Italian peace feelers.32 On the delicate question of how far to go in Italy, the Prime Minister assured the conferees that he was not committed to an advance beyond the Ancona-Pisa line.33 All were agreed on the desirability of capitalizing on the Italian fields as far north as they became available and thereby extending the range of the Combined Bomber Offensive.34
The Prime Minister indicated his hesitancy in placing Anglo-American divisions in southern France as a diversion for OVERLORD and said that he doubted that the French divisions would be capable of undertaking such an operation. Sir Alan Brooke pointed out that there were two routes by which such a diversion might be achieved: a drive west from Italy, if the Allied forces had been able to advance far enough north, and an amphibious operation against southern France. Such a diversion in southern France, he also maintained, would depend on what the German reactions had been. Troops would be landed in southern France only if the Germans had been compelled to withdraw a number of their divisions from that area. In the light of these conditions, the Prime Minister suggested an alternative plan he termed "air-nourished guerrilla warfare" in southern France. This proposal envisaged flying in supplies for French guerrillas at a rendezvous point in the mountains thirty miles inland from the southern French coast. The President went even further and voiced the belief that guerrilla operations could be conducted in south-central France as well as in the Maritime Alps.35 

As for operations in the Balkans, the President indicated his desire to have the Balkan divisions that the Allies had trained, particularly the Greeks and Yugoslavs, operate in their own countries. He expressed the belief that it would be advantageous if these Balkan divisions would follow-up and harass the Germans, should the latter decide to withdraw from the Balkans to the line of the Danube. The Prime Minister suggested that commando forces could also operate in support of the guerrillas on the Dalmatian coast. Neither the British nor the American leaders expressed an interest in offensive land operations by the United States and Great Britain in the Balkans.36
A persistent note pervaded the discussion of the American delegates-the fear of draining strength and means away from the cross-Channel operation and the consequent desire to restrict Mediterranean operations. How to keep the war in the Mediterranean a limited one contributing to OVERLORD and early victory over Germany was the problem. In any event, whatever measures were undertaken to eliminate Italy, establish bases on the mainland, seize Sardinia and Corsica, and launch an operation in southern France in conjunction with OVERLORD, should be carried out with the forces allotted at TRIDENT. To such limits the British raised objections. They argued strongly the need for more leeway in allocating resources in order to insure the success of the Mediterranean operations-all the more important now to pave the way for OVERLORD. Hence, they saw great danger in accepting rigid commitments for the Mediterranean-a straight jacket likely to jeopardize the Allied cause in the whole European-Mediterranean area.
Staff differences on the question of Mediterranean commitments were themselves symptomatic of more basic and lingering divergences in European strategy-on the role of preparatory operations and the timing of the main blow. Back of these divergences lay the even more fundamental differences in approach to strategy-the claims of waging attritional warfare versus those of concentration in a selected area. Though a definitive reconciliation of strategic methods and theories might be beyond the scope of the staffs assembled in conference, the practical issue on which the larger divergences came to settle-the question of Mediterranean commitments for the following year posed a problem for immediate compromise.
In the course of discussing operations in the European-Mediterranean area, British and American military leaders also considered the possibility of an emergency return to the Continent. For this a new plan was at hand-Plan RANKIN. Prepared by General Morgan's COSSAC staff, it provided for an emergency return to the Continent in the winter of 1943-44, or early spring of 1944 before the target date for OVERLORD.37 Just as OVERLORD represented

the culmination of the thinking on decisive cross-Channel operations that had been embodied in ROUNDUP and ROUNDHAMMER, so RANKIN signified a new version of the SLEDGEHAMMER concept of an opportunistic operation. Added weight was given to the urgency of this planning in view of the President's expressed interest in it at QUADRANT and particularly in the light of his expressed desire at the conference that the "United Nations troops . . . be ready to get to Berlin as soon as did the Russians." 38
Plan RANKIN set forth three contingencies for the emergency return: Case A: a substantial weakening of German resistance; Case B: a withdrawal of the German forces from occupied countries; or Case C: unconditional German surrender.39 The COSSAC planners considered that any of these three contingencies might evolve from the continuation of such encouraging developments as the German reverses on the Eastern Front, the growing threat to Germany in Italy and the Balkans, the setback to the German submarine campaign, and the increasing Allied air offensive. The COSSAC planners shaped their proposals for the emergency operations on the considerable alteration of the Allied strategic situation in the European-Mediterranean area since the incorporation of the SLEDGEHAMMER concept into the War Department BOLERO-ROUNDUP plan in the spring of 1942.
In Case A of Plan RANKIN the COSSAC planners set as the objective a lodgment on the Continent from which the U.S. and British forces could complete the defeat of Germany. The assault area was to be the same as that for OVERLORD-the Cotentin-Caen sector of northwestern France. If there was a sufficient disintegration in morale and strength of the German armed forces, an operation against organized opposition could be undertaken in January or February 1944 to capture the Cotentin Peninsula. Alternatively, a modified OVERLORD would be put into effect in March or April 1944. In either case, the COSSAC planners believed, the port of Cherbourg would have to be captured within the first forty-eight hours to provide adequate maintenance. In Case A of Plan RANKIN, as in OVERLORD, diversionary operations in the Pas-de-Calais area, and from the Mediterranean would probably be necessary. In Case B of Plan RANKIN the COSSAC planners also called for a lodgment on the Continent from which the Allies could complete the defeat of Germany. The first place of entry for the main Allied forces in this event was to be Cherbourg.
In Case B, moreover, substantial Allied forces were to be sent from the Mediterranean to occupy the ports of Marseille and Toulon and move northward as required.
In Case C Of RANKIN the COSSAC planners stated that the object was to occupy as quickly as possible areas from which the Allies could enforce the terms of unconditional surrender imposed by their governments on Germany. Under the general direction of the Supreme Allied Commander, France, Belgium, and the Rhine Valley from the Swiss frontier to Duesseldorf were to be under  

the control of the U.S. forces, with British representation in the liberated countries; Holland, Denmark, Norway, and northwest Germany from the Ruhr Valley to Luebeck were to compose an area under the control of the British forces, with American representation in the liberated countries.
The COSSAC planners concluded that the forces allotted for OVERLORD should also be considered available for RANKIN. In all three of the contingencies they emphasized the importance of rehabilitating the liberated countries. They therefore recommended that the United States and the United Kingdom lay down policies to govern the establishment of military governments in enemy territory to be occupied by Allied troops and of national administrations in the liberated Allied territories. Nuclei of combined Anglo-American civil affairs staffs in London for Germany and for each Allied and friendly country within the sphere of the Supreme Allied Commander should be established.40
In discussing RANKIN, Sir Alan Brooke stated the hope of the British Chiefs that fewer forces might be used for occupation purposes than set forth in the plan. Admiral Leahy replied that the JCS shared this view. The U.S. Joint Chiefs recommended that RANKIN be approved in principle and that it be continuously reviewed. 41 The CCS approved these suggestions, noting that, in line with the plan, the U.S. Joint Chiefs would appoint a commanding general, staff, and headquarters for the U.S. Army group in the United Kingdom.42
Compromises and Agreements
The operations in 1943-44 for the defeat of the Axis Powers in Europe, approved by the CCS, the President, and the Prime Minister at QUADRANT, represented another compromise of British and American views on strategy. The conferees agreed that Operation OVERLORD was to be the main Anglo-American effort in Europe with a target date of 1 May 1944, approved the outline plan of General Morgan for Operation OVERLORD, and authorized him to proceed with preparations. Because of differences in British and American views of the relationship between OVERLORD and Mediterranean operations, the U.S. delegation had to yield on its desire for overriding priority for OVERLORD and accept a compromise statement that, in the event of a shortage of resources, avail-' able means were to be disposed and utilized with the "main object" of insuring the success Of OVERLORD.43 The American delegation also accepted, as 3n added qualification to its original proposal that operations in the Mediterranean area be conducted with the forces allotted at TRIDENT, the clause upheld by the British representatives-"except insofar as these may be varied by decision of the Combined Chiefs of Staff." 44

The same proviso was attached by the British to the planned return of the seven divisions from the Mediterranean to the United Kingdom for OVERLORD, though Marshall had fought hard for a decision without strings. The conferees also accepted the British proposal that in the event OVERLORD could not be executed, JUPITER should be considered as an alternative and called for plans to be developed and kept up to date for such an operation. All agreed that the Combined Bomber Offensive (POINTBLANK) was to remain in the "highest strategic priority" and was to. be extended from all suitable bases-particularly from Italy and the Mediterranean -as a prerequisite for OVERLORD.
As for Mediterranean operations, the conferees agreed on the basic outlines of the three phases of operations in Italy that the JCS had suggested in their proposals to the CCS before QUADRANT. The first phase, as accepted at QUADRANT, called for the elimination of Italy from the war and the establishment of air bases in the Rome area and, if possible, farther north.45 For the moment at least, these general objectives in Italy represented a meeting ground between the aims of the Americans and the desires of the British.
The second phase involved, as the JCS had recommended, the seizure of Sardinia and Corsica. In this connection the delegates decided to request General Eisenhower to examine the possibilities of intensifying subversive activities on the islands in order to facilitate entry into them. This action stemmed largely from the American staff's urging, especially for Sardinia. The JCS had themselves been persuaded to make this proposal to the CCS by Generals Marshall, Handy, and Wedemeyer in the course of staff discussions during the conference. In the meeting of the JCS on 19 August 1943, Generals Marshall, Handy, and Wedemeyer had argued that, in view of the shortage of landing craft available to General Eisenhower, and in the light of the opportunity to test the effectiveness of the Office of Strategic Services organization, "fifth column activity" should be undertaken by the OSS on Sardinia.46
In the third phase of operations in Italy, the conferees accepted the JCS provision for maintaining constant pressure on German forces in northern Italy and creating conditions favorable for the eventual entry of Allied forces, including most of the re-equipped French Army and Air Force, into southern France. Also in keeping with the American proposal, offensive operations against southern France were to establish a lodgment in the Toulon-Marseille area and exploit northward in order to create a diversion in connection with  

OVERLORD. Omitted, however, was the qualifying phrase, "with available Mediterranean forces" that the JCS had sought.47 In line with the proposals of the President and Prime Minister, it was agreed that "air nourished guerrilla operations" in the southern Alps would be conducted if feasible.48 Also approved was the rearmament of French units up to and including eleven divisions by 31 December 1943. As a result of the deliberations at the conference, the CCS sent a directive to General Eisenhower calling for his appreciation and outline plan on operations in southern France, to be submitted to the CCS by 1 November 1943.49 In the preparation of the plan, Eisenhower was to consult with the Supreme Commander of the cross-Channel operations (whoever might be appointed) or his chief of staff so that his planning could be correlated with the requirements of OVERLORD.
With little debate, the delegations at QUADRANT rejected the idea of offensive ground operations by the United States and the United Kingdom in the Balkan area. Operations in that area were to be limited to supplying Balkan guerrillas by air and sea, minor commando raids, and bombing of strategic objectives. In keeping with the by then long familiar concern of the JCS for safeguarding the lines of communications in the Mediterranean, appropriate Allied forces were to be deployed in northwest Africa so long as the possibility of a German invasion of the Iberian Peninsula remained. From the military point of view, the time was not considered right for Turkey to enter the war, but the United States and Great Britain were to continue to supply such equipment to Turkey as they could spare and the Turks could absorb.50
Further measures were to be undertaken in the Atlantic to strengthen operations against the U-boats. Especially attractive was the possibility of using the Azores as a base for intensified sea and air operations and for the development of an air-ferry route.51 During the discussions, the British had indicated that negotiations over the Azores with the Portuguese Government, undertaken by the British Foreign Office in consultation with the British Chiefs of Staff, were approaching a conclusion. The Portuguese had agreed to the entry of a small British force into the Azores on 8 October (Operation ALACRITY. The U.S. delegation was assured by the British Chiefs of Staff that, upon gaining entry into the Azores, the British would seek to make arrangements for U.S. aircraft to use the airfields in the Azores as a base of operations and in transit.52
Finally, on the President-Prime Minister level, a significant agreement was reached at Quebec on the question of   

command for the projected coalition effort in the European-Mediterranean area. Earlier, the two leaders had agreed that, since the United States had the African command, it was but fair that the commander of the cross-Channel operation be British. With Presidential agreement, Churchill had gone so far as to nominate General Brooke, Chief of the Imperial Staff, for the post and early in 1943 had so informed him. The logic of events, however, now compelled a change. Churchill has since recorded:
 . . . as the year advanced and the immense plan of the invasion began to take shape, I became increasingly impressed with the very great preponderance of American troops that would be employed after the original landing with equal numbers had been successful, and now at Quebec, I myself took the initiative of proposing to the President that an American commander should be appointed for the expedition to France. He was gratified at this suggestion, and I dare say his mind had been moving that way.53
As already observed, the President's mind indeed had been moving in that direction. An American officer, they therefore agreed, would command OVERLORD, and a British commander would take over in the Mediterranean, the time for the change to depend upon progress in the war. The way was thus cleared, as Stimson had strongly urged before the conference, for an American leader to take over the command of the cross Channel operation. Whether that prize would fall to Marshall-as Stimson had hoped-or whether Marshall, like his British counterpart, would be passed over by the force of circumstances, remained to be seen.
Discussion on the War Against Japan
The British had originally intended to bypass the war against Japan at QUADRANT, but consideration of it consumed as much time and effort as it had at Casablanca and TRIDENT. Fear that the Pacific conflict might degenerate into a long war of attrition or stalemate made the United States anxious to spell out the future course and timing of operations. British reluctance to commit Allied resources too heavily in the Pacific until after the Germans collapsed was understandable, but could not withstand the growing American pressure for accelerated action. Two basic questions demanded consideration: selection of the main line of approach to the Japanese homeland and Great Britain's role in the Pacific after Germany was defeated. Exploration of these vital problems at QUADRANT brought to light areas of Anglo-American disagreement that would require still further examination.
As usual in the midwar international conferences, Admiral King took the lead in presenting the U.S. case for the Pacific war-an acknowledgment by the Army of the Navy's primary interest in the Pacific. Nevertheless, in backing the case for the Pacific position, Marshall was not unmindful of the Army's interests. Ever conscious of the need to link Pacific and European strategy, he sought, insofar as possible, to safeguard the plans and projects of Generals MacArthur and Stilwell for their respective theaters.
The Search for a Long-Range Plan
The basis for discussion of the Pacific war at QUADRANT was the over-all plan  

produced by members of the Combined Planning Staffs on the eve of the conference.54 The initial reaction of the JCS to this combined effort had been unenthusiastic, for they agreed generally that it overlooked many possible elements that might shorten the conflict.55 The American Chiefs' distaste for any plan that might prolong the war until 1947 or 1948 was keynoted by King and Marshall in the second meeting of the CCS at Quebec. King told the Combined Chiefs that the current lack of means in the Pacific to carry out operations directed toward Rabaul CARTWHEEL) was occasioned by Allied failure to consider the war against the enemy powers as a whole. Reverting to the mathematical approach of which he was evidently quite fond, he declared that if 15 percent of all Allied resources were now deployed against Japan, then an increase to 20 percent, or just 5 percent, would make one-third more resources available. The resulting decrease in resources available in the European war would amount to a mere 6 percent of the total. King and Leahy both thought it was most important to plan how to transfer the bulk of Allied forces from Europe to the Pacific-Far East once Germany was defeated. Marshall then went on to point out that not only were all operations in the Pacific related to those in Burma, but also affirmed that it was
. . . essential to link Pacific and European strategy. Movements of ships from the Mediterranean must take place in the next few days if operations from India were not to be delayed, and a decision must be taken. It was important that no time should be lost in agreeing on a general plan for the defeat of Japan since the collapse of Germany would impose the problem of partial demobilization and a growing impatience would ensue throughout the United States for the rapid defeat of Japan.56
Since the British would be faced with an even greater demand for demobilization as a result of their long participation in the war, Churchill and the British Chiefs of Staff were perfectly amenable to the early completion of a general plan for the war against Japan. They realized that the future role of Great Britain in the struggle would vitally affect British demobilization, particularly that of the ground forces. Since British land forces would probably be substantially de-  

creased after the defeat of Germany, the British desired to base their main contribution to the war against Japan on air and naval units. They hoped that Japan might be defeated by sea and air attack alone, but they agreed with the Americans that, for planning purposes, an invasion by land forces should be assumed as ultimately necessary.57
Using the proposed over-all plan as a basis, staff planners moved to define for the consideration of the CCS the points at issue and those on which the Americans and British agreed. American planners condemned the proposed schedule of operations in the Pacific as being too slow and suggested that the tempo be pitched to defeat Japan within twelve months of the fall of Germany. Although their British counterparts agreed on acceleration, they would not accept the twelve-month limit. Both staffs felt that the reorientation of forces toward the Pacific should be started about four to six months before the fall of Germany. They also agreed on an American advance toward Japan via the Central and Southwest Pacific and possibly the Northwest Pacific, and on a British drive via the Strait of Malacca and South China Sea, together with the development of a U.S. line of supply to China through Burma.58
Perhaps the sharpest difference of opinion between the British and U.S. planners revolved around the sequence and timing of the operations to take south Burma and Singapore. The Americans believed that south Burma should be cleared right after north Burma and visualized a target date of November 1944 for tile beginning of south Burma operations. The British, on the other hand, maintained that after the seizure of north Burma, south Burma should be bypassed until November 1946 and that an effort to take Singapore should be made in 1945.59  Wedemeyer advised the JCS that tile long period of inactivity between the close of north Burma operations (May 1944) and the initiation of the Singapore campaign (March 1945) would result in too great a time lag. Furthermore, operations in south Burma would provide more direct aid to China.60
The conflicting views between the two staffs were further complicated by Churchill. He sided with the United States in disapproving a Singapore expedition in 1945, since he did not think that the period from May 1944 to March 1945 should be a time of inaction. Instead, he wanted a move to take the northwestern tip of Sumatra-his favorite Far Eastern operation, which he pictured as the TORCH of the Indian Ocean and possibly of as great strategic significance as the Dardanelles operation of 1915. The Prime Minister received little comfort from the President in this direction, for Roosevelt looked at the problem from another angle:
The position occupied by the Japanese might be compared to a slice of pie, with Japan at the apex, and with the island barrier forming the outside crust. One side of the piece of pie passed through Burma, the other led down to the Solomons. He  

quite saw the advantages of an attack on Sumatra, but he doubted whether there were sufficient resources to allow of both the opening of the Burma Road and the attack on Sumatra. He would rather see all resources concentrated on the Burma Road, which represented the shortest line through China to Japan. He favored attacks which would aim at hitting the edge of the pie as near to the apex as possible, rather than attacks which nibbled at the crust. 61
The JCS were willing to forego any definite decision on the south Burma-Singapore question until the next conference, but pressed for the acceptance of the twelve-month target date. In support of their belief in a shorter war, they presented an AAF plan for the defeat of Japan, based upon the use of the new very long range (VLR) bomber-the B-29 Super fortress-which was due to become available in quantity in 1944. The 1,500-mile tactical radius of this new weapon would allow it to reach most of the important targets in Japan proper, if it operated from bases in the Changsha area in China, and its bomb load of ten tons would permit greater destruction to be inflicted by each plane. Since the Air plan was so recent that even the U.S. staff had not had a chance to study it carefully, the CCS referred it to the Combined Staff Planners for close consideration. In as much as use of Chinese air bases was part of the plan, the JCS recommended that the TRIDENT decisions regarding China's importance as an ally be reaffirmed and the capacity of air route to China be expanded. In the meantime, studies could be made of the possibility of operations at Moulmein in Burma and on the Kra Isthmus of the Malay Peninsula to isolate Rangoon-the gateway to north Burma and the Burma Road-and to facilitate the capture of Singapore. As to the Pacific, they urged that the U.S. plan for operations in 1943-44 be accepted in toto.62
The British met the American proposals more than halfway. Since Japan depended so heavily upon airpower, naval strength, and shipping to maintain its position, the CCS decided that greater emphasis should be placed upon attrition and that greater use should be made of the Allied air forces for this purpose. Through the build-up of the air route to China, the employment of lightly equipped, air-supported jungle troops, and the use of special equipment such as HABAKKUKS and artificial harbors, increased advantages might accrue to the Allies.63 The British accepted the twelve-month target date for future planning, but on the condition that the reorientation of forces toward the Pacific  

should proceed as soon as the German situation, in the opinion of the CCS, would so allow. Forces for the operations in the Pacific would be provided by the United States and those for the prospective operations in the Southeast Asia area by the British, except for special types available only to the United States. In the Pacific, as customary at the conferences, the U.S. schedule of operations was approved. Operations in north Burma would be carried out in February 1944, but the need for the amphibious landings at Akyab and Ramree would be investigated further. The CCS directed that studies be made on the south Burma-Singapore and Malaya-Sumatra operations. They also decided to examine fully the possibilities of developing the air route to China on a scale that would permit the use of the bombers and transports available after the defeat of Germany.64
The anxiety of the British to assure themselves of a proper place in the later stages of the war-an issue that led to heated staff discussion-was assuaged by assigning to the Combined planners the task of investigating further operations in which the British would play the major role. Evidently the Prime Minister was well satisfied that any doubts regarding Britain's desire to share in the final defeat of Japan had been effectively removed at QUADRANT-65 To the Americans, however, the perplexing problem of how and where to use the British naval and air forces in an area where bases were few and logistical difficulties many required answers that, at the moment, they felt in no position to provide.66
Although only certain features of the over-all plan were adopted by the CCS, there were several developments of especial significance. The acceptance of the twelve-month target date promised to shorten the war and to alter radically the timing and possibly the sequence of forthcoming operations. Similarly, the free hand given to the United States in the Pacific to conduct Southwest and Central Pacific offensives simultaneously promised to move the war into higher gear. The main portions of the over-all plan agreed upon by the CCS were essentially the short-term phases indicating immediate directions without committing the Allies to any definite ultimate roads. The relative importance of the different approaches to Japan-land, sea, and air-and the selection of the main line of offense for an invasion of Japan, if this should prove necessary, had not been considered. It remained to be seen whether this inability to determine conclusively where the weight of the Allied drive should be placed and which operations should be held as subsidiary foreshadowed the same prolonged deliberations on Pacific strategy as had marked the planning of European strategy.
Pacific and Far Eastern Operations 1943-44
The American decision to open up a new line of advance in the Central Pacific with the Gilberts-Marshalls operations in the fall and winter of 1943-44 was received by the British without protest. This increase of pressure upon the  

Japanese from the east, which would utilize the expanding U.S. naval forces profitably, made a favorable impression upon Churchill, who disliked the idea of fighting difficult land campaigns in Burma and China. Nevertheless, the British Chiefs at first did question the necessity for pressing forward in the Central Pacific and the Southwest Pacific with equal vigor and suggested that the New Guinea phase be limited to a holding operation, while the main effort took place through the mandated islands. Thus, they observed, resources might be released for OVERLORD.67
Admiral King attacked this proposal at once, holding that if there were resources that could be spared from the Southwest Pacific, they should be sent to the Central Pacific. He stated that he himself considered both advances essential, and Marshall pointed out to the British that the troops and resources for the New Guinea operations were already in or en route to the theater.68 The British did not press the point further.
Since the British seemed to have no other objections to the American plan, the CCS approved the U.S. proposals to proceed successively through the Gilberts, Marshalls, Ponape, Truk, and the Palaus to the Marianas in 1943-44. Consideration was also to be given to operations against Paramushiro in the Kurils. In the Southwest Pacific, eastern New Guinea as far as Wewak, the Admiralty Islands, and the Bismarck Archipelago were to be seized. As foreshadowed before the conference, Rabaul was to be neutralized rather than captured. 69 With these operations accomplished, a further move westward along the New Guinea coast to the Vogelkop Peninsula was to be made in step-by-step, airborne-waterborne advances.70 The decision to neutralize Rabaul marked the first official pronouncement of a policy of bypassing strong centers of resistance and foreshadowed the gradual replacement of the earlier conservative step-by-step method of operations in the Pacific.
In the CBI, the long wrangle over the importance of China and the value of capturing all of Burma came in for a full show of attention. The British appeared perfectly willing to carry out the north Burma campaign, but balked at the need for next clearing south Burma and were disinclined to go through with the amphibious landings at Akyab and Ramree. In fact, on several occasions the Prime Minister flatly opposed any commitment to conduct the latter and warned his Chiefs of Staff against taking any decision that he might later have to overrule.71
The JCS and the President had agreed before the conference that the Burma campaign should not be delayed, and Roosevelt had even gone so far as to mention the substitution of American resources and ships and possibly two U.S. divisions, should the British seek to withhold forces and supplies for use in their Mediterranean ventures.72 Marshall and 

King had then been particularly disturbed by the unilateral action taken by the British in the form of their "stand-fast order" preventing the scheduled departure of resources from the Mediterranean to the CBI. At the conference the JCS exerted considerable pressure on their British colleagues in support of the land campaign to take all of Burma. They pointed out the necessity of support for China, which ultimately would provide the necessary facilities for the huge air forces to be released for use against Japan after the defeat of Germany. To Marshall there appeared to be four issues to be decided: (1) the value of Chinese troops to future operations; (2) the likelihood that the existing Chinese Government might fall if there were no sustaining action; (3) the possible Japanese reaction to heavy air attacks; and (4) the need for a port on the China coast. The road to China would not be opened, he felt, by a Sumatra operation, but only through the capture of all of Burma, including Akyab and Ramree.73
The British were not convinced that the reconquest of south Burma was a prerequisite to assisting China, since they believed that the air route could be developed to the degree that it could supply most of the numerous Allied air forces that would become available for the CBI after Germany's defeat. Marshall agreed that, in view of the great difficulties of undertaking ground operations in the CBI, full advantage should be taken of the Allied air superiority. In the matter of Akyab and Ramree operations, however, the British were reluctant to make any decisions.74
In the face of this indisposition on the part of the British Chiefs, the intransigence of the Prime Minister, and the lack of Presidential enthusiasm for Akyab and Ramree operations, Marshall admitted to the JCS that he himself considered the plan for the landings unrealistic, since the British seemed unable to produce enough efficient troops to ensure success. The possibility of withdrawing some of the better-led Indian divisions from the Mediterranean was considered by the JCS, but the shipping implications for OVERLORD and the Pacific made such a move of doubtful value.75
Nature took a hand in the fate of Burma operations at this juncture, for Auchinleck reported that severe floods in Assam might force the cancellation of either the Ledo or the Imphal advance, and perhaps both. Even without the interference of nature, the limited capacity and inefficient operation of the Assam railroad promised to make difficult logistical support of the land campaigns in Burma. The strains on the Assam line of communications spurred the CCS decision to expand the railway capacity, and to lay gasoline pipelines between Calcutta and Kunming.76
The possibility of tonnage deficiencies in Assam led the British Chiefs of Staff to suggest that a policy decision be made on the priority of resources for undertakings in the CBI. They felt that the priority system should not be a rigid one. Nevertheless, they favored putting the

GENERAL ARNOLD WITH LORD LOUIS MOUNTBATTEN, Quebec conference, 20 August 1943.
GENERAL ARNOLD WITH LORD LOUIS MOUNTBATTEN, Quebec conference, 20 August 1943.
main stress on north Burma operations, so necessary to establish land communications with and to improve and secure the air route to China. This primary emphasis on north Burma operations would serve as a guide for the supreme commander of the Southeast Asia Command -still to be appointed-. The commander would, of course, also have to keep in mind the importance of long-range development of the Assam line of communications, so fundamental for all CBI undertakings. Although the JCS recognized that the proposed priority might affect adversely the supply delivery to China over the Hump, they accepted the British recommendations.77
Logistical difficulties involved in supporting both the air effort and the projected ground offensive in China, which the TRIDENT decisions had failed to appraise adequately, had to be faced more realistically at QUADRANT. Despite British and Chinese preference for more emphasis on the air build-up, the end result was an apparent reversal of the priorities set up at TRIDENT. After the enthusiasm for the Chennault air plan that had been so manifest at TRIDENT, this volte-face seemed to be a major change in policy in the CBI. How strictly the reversal would be adhered to in the future remained a matter of conjecture, for although the unimpressive showing made by the air forces in China during the summer may have influenced the President to give the Stilwell-Marshall-Stimson school a chance to demonstrate the efficacy of a ground approach to aiding China, he gave no clear indication that he had completely abandoned his predilection for the Chennault plan.
The Southeast Asia Command
Extensive negotiations after TRIDENT between the British and Americans over the organization and leadership of the Southeast Asia Command had brought no solution to this vexing problem, although they served to underline the  

points of issue.78 To stave off another postponement of the campaign in Burma, the U.S. Chiefs of Staff had tried, without success, to secure a settlement of the command problem before the conference convened.79
The opportunity to discuss the problems face to face allowed the leaders on both sides to modify their positions and seek satisfactory compromises. Marshall, after a meeting with Churchill, admitted to the JCS that the Prime Minister had some ground for complaint on the absence of information from SWPA. In fact, to remedy the situation Marshall consented to the Prime Minister's having a personal representative on General MacArthur's staff. The Chief of Staff did not desire a similar situation in reverse to arise in the proposed southeast Asia command, but felt that a more moderate version of the MacArthur headquarters might be established that would avoid its defects and yet still be acceptable to the British. Changes were suggested that brought British proposals for a command closer to the Eisenhower pattern, while allowing operational control to remain under the British. 80
The British in turn dropped their brief for Sir Sholto Douglas and finally proposed Lord Louis Mountbatten, then British Chief of Combined Operations, for the post. Churchill found the President and General Marshall "very keen" on Mountbatten's appointment. His record and personality bespoke the kind of young, energetic, and offensive-minded leader that the Burma operations would require.81
The most difficult administrative feature of the Southeast Asia Command concerned the appointment of Stilwell as Deputy Supreme Allied Commander. The intricate Sino-American command system, which already had Stilwell serving as chief of staff to the Generalissimo, commander of U.S. Army forces in the CBI, and lend-lease administrator, required delicate handling. His responsibilities to the Generalissimo and the JCS were now to be complicated further by a responsibility to Mountbatten and the British Chiefs of Staff.82 As Stilwell himself expressed it:
The command setup is a Chinese puzzle with Wavell, Auk [Auchinleck], Mountbatten, Peanut [Chiang Kai-shek], Alexander and me interwoven and mixed beyond recognition.83
This complex situation came about because the Allies recognized that Chiang would not permit Chinese forces to serve directly under British command and that the only way to ensure Chinese co-operation in the north Burma campaign would be to make Stilwell the middleman or point of contact and liaison. As Deputy Supreme Commander, he would control the Chinese forces engaged in the operations and co-ordinate the efforts of the Fourteenth and Tenth Air Forces so that they could also play their parts  

in Burma. Integration of forces and fulfillment of his varied responsibilities would give Stilwell a task hard to perform successfully, but in view of the already entangled command setup, it appeared to the Americans to offer the only practicable solution. With some doubt and hesitation over the ability of Stilwell to cope with the added duty, the British accepted the American proposals.84
Geographically, the Southeast Asia Command would embrace Ceylon, Burma, Thailand, Malaya, and Sumatra, all but the first then under Japanese dominion. Under Mountbatten and Stilwell would be British commanders in chief of the ground, naval, and air forces. The Supreme Commander would have direct access to the British Chiefs of Staff on all matters, since his administrative and logistic support must be based upon India. The Viceroy of India had authority to settle any question of priorities, but Mountbatten could appeal the Viceroy's decisions to the British Chiefs. General jurisdiction over strategy in SEAC and the allocation of U.S. and British resources of all kinds between China and SEAC would remain under the CCS, but the British Chiefs of Staff would exercise operational jurisdiction and would be the executive agent for transmitting all instructions to the Supreme Commander. To provide for exchange of information and intelligence co-ordination in India and in SEAC, a Combined Liaison Committee would be established in New Delhi.85
Having resolved the Anglo-American differences, the ticklish question of presenting the combined agreement to the Generalissimo as a proposal rather than as an accomplished fact required the utmost tact and diplomacy. Since Thailand originally had been included in the China theater, its transfer to SEAC would have to be explained to Chiang gracefully in order to prevent any Chinese loss of face. Colonel Roberts of the Operations Division suggested that Dr. Soong would be the ideal intermediary to break the news. To emphasize the importance attached by the United States to the need for full Chinese cooperation with SEAL in the coming campaign, Marshall personally took Soong aside, after the CCS had informed him in general terms of their proposals, and underlined the necessity for Soong to convince the Generalissimo of the importance of the arrangement.86
If Chiang would accept the boundaries of SEAC and co-operate with Mountbatten and Stilwell, the prospect for successful Burma operations appeared bright. Besides Mountbatten, the British had brought with them Brigadier Wingate, leader of the daring British raid behind the Japanese lines of communications in Burma earlier in 1943. Wingate's imaginative and aggressive spirit captured first the favor of the Prime Minister and later the enthusiastic support of Marshall and the other American leaders.87
Wingate explained to the CCS the tactical employment of his long-range jungle troops, which operated behind  

the enemy's main forces disrupting communications and conducting guerrilla activities until the main advance could reach them. The British now proposed to organize six brigades of lightly equipped, mobile, air-supported troops similar to the Wingate groups. The plans for the units would be elastic and open to alteration in the light of the enemy reaction.88 Marshall was convinced that Wingate was a "best bet" and suggested that the United States might add a long-range column of its own to work under him in order to encourage the Chinese to act more favorably toward the project.89
Injection of new blood into the Burma operation held forth a possibility that the campaign might actually be launched during the coming dry season, and the U.S. Chiefs of Staff were ready to help make the possibility a reality in any way that they could. First, however, would come the task of eliminating the attitudes in India-Burma that had produced nothing but delays and excuses during the previous months. New leadership would make little progress until the old attitudes had been supplanted.
Emerging Strategic Patterns
By the close of QUADRANT, War Department planners could point to encouraging signs of progress in planning against Japan as well as against Germany. A mere surface appraisal of the record of accomplishments at the conference in regard to the Japanese would have been misleading. More or less perfunctory approval by the British of the American 1943-44 plan of operations in the Pacific, long overdue establishment of SEAC, lack of decision on subsequent operations in southeast Asia and China after north Burma, de-emphasis of the Chennault air plan after a short trial period, and failure of the CCS to evolve a long-range plan by no means told the whole story. Several essentially new features had entered the planning picture. Paramount among these was what might be called the new urgency-the recognition that the war on all fronts must be pushed forward faster, lest a stalemate develop and home morale bog down. This gathering momentum was manifested at QUADRANT in the fixing of the twelve-month target date for the defeat of Japan once Germany was beaten. The target date for ending the European conflict-October 1944-embodied in the OVERLORD plan accepted at QUADRANT, gave promise of clarifying one of the basic unknowns in over-all strategic planning against Japan -timing-and thus introduced a controlling objective hitherto lacking in that planning. Secondly, the evident intention of the British to get north Burma operations under way was demonstrated by the appointments of Mountbatten and Wingate. For the first time there was confidence that Burma operations would be launched. Going further, the British had served clear notice of their desire for an eventual "full and fair place in the war against Japan"-to borrow Churchill's phrase.90 In the third place, the American drive to improve the Assam line of communications and to expand the air route to China gave promise of a vastly increased operational effort both in Burma and in China.  

Lastly, the projected introduction of a new weapon, the B-29, bade fair to revolutionize strategic concepts on the approach to Japan. As ranges of strategic weapons increased and distances shrank, bold new plans for shortening the war might be formulated and carried out. The promise of increased activity in the Pacific and the CBI, coupled with the prospective employment of fast aircraft carriers and very long range bombers, made the outlook for the war against Japan decidedly more encouraging. Emphasis on the strategic offensive against Japan, signalled by the U.S. staff first at Casablanca and then at TRIDENT, was thus confirmed all the more strongly at Quebec, and at least some of the components for an eventual strategic synthesis for defeating Japan appeared to be taking more positive form.
Perhaps the single greatest failure of the conference in respect to ,the war against Japan, from the viewpoint of the Army planners, was the inability to evolve a long-range strategic pattern. Aside from the twelve-month theory, an agreed blueprint on the "how," "when," and "where" of the ultimate defeat of Japan had still not been formulated. The question of whether to rely on bombardment and blockade or the invasion of Japan proper was left open. Selection of the main route was also postponed. That the Army welcomed whatever new plans and "modern and untried" methods of warfare-revolving about carriers, B-29's, the new policy of neutralization, and inventions and techniques still in the experimental stage-that might keep the Pacific war progressing without absorbing Army troops and resources needed for defeating Germany is obvious. But there could be no certainty over whether the newly emerging strategic elements for speeding the Pacific offensive would eventually sup plant the need of an over-all plan against Japan or, in the absence of such a plan in the long run, might not themselves add to the "suction pump" pressures of the dynamic war in the Pacific.
Though there might still be reason after QUADRANT for Army planners to question whether planning in the war against Japan had been linked firmly enough to European strategy and whether strong enough barriers had beer erected to keep the Pacific conflict a limited war, there appeared to be a more solid basis for believing that the trend to the Mediterranean was finally coming under control. Hitherto, that trend lead been one of the most important keys to grand strategy-threatening, in the opinion of the American staff, not only any major cross-Channel operation but Burma and Pacific undertakings as well, The U.S. staff had come to Quebec seriously disturbed by the British "stand fast" order in the Mediterranean and by the likely effect of Presidential overtures and British predilections for postponing the departure of the seven divisions from the Mediterranean to the United Kingdom.91 To keep the Mediterranean war a limited one contributory to the success of OVERLORD and early victory over Germany was the fundamental aim of the American staff. By the close of QUADRANT the Army planners could point to  

a number of steps taken to realize this goal.
In the strategic synthesis emerging from QUADRANT provision was made for drawing together a major cross-Channel operation, the Mediterranean undertakings, and an extended Combined Bomber Offensive from the Mediterranean as well as the United Kingdom, and weaving them into an over-all scheme for defeating the Western Axis Powers. The new synthesis seemed to contain the formula-for which the War Department staff had been searching since the diversion from BOLERO-ROUNDUP to the Mediterranean in TORCH-that would retain the primacy of the cross-Channel operation from the United Kingdom, define the role of subsequent Mediterranean operations in relation to that main effort, and use to the fullest extent the potentialities of the Combined Bomber Offensive.
The War Department staff could take comfort that the new synthesis was developing to a considerable extent in line with their thinking. They could point to the fact that the central element in the new strategic pattern about which the other factors were woven was Operation OVERLORD. The adoption of that plan marked an important milestone in the campaign that had been waged by the American staff since General Marshall had taken the BOLERO-ROUNDUP proposals to London in April 1942. After long debate and many decisions and revisions on the highest levels, the United States and Great Britain had at last agreed upon an outline of a plan for concentrating forces for a major cross-Channel operation for the early and decisive defeat of Germany. Steps had been taken, moreover, to tie together the Mediterranean and OVERLORD operations in order to create favorable conditions for launching and supporting OVERLORD. The stress by British and U.S. leaders on central and western Mediterranean operations in support of OVERLORD and their disavowal of interest in offensive ground efforts in the Balkans and the eastern Mediterranean seemed to be added proof of their firm resolve to make the main effort against the Axis citadel from the United Kingdom in the spring of 1944. Limits in the form of compromise agreements-qualified as they were-had been set on priorities, magnitudes, and timing for operations in the European-Mediterranean area. Along with authorization for the definite allocation of forces in the approved cross-Channel operation-twenty-nine divisions for target date 1 May 1944 -went, therefore, definite restrictions on future Mediterranean advances. The pattern of strategy agreed upon at Quebec pointed the way more clearly than before to the final halting of the diversionary trend from BOLERO to the Mediterranean that had begun with TORCH.
In retrospect, QUADRANT was a critical conference in the evolution of Anglo-American strategy in the war against Germany. If Casablanca represented for U.S. strategists initiation in planning for the offensive phase of coalition warfare, and TRIDENT a halfway mark, QUADRANT was the beginning of the final stretch. The results showed that the American staff had made marked progress in preparing and presenting its case and was mastering the art of military diplomacy. Most encouraging from the Army point of view was the fact that at this conference the President had held through and backed his Army Chief of  

Staff on European strategy, enabling the Americans to present a united front to the British. In Army terms the Allies, at the crossroads in European strategy, had chosen correctly. The choice gave promise of realizing the basic objectives of Marshall and his staff in the conflict with Germany-a decisive war waged with a minimum of loss, expense, and time.
Nevertheless, the compromises adopted and the lingering debate over European Mediterranean strategy in the following year indicate that, in the final analysis, negotiations at Quebec fell short of the final showdown desired by Marshall and his staff. Subsequent events would show that the Mediterranean issue was still far from permanently settled. Not only were the questions of advance in Italy and of eastern Mediterranean operations to rise again, but firm agreement on prospectively one of the most important links between the European and Mediterranean theaters-a southern France operation-still had to be reached. Aside from the need to put Anglo-American (and French) undertakings against Germany into final form, moreover, the conceptions of the Russians on Allied planning for concluding the war in Europe remained to be heard and the Western pattern of operations somehow had to be coupled with the Soviet effort to crush Germany. These problems and their solutions still lay in the future. For the Army staff after QUADRANT the immediate question was whether the outcome of the fight General Marshall had led at Quebec would prove successful and the British would follow through all the more important now since the Russian bear, long impatient for the "second front" in Europe, was beginning to grumble again.


Previous Chapter     Next Chapter

Return to the Table of Contents

Search CMH Online
Return to CMH Online
Last updated 1 June 2004