Part Three: Support Activities
Chapter 10: Anglo-American Collaboration
On 15 December 1942, the Military Policy Committee submitted its first report to the Top Policy Group on the “present status and future program” of the Manhattan Project. The report dealt at length with such matters as scientific progress, the organization of the project, the need for funds, the availability of raw materials, and the status of the Anglo-American atomic partnership. The latter, reported the committee in something of an understatement, needed “clarification.”1 In effect, at the urging of OSRD Director Vannevar Bush, S-1 Chairman James B. Conant, and General Groves, the Military Policy Committee was proposing a reconsideration of American policy on the exchange of information and a presidential decision not only on the immediate problem but also on the far-reaching one of postwar relations in the field of atomic energy.2
This call to reevaluate Anglo-American collaboration on atomic energy research and development was a result of the extensive and rapid expansion of the Manhattan Project during the past six months. Until then the American effort had faced serious problems and its leaders had been willing, even eager, to compare notes with their British counterparts. But, by the fall, with both the scientific and engineering programs moving ahead, the project’s military and civilian administrators had made an impressive start at cutting away red tape, thus assuring the atomic program a strong and solidly backed position in the American war effort. As the need for British assistance seemed less urgent, a new attitude about interchange took hold, and in December project leaders voiced their increasing reluctance, reinforced by growing security considerations, to give the British the fruits of American labors.
Breakdown of Interchange
The atomic partnership between the United States and Great Britain, which the allies had begun on a small scale in the fall of 1940 and developed into a full exchange program by late 1941, first underwent a slight modification in the early summer of 1942. Meeting at Hyde
Park on 20 June, President Franklin D. Roosevelt and Prime Minister Winston S. Churchill agreed that the United States should take the major role in atomic weapons production and that Great Britain should devote its already severely limited resources to the more immediate problems of fighting the war. In spite of this somewhat qualified yet carefully considered arrangement, which would permit the British to avoid the risk that large-scale atomic installations might be damaged or destroyed by German air raids, Churchill left the conference with the “understanding ... that everything was [still] on the basis of fully sharing the results as equal partners,” and shortly thereafter Roosevelt reported to Bush that he and Churchill were “in complete accord.”3
It appeared that the two wartime leaders had reaffirmed continuation of the free and open exchange of atomic information; however, developments in the months following the Hyde Park summit clearly illustrate the slow waning of Anglo-American collaboration. On 5 August, six weeks after the Roosevelt-Churchill talks, the British Cabinet officer in charge of atomic energy, Sir John Anderson, Lord President of the Council – who was to the Tube Alloys program what Secretary Stimson was to the DSM program – wrote to Bush. He proposed integrating the British gaseous diffusion project into the American program and, as a consequence, providing British representation for the OSRD S-1 Executive Committee.4 Thus, with the simultaneous transfer of the British heavy water research group to Canada, which Sir John concurrently was suggesting to Canadian authorities, most Tube Alloys activities would be removed beyond the danger of German air attacks. Sir John also made reference to the broader question of controlling atomic energy, both during the war and afterwards. For this, he recommended immediate implementation of a joint policy on patents and raw materials and the early establishment of an Anglo-American commission on atomic energy.
Anderson’s proposals reached Bush at a time when the Manhattan Project still was beset with major difficulties; scientific problems loomed large, adequate priorities were lacking, no decision had been reached on site questions, and even the basic matter of organization remained unresolved. Bush, accordingly, was in no position to commit himself to anything far-reaching, no matter how much he still desired British assistance. Finally on 1 September, after checking with Stimson’s assistant, Harvey Bundy, Bush replied to Anderson, expressing general approval of close Anglo-American collaboration but putting off for the moment any specific implementation of this principle. Only
the transfer of the heavy water group to Canada – an action subsequently taken – elicited his immediate concurrence. For the rest, he said, he would reply “somewhat later when other broad phases have been resolved.”5
Within the next few weeks, Manhattan leaders were successful in overcoming many of the uncertainties. Yet a strong desire for the kind of close partnership Anderson had suggested still was lacking; indeed, when General Groves raised the question of Anglo-American relations at the S-1 Committee meeting in Stimson’s office on 23 September, no one pressed for immediate action. Because some members felt working closely with the British might even slow down American research, the committee agreed to delay any decision until Stimson had talked with the President. When Bush wrote to Anderson a week later, he outlined the new American organization and urged continued close contact, but he purposely avoided a precise commitment, pending word from the President.6
It was the end of October before Stimson was able to discuss the issue with Roosevelt, for this was a period when relations between the Secretary of War and the President were somewhat strained by disagreement over the forthcoming North African operations and Stimson saw Roosevelt infrequently. Finally, following a Cabinet meeting on the twenty-ninth, he seized the opportunity to talk with Roosevelt alone. After pointing out that the United States was doing most of the work on atomic energy, the Secretary added that Manhattan leaders wanted to learn what commitments the President had made to the British. When the President assured him his conversation with Churchill had been “of a very general nature,” Stimson suggested going “along for the present without sharing anything more than we could help.” The President agreed but indicated that he, Churchill, and Stimson had better talk over the whole problem before too long. And there the matter rested.7
Meanwhile, as the American Army took over management of more aspects of the atomic project, the British were becoming disturbed at the trend toward an independent course that minimized Anglo-American cooperation. Hence, no one was surprised when Anderson proposed that Wallace A. Akers, the engineer who headed the British Directorate of Tube Alloys (which was comparable to Conant’s position as chairman of the S-1 Executive Committee) should visit Washington, D.C. During the weeks that followed Akers’ arrival in early November, he assiduously consulted with Bush, Conant, and Groves, seeking ways to link more closely the American-British atomic energy programs but achieving only an agreement on steps to set up and support the British heavy water research group in Canada.
What Akers wanted, based on his understanding of agreements reached “at the highest levels,” was a “really cooperative effort between the two countries.”8 This would include joint research, development, and production efforts, and complete interchange of information on all aspects. British scientists and engineers would work in American plants and their American counterparts would do the same in England. Each country would make available to the other all atomic data in its possession, including theoretical and developmental information, plant designs, and operational details. This approach, insisted Akers, was the most efficient way of assuring success for the program and, moreover, would be in harmony with the understanding between the President and the Prime Minister.
The position taken by Bush, Conant, and Groves – as worked out among themselves and at meetings of the Military Policy and S-1 Executive Committees – fell considerably short of Akers’ view. They were still uncertain about what Roosevelt had told Churchill, and especially about what he now desired, and because they were not convinced that complete cooperation on all phases of the program would necessarily build an atomic bomb any sooner, they preferred that cooperation and interchange of information be restricted to matters that would be of use to each partner in the successful prosecution of the war. The three Americans also shared the suspicion that Akers’ arguments most probably were “influenced by an undue regard for possible postwar commercial advantages.”9 Another serious concern was the growing problem of security, which would increase if British scientists were permitted access to all project developments. Finally, too, joint Anglo-American production certainly would complicate production efforts in the United States and might actually impede, rather than speed up, the manufacture of atomic bombs.
From the American view, the extent of atomic cooperation that would be desirable varied according to the specific phase of the program concerned. Bush, Conant, and Groves felt there should be no interchange whatsoever on the electromagnetic separation process, because the British were not working on this method and presumably had no “need to know.” Akers replied with the argument that complete cooperation had been agreed upon, regardless of which country developed the idea or of where the production plants were to be built. Progress on one method had a direct bearing on work being done on other methods, he insisted, and there ought to be full interchange on the electromagnetic process.
On the gaseous diffusion process, where the British had done considerable work, the American project leaders were willing to permit unrestricted
interchange on experimental and design problems, but felt that exchange of information beyond this was unnecessary. Akers argued that limiting exchange on gaseous diffusion to these aspects was not acceptable. The British should be given full information on construction and operation of the production plant, and British engineers and scientists actually should be employed in it. Not only did this fall within his understanding of the Churchill-Roosevelt agreement, but also, as he emphasized, the British were already working on a diffusion plant.
As for production data on U-235, Bush, Conant, and Groves held that none should be given to the British because of the fact that their interest in uranium production was only for experimental purposes. The same applied to plutonium. The three Americans were willing to exchange information about scientific findings, but not about the design, construction, or operation of production plants. Heavy water, which might be used to manufacture plutonium, fell into the same category. Akers continued to argue, although in vain, for full British participation in American efforts.
Regarding the work at Los Alamos, Bush, Conant, and Groves proposed that there should be no interchange with the British on information pertaining to weapon research and development. Once again Akers urged full reciprocity of information, and again his arguments were without effect.
As a direct result of these extended discussions, the Military Policy Committee prepared a comprehensive progress report on its views on future U.S.-British relations in the field of atomic energy. The report, dated 15 December, identified “only one reason for free interchange of secret military information between allied nations – namely, to further the prosecution of the war in which both are engaged.”10 The consensus of the committee was that, because the British had now given up any intention of manufacturing atomic bombs or significant amounts of fissionable materials during the war, making production data available to them would not increase their military capabilities. Although the work of British scientists on diffusion and heavy water was well along, the results of their research was not essential to the Manhattan Project; American efforts in these areas were considerably advanced. A complete halt of interchange on diffusion and heavy water would be an inconvenience, but it would not seriously hinder progress of the American program.
Nor did the committee see any moral objections to halting interchange. Both countries had worked on the basic concept, as, indeed, had the Germans. British studies on diffusion probably had benefited from American research, and vice versa. Heavy water had been used in a uranium pile first in France and then in Britain at the instigation of refugee French scientists. But only after the discovery in the United States that plutonium was fissionable by fast neutrons had the British given a high priority to the heavy water program. And, as Conant emphasized, the British had not followed a policy of unrestricted
interchange in the past. They had been unwilling to share with American scientists information about several of their own developments – a secret bomb disposal method for one – because they would not help the American military effort.11
The committee concluded that halting interchange would not unduly hinder the Manhattan Project, could hardly be regarded as unfair, and had obvious security advantages. However, complete cessation certainly would cause friction with the British and might adversely affect the flow of uranium from Canada and other areas. Thus, in its report the committee recommended that a policy of limited interchange, confined to information that could be used to win the war, should be adopted as national policy.
With the approval of three members of the Top Policy Group, the Military Policy Committee report, a copy of a letter from Akers to Conant restating the British position, and a separate summary by Bush of both British and American views reached the White House on 23 December. Two days after Christmas, Stimson went to see Roosevelt. The British, he had just learned, had signed a treaty with the Soviet Union in September to exchange information on new weapons, including any that might be developed in the future. The treaty, said Stimson, came as a complete surprise and had a direct bearing on any Anglo-American exchange of information. Obviously, it posed the possibility that weapons development data passed on to the British eventually would reach the Russians. This news apparently reinforced the arguments set forth by the Military Policy Committee, and the next day, 28 December, the President told Bush that he approved the committee’s recommendations.12 In so doing, he adopted for the United States a new policy of limited interchange with its atomic partner across the Atlantic – one that restricted collaboration to information of use during the war.
The Quebec Agreement
With the United States’ position on limited atomic partnership solidly affirmed, Conant undertook the task of informing both the British and the Canadians. The day after New Year’s (2 January 1943), he wrote to Dean C. J. Mackenzie, head of Canada’s National Research Council, and explained how the new American policy would affect the work on heavy water under way in Montreal.13 Then on the seventh, he prepared a lengthy memorandum in which he outlined the specific regulations for Anglo-American cooperation. Because Conant never officially presented this memorandum to the British, it was in effect only a working paper. Its contents, however, generated
considerable controversy, and Churchill later complained to presidential aide Harry Hopkins that Conant’s memorandum “drastically [limits] interchange of technical information and entirely destroys [Roosevelt’s] ... original conception” of a “ ‘coordinated or even jointly conducted effort between the two countries.’ ”14
Although Akers had read Conant’s memorandum, he apparently had elected to keep his thoughts to himself. But on the twelfth, he ran headlong into the practical effects of the new policy at a meeting with Colonel Nichols of the District staff and Percival C. Keith of Kellex. When Akers asked for full exchange of information and access for British scientists to the American diffusion production plant, Nichols informed him that such requests would be “subject to General Groves’ decision,” the outcome of which the British representative could by now undoubtedly guess.15
The problem came to a head on the afternoon of the twenty-sixth at a meeting with Groves and Conant. Akers protested, argued, and bargained, largely in vain, for a relaxation of the American attitude. All he achieved was Groves’ statement that America probably would be willing to reopen information exchange on heavy water production if Great Britain would make significant use of it before the end of the war and would indicate a willingness to make slight adjustments regarding interchange on the diffusion process. On other matters – electromagnetic separation, the use of heavy water in a chain reaction, the furnishing of uranium metal and purified graphite to the Canadian group, the chemistry of plutonium, and the design and construction of a weapon – the American project leaders remained adamant. Unsuccessful in his mission, Akers returned home to England a few days later.16
Meanwhile, word had reached the Moroccan town of Casablanca, where Churchill and Roosevelt were meeting to discuss Anglo-American strategy. During the military talks the subject of atomic energy was not even mentioned; but, in confidence, the Prime Minister asked the President about the American position on Tube Alloys. Roosevelt’s reply, as the Prime Minister related it to Bundy, was to assure Churchill again that atomic energy was a joint enterprise. Hopkins, also present during the exchange, quickly added that the problem could be easily straightened out as soon as the President returned to the White House.17
The two leaders parted, Roosevelt to Washington and Churchill to London via the Middle East. Soon after returning home in early February, the Prime Minister apparently received a thorough briefing on Akers’ disturbing experience in the United States and, on the sixteenth, he cabled Hopkins to remind him of his assurances given at Casablanca. “The American War Department,” complained Churchill, “is asking us to keep them informed of our experiments while refusing altogether any information about theirs.”18
That Churchill had expressed his perturbation to Hopkins rather than directly to Roosevelt indicates the importance he attached to the problem. The Prime Minister was well aware of Hopkins’s close relationship with Roosevelt and regarded him as a “most faithful and perfect channel of communication.”19 A personal representation by “Lord Root of the Matter,” as he once called Hopkins, would be more effective than a simple cable direct to the President. Yet, curiously enough, Hopkins apparently knew little about atomic energy matters. Certainly his ready assurances at Casablanca indicated his unfamiliarity with the complexities of the problem.20
But in the weeks that followed Churchill’s cable, Hopkins set about familiarizing himself with the problem of Anglo-American interchange. Now well briefed by Conant and Bush, and perhaps by Lt. Gen. Brehon B. Somervell, the Army Service Forces (ASF) commander, he replied to Churchill’s continued prodding with cables that avoided a direct answer and thus left the American position unchanged. Bush, in turn, reviewed the policy separately with Conant and Stimson and collectively with fellow members of the Military Policy Committee at its 30 March meeting. “None of us,” he reported to Hopkins on 31 March, “can see that the present policy, which was approved by the President after it had had the careful review and approval of General Marshall, Secretary Stimson, and Vice President Wallace, is in any way unreasonable, or such as to impede the war effort on this matter. Neither can we see that the application is at present unwise.” Supporting a strongly worded memorandum from Conant, which he enclosed, Bush stressed, as had Conant also, the growing American belief that British desire for information about the American program was not for wartime weapons development but, rather, for postwar commercial and industrial application. This might perhaps be considered in another context, said Bush, but it should in no way be allowed to interfere with the Manhattan Project or with the “proper conduct of the secure development of a potentially important
weapon.”21 Apparently convinced of the correctness of the American policy, Hopkins allowed the matter to drag on through April without resolution. Even though he had promised British Foreign Secretary Anthony Eden a telegram that would give his “views fully,” he never sent it.22
Except for limited exchange between the Montreal and Chicago groups, Anglo-American collaboration slowed almost to a standstill. Sir John Anderson, fearing a weakening of Churchill’s negotiating position, refused an American request that chemist Hans von Halban, a refugee from the French atomic program, be permitted to come to New York to confer with Fermi and Urey on heavy water problems, and in partial reaction the Military Policy Committee reduced American support of the heavy water project at Montreal. Hopkins’s procrastination did nothing to improve the steadily deteriorating situation, and British scientists began thinking seriously of building their own U-235 plant. During this time, the only answer the British received to Churchill’s protests was an indirect one: an explanation of the American position by Bush and Conant to Dean Mackenzie of the Canadian project as he passed through Washington, D.C., on his way to London to discuss the problem with his British colleagues.
By then, however, Churchill had decided to pay a personal call on Roosevelt.23
The TRIDENT Conference, as Churchill dubbed his third major wartime meeting with Roosevelt, began in Washington on 12 May; however, it was not until the twenty-fifth, his last day in the national capital, that he raised the problem of atomic interchange. Hopkins telephoned Bush, and that afternoon the two Americans met with Professor Frederick Lindemann (Lord Cherwell), the British physicist who was one of Churchill’s closest advisers. An able negotiator, Lord Cherwell had already formed some strong opinions about who was responsible for the new American position. The whole situation, he had told Canadian Prime Minister William Lyon Mackenzie King a week earlier, was the fault of the American Army, which had taken over the atomic energy program from the scientists. “They are as difficult about it in their relation with Britain,” King noted in his diary, “as Stalin had been in telling of what was being done in Russia.”24
The meeting resulted in an impasse, although it did clarify matters to some extent. After Bush restated the American position and explained the reasons, Lord Cherwell pressed for a change. He denied Great Britain was aiming at any postwar commercial advantage, but admitted the British wanted to be in a position to build atomic weapons once the war was over. During the war, he added, his government was willing to depend on the United States for these weapons, but in the postwar period it could not afford to rely on any other power for military security. Bush and Hopkins immediately pointed out this was a far different question than had been previously discussed. It concerned broader problems of postwar international relations, the solutions to which, Hopkins noted, the Roosevelt administration constitutionally could not commit its successor: Lord Cherwell indicated that if the United States refused to provide the desired information on atomic production, the British might – to guarantee their own future security – have to undertake an immediate production program of their own, diverting whatever was necessary from the main war effort. But he did not put this in the form of an outright threat.
The main question had at last been isolated: Was it necessary for America to provide Britain with production data during the war to ensure her military security in the postwar era? It was clear to Hopkins where the problem lay and he told Bush to do nothing further on the matter. Presumably, Hopkins would take it up with the President.
That evening, Churchill apparently discussed the problem privately with Roosevelt. There is no record of this meeting. Indications are that the President was not informed of the Bush-Hopkins-Cherwell conference. Once again he showed his earlier willingness to cooperate fully with the British. The next morning Churchill cabled Sir John Anderson that the President, foreseeing that the general agreement on wartime interchange would be fulfilled by the almost certain use of the bomb in the war, had agreed that the exchange of information on Tube Alloys should be resumed and that the enterprise should be considered a joint one.”25
Whatever Roosevelt told Churchill, he did not pass it on to Bush or Stimson. How much Hopkins knew is not clear, but he was at least aware that Roosevelt had promised Churchill something. A month after TRIDENT, Bush had his first opportunity to brief the President on this talk with Lord Cherwell. Roosevelt seemed impressed, but he said nothing about any arrangements he might have made with the Prime Minister and simply told Bush to “sit tight” on interchange.26
Meanwhile, the British had sent Akers to Ottawa, and during his stay in the Canadian capital, Churchill had cabled Roosevelt once again, seeking to implement their agreement on atomic energy. He received no satisfactory reply. Finally in mid-July, the President asked Hopkins what to do about interchange. Hopkins replied that he [Roosevelt] had “made a firm commitment to Churchill in regard to this when he was here and there is nothing to do but go through with it.”27 Accepting this fact, on the twentieth the President cabled Churchill that he had arranged matters “satisfactorily.” The same day he wrote Bush, who was in London attending to other scientific matters, that because “our understanding with the British encompasses the complete exchange of all information,” he should “renew ... the full exchange of information with the British Government regarding the Tube Alloys.”28 The President’s letter should have settled the matter. Yet, by one of those peculiar quirks of fate, the new directive did not reach the OSRD director in time to be effective.
On the fifteenth, an unexpected confrontation by an agitated Prime Minister, who daily was becoming more and more disturbed over the interchange problem, had occasioned Bush to refer him to Secretary Stimson, who, with Bundy, also was visiting England. Two days later, Churchill asked Stimson to “help him by intervening in the matter.”29 Harboring strong feelings about the value of close Anglo-American collaboration on all wartime activities, Stimson arranged for a conference on the twenty-second. Shortly before the meeting, the Secretary met with Bush and Bundy. Particularly concerned about the need for careful international cooperation under the new world conditions that atomic energy would create, Stimson questioned Bush carefully and forcefully, and at times the OSRD director felt almost as if he were being cross-examined by the distinguished lawyer. When Bundy suggested constitutional limitations on the President’s power to make long-term commitments, Stimson dismissed this as “the argument of a police-court lawyer.” But in the end, he agreed that Bush should present the American position to the British as he saw it.30
That afternoon, the three Americans sat down with the Prime Minister, Anderson, and Lord Cherwell. Because Churchill, for reasons that are not known, had not yet received Roosevelt’s cable, none of the participants were aware of the
President’s decision to reaffirm a policy of full interchange. Churchill opened the session with a vigorous defense of the British position, emphasizing his fear that unless Great Britain had the means and knowledge of how to develop atomic weapons, Germany or Russia might “win the race for something which might be used for international blackmail.” He seemed particularly concerned about the possible atomic threat from Russia, which appeared to be at the root of his worries about the postwar world. If the United States would not “interchange fully,” he said, Great Britain would have to undertake its own development “parallel” to that of the Manhattan Project, no matter how this might affect the rest of the war effort.31
As diplomatically as possible, Bush attempted to restate the American view and to point out that the main U.S.-British differences lay in the area of “postwar matters.” Stimson seconded this approach by reading aloud a short, clear analysis of the situation he had written in preparation for the meeting.32 The Prime Minister then proposed a five-point agreement to be signed by Roosevelt and himself. Under this agreement, there would be “free interchange” of atomic information within a “completely joint enterprise”; neither government would “use this invention against the other”; neither would “give information to any other parties without the consent of both”; neither would use atomic weapons “against any other parties” without the other’s consent; and, finally, “in view of the large additional expense incurred by the U.S.,” British commercial or industrial use “should be limited” in whatever way the President deemed “fair and equitable.”33
Stimson agreed to pass these proposals on to the President. He could not comment officially, but he was obviously pleased. “Satisfactory atmosphere produced,”34 he noted in his diary. Bush, too, felt somewhat better, for while the Prime Minister’s proposed free interchange still seemed dangerous from a security viewpoint, Churchill had made a convincing disclaimer of any postwar commercial motivations. When Churchill received Roosevelt’s 20 July message several days after the conference, he was unable to determine from the general terms of the message that the President, in fact, had completely reversed the American position. Only Roosevelt’s explicit instructions in his 20 July letter to Bush would have indicated this shift in policy. But the letter of instructions, which the OSRD cabled to Bush on the twenty-eighth, was somehow garbled in transmission or decoding; it ordered Bush to review, rather than renew, full interchange. Even this mild wording gave Bush some concern, but not nearly as much as the original version would have.35
On the same day Churchill approved a formal draft of the British proposal, which he forwarded to Stimson on the thirtieth. This version, drafted by Anderson and revised by Churchill, was basically the same as the one the Prime Minister had presented orally. It eliminated the specific references to “free interchange” within a “completely joint enterprise,” substituted a general statement about pooling “all available British and American brains and resources,” and made even more explicit the British disclaimer on “industrial and commercial aspects.” Sir John Anderson would go to Washington at once, said Churchill, to help arrange “for the resumption of collaboration.”36
Back in Washington, Bush learned the actual wording of the President’s instructions. He also found awaiting him a strong memorandum from Conant, which reiterated the Harvard president’s “conviction ... that a complete interchange with the British is a mistake” and authorized Bush, if he saw fit, to quote him “on this point to those in higher authority.”37 This proved unnecessary, for the British remained unaware of Roosevelt’s actual position and continued negotiating on the basis of American policy as explained by Bush in London.
With the approval of Secretary Stimson, Bush carried out final negotiations with Anderson. He kept in close touch with the Secretary, Bundy, and General Marshall – Vice President Wallace and General Groves were out of town – and especially with Conant, who participated in the opening talks with Anderson on 3 August. Stimson and Marshall also had lunch with the British representative, but their conversation appears to have been more of a general discussion than a bargaining session.
On the sixth, after an exchange of letters, Bush and Anderson came to a meeting of minds on a proposed agreement to be signed by Roosevelt and Churchill. This agreement was based on the four-point draft Churchill had sent Stimson a week earlier, but added a fifth section “to ensure full and effective collaboration.” This section provided for establishment of the Combined Policy Committee, which would determine the role of each country, maintain an overall review of the project, allocate critical supplies, and have the final say in interpreting the joint agreement. There would be interchange on all sections of the project. Details would be regulated by ad hoc agreements, subject to committee approval, and Bush stipulated that information made available to committee members would be general in nature. Anderson also agreed that the committee would not interfere with the
Army’s control of the Manhattan Project.38
The next day, Bush forwarded the draft agreement and copies of his correspondence with Anderson to the President. He acknowledged the delayed directive of 20 July sent to him by Roosevelt, but then went on to state his conviction that his understanding with Anderson “provided adequately for appropriate interchange, with due regard to the maintenance of security, and with the object of providing the British with all of the information they can utilize in this connection in the prosecution of the war, in return for the benefit of the deliberations of their own scientific and technical groups.”39 In a separate note to Bundy, Bush urged that Secretary Stimson “impress upon the President” the desirability of limiting agreements to wartime objectives and the dangers of making commitments for the postwar period.40
General Marshall, too, urged caution, and Bundy strongly recommended to Stimson that the President talk with Bush, or at least carefully read the Bush-Anderson correspondence, before signing any agreement with Churchill. He emphasized to the Secretary that Bush and Conant were trying to protect Roosevelt from any possible charges that he was exceeding his legal authority or acting from any other motivation than a desire to win the war. Strongly impressed by Bundy’s urging, Stimson went to the White House on 10 August, determined to make these points. Whether or not he did is unclear, but he did describe the negotiations with Churchill and raise at least one caveat. He asked the President whether a problem might arise from Churchill’s proposal that neither country would use atomic energy against third parties without the consent of the other. Roosevelt indicated he saw no danger in the provision.41
Even as Stimson met with Roosevelt, the Prime Minister was settling himself in Quebec, in preparation for meeting with the President at the QUADRANT Conference that would begin in a few days. Only then did General Groves, who had been busy on inspection trips to the West Coast and New York, learn of the forthcoming conference and realize the proposed agreement would be discussed. More than half a year had passed since the President had had a report on the Manhattan Project from the Military Policy Committee, and Groves felt Roosevelt should have an up-to-date summary before his meeting with Churchill. Groves drew up a twenty-page report; cleared it with the committee; and, on 21 August, without
showing it to Wallace or Stimson, directed Colonel Nichols to hand carry it to General Marshall in Quebec, where QUADRANT was already under way. The report, which covered all Manhattan activities, included a brief summary of relations with the British and, in the light of the Bush-Anderson negotiations, asked the President for further instructions. But when Colonel Nichols arrived in Quebec with the document, General Marshall informed him that Roosevelt and Churchill had already signed an agreement on atomic energy.42
The two leaders had approved the proposed agreement at Hyde Park, where Churchill had visited Roosevelt from 12 to 14 August.43 But it was not until the nineteenth, in Quebec’s historic fortress known as The Citadel, that they actually affixed their signatures to the “Articles of Agreement Governing Collaboration Between the Authorities of the U.S.A. and the U.K. in the Matter of Tube Alloys,” or, simply, the Quebec Agreement. It called for the earliest possible completion of the Tube Alloys project, ruled out “duplicate plants on a large scale on both sides of the Atlantic,” and acknowledged the “far greater expense” borne by the United States. It agreed “never” to “use this agency against each other” and “not to use it against third parties without each other’s consent,” and it prohibited giving “any information about Tube Alloys to third parties except by mutual consent.” In view of the heavier burden carried by the United States, “any postwar advantages of an industrial or commercial character” would be “dealt with ... on terms to be specified by the President ... ,” and the Prime Minister specifically disclaimed “any interest” in them “beyond what may be considered by the President ... to be fair and just and in harmony with the economic welfare of the world.” Finally, using the Bush-Anderson arrangement for interchange as the basis, the Quebec Agreement established the Combined Policy Committee in Washington, D.C., and designated six members.
On the choice of members, Roosevelt apparently did not consult any of his advisers, except possibly Hopkins. American members were Stimson, Bush, and Conant; British members were Field Marshal Sir John Dill, head of the British Joint Staff Mission in Washington, and Col. John J. Llewellin, Washington representative of the British Ministry of Supply. The sixth member was Canada’s Minister of Munitions and Supply, Clarence D. Howe, an American-born engineer whose appointment Churchill had cleared earlier with Mackenzie King. The British had felt that the Canadians, even though they were not a party to the Quebec Agreement, should have representation on the high-level committee because they would be making important contributions to the atomic energy project in Montreal.
The Quebec Agreement set the official basis for Anglo-American atomic relations for the rest of the wartime period. It did not establish the free and open interchange the British had desired and that the President,
indeed, had offered in his letter of 20 July. It called for “full and effective collaboration,” and both Roosevelt and Churchill believed they had provided the basis for it; however, in reality, collaboration would comprise only what was necessary for the war effort, avoiding any form of interchange that might conceivably hinder progress of the Manhattan Project.44
Implementing the Agreement
Combined Policy Committee
Despite pressure by Bush and General Marshall, and the presence in Washington of Akers and four leading British scientists who were anxious to implement interchange,45 two weeks passed before the President revealed the details of the Quebec Agreement to Manhattan officials, including particulars on the Combined Policy Committee. With Churchill visiting at the White House, the President first wanted the Prime Minister’s concurrence in the contents of the Military Policy Committee’s report before any meeting of the new committee took place.46
On 8 September, after lunch at the White House, Stimson discussed the Quebec Agreement with Roosevelt and Churchill. Having learned only that morning that he was to be chairman of the Combined Policy Committee, he asked permission to name ASF chief of staff, Maj. Gen. Wilhelm D. Styer, as his deputy – a request the President and Prime Minister readily approved.47
An hour or so later, the first informal meeting of the Combined Policy Committee took place in the Pentagon. One reason for the hasty convening was to accommodate the four British scientists, waiting impatiently to exchange data. Bush was out of town and Howe had not yet arrived from Canada, but Stimson, Conant, Dill, and Llewellin proceeded without them. General Styer was also present, as was Bundy, acting as secretary. They formed a technical subcommittee, with Styer as chairman, to make recommendations on the American and British programs, to prepare directives for interchange of research and development data, and to propose ad hoc arrangements for interchange in the area of plant design, construction, and operation. The subcommittee consisted of three scientists who had a thorough knowledge of the American, British, and Canadian projects – Richard C. Tolman, who
was General Groves’ scientific adviser; Sir James Chadwick, the eminent British physicist; and C. J. Mackenzie of the Canadian National Research Council. Despite some hesitation by Dill and Llewellin about delegating their authority, the Combined Policy Committee authorized the subcommittee to act independently on interchange whenever there was unanimous agreement among its four members.48
Working Out Interchange Arrangements
Styer’s subcommittee met on 10 September, to consider a plan drafted by General Groves and submitted by the Military Policy Committee. Because this plan hewed fairly closely to the earlier American proposals on interchange, it fell considerably short of what the British desired. On weapon development it recommended assignment of two British scientists to Los Alamos under the same security restrictions governing American scientists there. On the gaseous diffusion and heavy water pile processes it suggested interchange of scientific information through a joint committee. On the centrifuge and thermal diffusion processes, which would probably soon be dropped, Styer’s subcommittee should decide whether interchange “might affect this decision.” As for the electromagnetic and graphite pile processes, on which the British had done little work, interchange would serve no useful purpose, for these methods had reached the stage where changes “would result in serious delay in completion.”49
The subcommittee, largely at the insistence of Chadwick, recommended some modifications to the plan favorable to the British view. On the gaseous diffusion and heavy water pile processes, interchange should extend to some aspects of development and production. There should be exchange of scientific data on the graphite pile to the extent it might be helpful in the Anglo-Canadian development of the heavy water pile process. Chadwick’s contention that the British might be able to contribute to development of the electromagnetic process should be explored by a committee consisting of Groves, Tolman, and Australian physicist Marcus L. E. Oliphant. In keeping with the Military Policy Committee’s recommendations, the subcommittee reached agreement on possible personnel for other committees or representation needed to carry out interchange on the various processes. Chadwick and Sir Rudolph E. Peierls, the University of Birmingham physicist, would serve as British representatives at Los Alamos; von Halban with Metallurgical Project Director Arthur Compton, or one of his principal assistants, as a committee to exchange data on the heavy water pile process; Sir Francis E. Simon, physicist at Oxford’s Clarendon Laboratory, and Peierls with Keith, the Kellex head, and Urey on a gaseous diffusion committee; and Oliphant, Simons, and Peierls on a committee with American representatives designated
by Bush or Conant to decide the extent of interchange on the centrifuge and thermal diffusion processes.50
Despite the considerable progress made by the subcommittee, there was little specific interchange in the weeks that followed. Part of the difficulty lay in the lack of specific working procedures. To set these up, Tolman went to England in October to consult with Chadwick and other British scientists and with Sir John Anderson. General Groves, who was becoming increasingly impatient to implement interchange in those areas where it was sanctioned, closely monitored Tolman’s negotiations from his Washington office and attempted to facilitate Anglo-American coordination by keeping members of the Military and Combined Policy Committees regularly informed.51
When Groves received word that another team of British scientists soon would be arriving in the United States, the need for a speedy procedural agreement on interchange became even more critical to him. Yet not until mid-December were the British and Americans able to complete interchange procedures. With the approval of the Military and Combined Policy Committees, the new procedures went into effect on the fourteenth. Naming Chadwick as the “immediate scientific adviser to the British members” of the Combined Policy Committee, the terms of this agreement permitted that he have “access” to all work on “research and plant scale” on both sides of the Atlantic. The slight and unassuming Cambridge professor, who, surprisingly enough, got along exceptionally well with the robust and outspoken Groves, would also help guide experimental work at Los Alamos, where he would be joined by a small number of other British scientists. Peierls, and one or two others, would work with Kellex on the diffusion process and also would discuss theoretical problems of bomb construction with American scientists; he could not, however, visit Los Alamos. Oliphant and six assistants would work with Ernest Lawrence at Berkeley on research and design and then move to Los Alamos to assist on ordnance problems. Oliphant would continue his close contacts with the electromagnetic project during production and would be free to visit England to supervise any similar research there. About fifteen British scientists and industrialists, led by Akers, would exchange information on diffusion at Columbia University and Kellex. Research on heavy water piles at Montreal would be continued under a joint program to be worked out with those doing similar work in the United States.52
This arrangement was, in effect, the implementation of the Quebec Agreement. While it did not actually provide full information exchange, it went further than most members of the Manhattan Project administrative staff would have preferred. Certainly the arrangements were more liberal than Groves would have wished, although he later claimed full credit for having drawn up these “rules regulating the ... British scientists.”53 While anxious to get any British help that might speed the progress of the American program, he was generally opposed to providing Great Britain with anything more than was absolutely necessary to gain this aid. “I was not responsible for our close cooperation with the British,” he asserted a decade later. “I did everything to hold back on it.”54
By the end of January 1944, eighteen British scientists had reached New York, Washington, D.C., Berkeley, and Los Alamos, and more were expected. Only one problem remained outstanding, namely, arrangements for cooperation between the Montreal and Chicago scientists on
pile research. Around the middle of the month, senior members of both groups had discussed a joint program of research that would lead to the construction of a heavy water pile. Yet to Manhattan leaders in Washington, it seemed doubtful the venture would be of significant value during the war, and Groves and Conant, at least, preferred that it should not begin.55
On 17 February, however, at the next meeting of the Combined Policy Committee, Chadwick pressed for approval of a Canadian heavy water pile to undertake large-scale production of plutonium. Great Britain and Canada would provide the funds, the United States the heavy water, and the three nations would exercise joint control over the project. Neither Groves, who was not a committee member, nor Styer was present, but Bush and Conant apparently raised some questions. Would the project result in militarily significant production before the end of the war? Was it advisable to use up resources, especially ore? The committee turned the problems over to a subcommittee composed of Groves, Chadwick, and Mackenzie.56
The subcommittee discussed a heavy water pile with Compton, Fermi, and others at Chicago and with von Halban and his colleagues at Montreal. Then, on 6 April, it submitted its report to the Combined Policy Committee. The Hanford Engineer Works, the subcommittee concluded, would produce enough plutonium to satisfy “essential military needs” for the war, and production at the proposed Canadian plant could not begin in time “to have an appreciable influence on the outcome of the present war.” On the other hand, the potentialities of the heavy water pile were so great that its development could not be “wholly neglected.” Accordingly, it recommended continued research and development at both Chicago and Montreal, with an increased staff and the appointment of a director for the Canadian project; the design and construction of a heavy water pilot pile in Canada by the United States, Great Britain, and Canada; and future consideration of a small production pile when the experimental stage was further advanced. A week later the Combined Policy Committee adopted this program, and in the ensuing months Groves, Chadwick, and Mackenzie continued to keep an eye on the project for the committee and see to it that the approved recommendations were carried out.57
The new Montreal director was physicist John D. Cockcroft, and his staff was rapidly reinforced with British and Canadian scientists. In early May, as plans for construction of the pilot plant matured, General Groves approved an isolated site previously selected by the Canadians, near Chalk River, Ontario, on the south bank of the Ottawa River and about 110 miles
northwest of the Canadian capital. (See Map 2.) Late in the month, Cockcroft, von Halban, and others from Montreal visited Chicago. A second meeting was held in Montreal two weeks later. Discussion was limited by the rules governing interchange that Groves, Chadwick, and Mackenzie were in the process of drafting. These regulations, which the Combined Policy Committee approved formally on 19 September, limited interchange to information necessary for the design, construction, and operation of the Chalk River pilot plant. Scientists at Montreal could learn about the pilot pile at Clinton and the research piles at Argonne, and receive basic scientific data essential to the heavy water pile. They were not to be furnished with information about production plant construction at Hanford or the chemistry of plutonium or the method of separating that element, because these developments were not necessary for work at Chalk River. Finally, the regulations directed that the Montreal group should establish strict security in the transmittal of all data.
General Groves designated Maj. Horace S. Benbow as his liaison officer at Montreal, or Evergreen, to use its code name, and directed that the Chicago area engineer handle all Evergreen requests. For scientific liaison, Groves assigned physicist William W. Watson and chemical engineer J. R. Huffman to report directly to him rather than the Metallurgical Laboratory director.58
The policy established in the spring of 1944 for interchange on the Canadian project completed the arrangements approved the previous December for Anglo-American information exchange on atomic energy and fulfilled the terms of the Quebec Agreement of August 1943. British scientists were now working with Americans in the United States on several phases of the overall program and were reviewing a limited amount of information. In the remaining months of the war, Anglo-American relations steadily improved, although, inevitably, minor problems arose.59
One of the problems relating to interchange with which the Combined Policy Committee had to concern itself periodically during 1943 and 1944 was patent rights. The United States and Great Britain in August 1942 had concluded an executive agreement on exchange of patent rights that provided a general basis for negotiating more specific arrangements applicable to particular areas of interchange.60 At the time of this agreement,
Sir John Anderson had suggested to Bush the adoption of a joint patent policy relating specifically to atomic energy as an important aspect of international control. Bush, however, did not think the time was propitious for establishing such a policy; instead, he recommended that participating countries could facilitate eventual establishment of controls by seeing to it that most patent rights concerning atomic energy within their own borders were publicly owned.61
The need for patent arrangements became even more obvious after the signing of the Quebec Agreement. With scientists of both countries working together, a common policy was necessary to protect both individual and national rights. Secrecy and security aspects further complicated the difficult technicalities inherent in all patent matters.
In the fall of 1943, Arthur Blok, patent expert in the British Department of Scientific and Industrial Research, and Capt. Robert A. Lavender, retired American naval officer who advised Bush and later Groves on patent questions, attempted to reach some agreement. They concluded that the 1942 agreement did not apply to atomic developments and drew up a new proposal.62 When Bush pointed out certain inadequacies in the Blok-Lavender proposal at the Combined Policy Committee on 13 April 1944, the committee referred the problem to its recently appointed joint secretaries, Harvey Bundy and W. L. Webster of the British Supply Council. During the summer the two men studied the question, conferring frequently with Lavender, Blok, Bush, and others; and in September, they drew up a lengthy administrative procedure, which the committee approved at its meeting on the nineteenth. But project lawyers found that the procedure was in conflict with the United States patent law, and not until February 1945 was it properly amended. As finally approved at the 8 March committee meeting, the arrangement was still an ad hoc procedure, neither final nor complete, leaving the negotiation of a permanent settlement to the future.63
New Partnership Strains: Repatriation of French Scientists
The liberation of France following the Allied invasion of Western Europe in the summer of 1944 placed new strains upon the British-American atomic partnership.64 The immediate
source of the dispute was the repatriation of five French scientists – Hans von Halban, Pierre Auger, Lew Kowarski, Jules Gueron, and Bertrand Goldschmidt – who had fled to England from France after the German invasion in 1940 and then gone on to Montreal in 1943 to work in the Canadian atomic program. When they began to apply for permission to visit or return permanently to their homeland, American atomic leaders contended such visits posed too great a security risk, particularly because physicist Frederic Joliot-Curie, head of the French atomic program, was known to be a member of the Communist Party.
The Americans, and especially General Groves, took the view that the French should not be allowed to go back to France until the war was over. In May 1944, when Pierre Auger terminated his employment with the Canadian project, citing a desire to return to France to assist Joliot-Curie in rebuilding French science, Groves and the British representatives in America agreed that neither he nor any of the other French scientists in Canada should be permitted to do so and that measures should be taken to prevent any atomic information from reaching that country. Nevertheless, when Auger went to London in August to become a full-time member of the French Scientific Mission in that city, British authorities permitted him to visit France.
In October, Gueron requested permission to visit France on personal matters. Groves, who had learned that Gueron planned to see Joliot-Curie, opposed the visit because Gueron knew a great deal about the atomic project and was reputed to be an “ardent Free Frenchman” and supporter of General Charles de Gaulle. But British authorities indicated they had agreed to let Gueron go. When Groves learned this, he determined to have the French scientist kept under surveillance by Manhattan security personnel while in France. The British objected strongly. Gueron was “a man of integrity,” they asserted, and ought not to be treated as if he were a prisoner.65
Manhattan leaders interpreted these British actions to be a clear violation of the terms of the Quebec Agreement, which forbade communication of atomic information to third parties without mutual consent, and requested the American ambassador in London, John G. Winant, to secure an explanation. Sir John Anderson replied that the British had made agreements with the French scientists before they went to Canada. The first to come to England – von Halban and Kowarski – had negotiated an agreement for exchange of patent rights relating to atomic energy between France and the United Kingdom. Later when Auger, Gueron, and Goldschmidt reached England, they had worked out employment arrangements
that assured them their right to return to France as soon as the war made it feasible and also their status as French civil servants and as adherents of General de Gaulle and the Free French. Because the French scientists had made a “very special contribution” to the Tube Alloys project, in the form of “research already started by Joliot and by his action at the time when France was over-run,” Sir John contended the French had “a better claim than any other fourth country to participate in any postwar T. A. arrangements,” and he did not think it wise to embark on a course of action that would “lead the French authorities to raise the matter prematurely and with a sense of grievance already established.”66
Anderson’s revelation came as a shock to leaders of the American program. Except for some information on British acquisition of rights under von Halban’s patents that Vannevar Bush had learned about earlier, they had known nothing about the agreements between the British and French scientists. Sir John had not mentioned them during negotiations for the Quebec Agreement, yet, as Groves saw it, these third-party obligations were in obvious contradiction to that agreement. He also thought Sir John was wrong to feel he had to placate Joliot-Curie and furnish him with information about the American project.
Consequently, Groves expressed some reluctance in consenting to a British request in November 1944 that von Halban be allowed to visit London, with the understanding that the French scientist would not be allowed to go to France. But as soon as von Halban arrived in England, Sir John went to Ambassador Winant with the plea that von Halban should be permitted to see Joliot-Curie to ensure preservation of the status quo with France. Faced with Sir John’s insistent request, Winant asked Groves to come to London to talk with the Chancellor, but Groves did not go because he was too involved in urgent atomic project matters. Under continuing pressure from Sir John, Winant finally consented to von Hal-ban’s visit to Paris. The British provided the French scientist with an agenda establishing limits for information about the American atomic project that he was to give to Joliot-Curie, but Manhattan intelligence agents learned subsequently that von Halban had furnished the French atomic chief with much additional highly secret data about the American project. There were strong indications, too, that Joliot-Curie himself was shortly going to request assignment to work on the Manhattan Project.67
When Groves learned of von Halban’s visit more than a week after it
had taken place, he determined to bring an end to what he perceived as a deliberate British policy to secure postwar commercial advantage in the atomic energy field largely at the expense of the United States. As Groves saw it, Anderson was continuing to permit disclosure to the French of important information relating to atomic research that had been “developed by Americans with American money, and given to the British pursuant to interchange agreements subsidiary to the Quebec Agreement.”68
On 14 December, Groves wrote to the Secretary of War, stating that “pending the receipt of instructions from you, I will take steps to safeguard the security of the DSM project by delaying insofar as practicable the passing of vital information concerning it to the representatives of any government other than our own.”69 Stimson met with Groves, Bundy, and Harrison the next day. He informed them he would take the matter up with the President at the earliest opportunity. He instructed Groves to prepare a complete resume of the French situation and requested Bundy to notify Ambassador Winant that, until the Combined Policy Committee met to discuss the situation, he should refer to Washington “any further British proposals for disclosures or contacts which might lead to disclosures to the French. ...”70
It was not until 30 December that Stimson was able to see the President. Groves accompanied the Secretary to the White House and the two reviewed for Roosevelt the entire French problem, emphasizing that Anderson appeared to have deliberately deceived Winant and other American representatives in England regarding Britain’s commitments on atomic energy matters to France. Roosevelt’s reaction was that Winant had been “hoodwinked.” What, he wished to know, were the French after? Stimson and Groves said they believed France wanted to secure a full partnership in the tripartite atomic agreement. Roosevelt indicated that France in its current unstable political situation was not a suitable partner and, even if it were, he saw no justification for letting it share in the partnership. The discussion then turned to other matters relating to the atomic energy program.71
With the backing of the President, Stimson and Groves, assisted by Bundy, endeavored to prevent further disclosures of atomic secrets to the French during the winter and spring of 1945.72 They had a statement approved
by the Combined Policy Committee for Sir John Anderson to use if, as anticipated, the French requested full participation in the atomic energy program. The gist of this statement was that, for reasons of security, all detailed discussion of atomic matters with the French must be postponed until the end of the war, when the British would guarantee “fair treatment of any claims ... relating to commercial or industrial applications of nuclear sources of power.”73
When Sir John met with Joliot-Curie on 23 February in London, he did not present the formal statement, but he did adhere generally to the policy set forth in it. He indicated that, because of the continuation of the war and because British leaders could not readily get together with their French counterparts, progress on shaping postwar policies had not been possible. Anderson found that Joliot had concluded from the favorable British actions with regard to von Halban and Gueron, and the other French scientists, that Great Britain recognized the interests of France in atomic energy matters and, in the postwar period, would strongly support her in the pursuit of these interests.
Fear that there might be another breakdown in Anglo-American interchange if he persisted in his strong support of French atomic interests appears to have engendered a modicum of moderation in Sir John; however, he persisted in efforts to have British leaders propose that the French be assured of greater participation in atomic matters as soon as security considerations made this feasible. Two pressing concerns motivated Sir John’s actions: his belief that Britain owed this support to the French atomic scientists for their contribution to the British and Canadian atomic programs, and his fear that any policy that offended France might drive her into the Russian camp in the postwar period.
In March 1945, Bundy and Groves worked out an acceptable arrangement with the British and Canadian authorities for keeping the French atomic scientists (except Auger who was now in Paris working with Joliot) in the United States or Canada until the war was over. In early May, Auger’s status temporarily gave cause for concern when word reached Groves through Chadwick that Joliot, under pressure from one of the ministers in the French government, felt compelled to begin an active atomic energy program, including a survey of French territories for uranium and the start of research projects for the preparation of pure uranium metal and graphite. But Auger assured British scientists he would take no active part in the proposed program, and by summer of 1945 atomic developments in the United States had reached a point where the French problem no longer constituted a major threat to the security of the Manhattan Project.74