For about the first year, internal security in the MED was supervised by War Department Counter Intelligence and was thus a responsibility of the Army G-2, Major General G. V. Strong. As far back as February, 1942, he and J. Edgar Hoover of the FBI had agreed upon the various phases of security each organisation would cover. The War Department’s area of responsibility was to include all its civilian employees, as well as all civilians on military reservations or under military control.
On March 18, 1943, General Strong requested that the FBI discontinue its investigation of one of the scientists working at the Berkeley, California, laboratory. This was to lessen the chances that the suspect would find out he was being watched. Through its surveillance of the Communist party leaders in the San Francisco area the FBI had got wind of the project, for Russian espionage of activities in this laboratory had been vigorous for a long time. But it was not until a meeting on April 5 that the Bureau was first officially notified of its existence. At that time Strong told E. A. Tamm, an assistant to Mr. Hoover, of the Army’s plans for protecting the development, and it was agreed that the Army’s responsibility would include the Manhattan Project.
The MED’s own security organization was then limited to a few officers and men who performed minor security tasks and carried out certain liaison responsibilities with G-2. This was in accord with our general practice of not undertaking any task ourselves that we could have competently done for us. Through frequent meetings with the responsible officers in G-2,1 maintained proper contacts with this phase of the project.
It was not until much later in 1943 that a reorientation in the philosophy of the counterintelligence operations of the War Department made it impossible to rely any longer on the formerly very satisfactory centralized organization. Our only recourse was to set up our own complete security staff.1
For the staff control of this activity I selected Major John Lansdale, Jr., a graduate of VMI and a successful young lawyer in civil life. For some months he had been devoting almost his entire time in G-2 to our problems. He had, with the full support and approval of General Strong, set up a security organization operating outside of regular military channels. Selected officers and agents in every service command reported directly to Major Lansdale, who in turn reported directly to Strong and also to me. Thus we had been able to utilize all the resources of the Army counterintelligence organization without having to disclose through regular channels the nature of our work.
When it became necessary to move this activity into the MED a special counterintelligence group was formed, into which the existing security force was merged with little change. Later a few additions were made to this, but in the main it was kept as it was originally formed.
Throughout the entire history of the MED, both before this changeover and after, there was the fullest co-operation between the FBI and our security organization. This was vital, for the FBI had a great deal of background information that was of much value to us, and we were acquiring information that was of interest to them. There were also persons with Russian connections entirely outside our control—that is, not in the MED—who were attempting to procure information about the project. Since they were also trying to secure other military information, they were under the surveillance of the FBI. Obviously the closest co-operation was essential.
Our counterintelligence operations were under the District Engineer, who had a District Security Officer, first Captain H. K. Calvert, and later Lieutenant Colonel William B. Parsons. At every laboratory, plant and other facility there was a security officer with such assistants as he needed. He was responsible to the District Security Officer, our local MED representative and to the head of the installation.
Because of the nature of the work and our close co-ordination with the FBI and the Office of Censorship, more of the details than I liked had to be handled in my office in Washington by Lansdale and his assistants. I found it difficult at times to control to the required degree of smoothness the direct communications between my office and the many field offices. This on more than one occasion caused unnecessary, but not too serious, friction with the district organization.
The basic security problem was to establish controls over the various members of the project that would minimize the likelihood of vital secrets falling into enemy hands. Dr. Bush had already expressed concern over the risks incurred through the free exchange of information among the various people in the project. This flow had to be stopped, if we were to beat our opponents in the race for the first atomic bomb. Compartmentalization of knowledge, to me, was the very heart of security. My rule was simple and not capable of misinterpretation—each man should know everything be needed to know to do his job and nothing else. Adherence to this rule not only provided an adequate measure of security, but it greatly improved over-all efficiency by making our people stick to their knitting. And it made quite clear to all concerned that the project existed to produce a specific end product—not to enable individuals to satisfy their curiosity and to increase their scientific knowledge.
Nevertheless, security was not the primary object of the Manhattan Project. Our mission was to develop an atomic bomb of such power that it would bring the war to an end at the earliest possible date. Security was an essential element, but not all-controlling.
Never once was any definite country named to me as the one against which our major security effort should be aimed. At first it seemed logical to direct it toward the Axis Powers, with particular emphasis on Germany. She was our only enemy with the capacity to take advantage of any information she might gain from us.
Japan did not in our opinion have the industrial capacity, the scientific manpower or the essential raw material. Italy was in the same position, with the further disadvantage that any large plants would be exposed to Allied bombing attacks. We did not feel that information secured by Japan would reach Germany accurately or promptly, and we suspected that the Italian-German intelligence channels were not too smooth either.
I had learned within a week or two after my assignment that the only known espionage was that conducted by the Russians against the Berkeley laboratory, using American Communist sympathizers. Our security aims were soon well established. They were threefold: first, to keep the Germans from learning anything about our efforts or our technical and scientific advancements; next, to do all we could to ensure a complete surprise when the bomb was first used in combat; and, finally, insofar as we were able, to keep the Russians from learning of our discoveries and the details of our designs and processes.
All procedures and decisions on security, including the clearance of personnel, had to be based on what was believed to be the overriding consideration—completion of the bomb. Speed of accomplishment was paramount.
Naturally, we made every effort to find out before employing anyone whether there was anything in his background that would make him a possible source of danger, paying particular attention to his vulnerability to blackmail, arising from some prior indiscretion. A new person was usually kept on nonsecret phases of the work until a hurried investigation could be completed. Since there was no assurance that secret information might not be disclosed during the original discussion of employment, we tried to be certain, even before approaching him, that there was no likelihood that he would prove unqualified.
A number of foreign-born persons were employed throughout the project, although it was usually impossible to obtain any but the most meager information regarding their past activities. Several of these people were refugees from countries with which we were now at war, or from other countries whose ideologies they had been unable to stomach. Although it seemed highly probable that they could be entrusted to share secret information affecting the security of the United States, it was always possible that someone with disloyal intentions might slip through our screening procedures. Despite the fact that there was so much at stake, a number of critics within the project enjoyed talking about our “Gestapo” methods. However, we thought it was absolutely necessary to maintain reasonable security checks on those whose records of prior affiliations were not available to us or to the British Government.
When I was first placed in charge of the MED I found that a number of people in the project had not as yet received proper security clearances, though some of them had been engaged in the work for months.
Any question of the trustworthiness of any one of these people was troublesome, for he would already be in possession of valuable information. To remove him would create only a greater hazard, particularly if he thought our suspicion of him unjustified. (I remembered that Benedict Arnold’s treason had been sparked by his feeling that he had been unfairly treated.) Moreover, if we were to dismiss a person without publicizing the proof, which we would not want to do, the understandable resentment of his friends and associates in the project might seriously interfere with their work.
Almost all our original scientific workers came from academic surroundings. Most of them had been in universities as students or young teachers during the depression years, when there was more than the usual amount of sympathy for Communist and similar doctrines. Almost all of them at one time or another had been exposed to Communist propaganda and had had friends who were secret or even semi-open Communists.
I realized what the temper of the times had been, even though I never had any sympathy for the philosophy or for the educated Americans who adopted it. Discussions with others experienced in this area led me to the belief that among those whose employment would be to the advantage of the United States a reasonable distinction could be made between individuals whose use might be dangerous and individuals whose use would probably not be.
We gave a great deal of weight to how closely the person had followed the party line and for how long. We were particularly interested in how closely he had followed the twists and turns of Soviet relations with Germany. In most doubtful instances this was a deciding factor.
Our problem was made much more difficult by the very limited number of qualified atomic scientists available in this country. We could not afford not to use everyone possible.
The most disastrous break in security was that resulting from the treasonable actions of the English scientist, Klaus Fuchs. Fuchs was born in Germany and had fled to England, where he completed his education. The British authorities had been informed by the Germans prior to the war that he was a Communist. For some reason they ignored this and did not even record the information where they would find it. After the outbreak of the war he was interned as an enemy alien, first in the British Isles and then in a prisoner of war camp in Canada. After some time there he was released and returned to work in England on atomic research. After his return he was made a British citizen.
Our acceptance of Fuchs into the project was a mistake. But I am at a loss when I try to determine just how we could have avoided that mistake without insulting our principal war ally, Great Britain, by insisting on controlling their security measures.
When I received the names of the first group of British scientists coming over to work in the Manhattan Project, under the terms of the Quebec Agreement, I observed that there was no mention of their reliability. I told the British official with whom I was dealing that I would have to have a statement that they had been properly cleared. The statement furnished in reply was inconclusive, in my opinion, and I asked for a more definite one. This was given me; it said that each member had been investigated as thoroughly as an employee of ours engaged on the same type of work.
Since the disclosure of Fuchs’ record, I have never believed that the British made any investigation at all. Certainly, if they had, and had given me the slightest inkling of his background, which they did not, Fuchs would not have been permitted any access to the project. Furthermore, I am sure the responsible British authorities would have withdrawn his name of their own volition, before giving me his history.
If Dr. Chadwick had been in charge of the British mission at that time, as he was later, I am sure that no such deception would have been attempted. Chadwick was always most punctilious in informing me of the slightest question of background, including that of German blood. Unfortunately for the free world, Chadwick did not take over until a few weeks later.
Since Fuchs was uncovered, it has often been suggested that I should have investigated each British subject before he was admitted to the project. This would have been most presumptuous and, in fact, impossible without complete infringement of British rights and without the co-operation of the British Government, which we would not have obtained.
It was a British responsibility. As partners in the atomic field each nation had to be responsible for its own personnel. The United Kingdom not only failed us, but herself as well.
I have always felt that the basic reason for this was the attitude then prevalent in all British officialdom that for an Englishman treason was impossible, and that when a foreigner was granted citizenship he automatically became fully endowed with the qualities of a native-born Englishman. With the uncloaking in recent years of Fuchs, May, Maclean and Burgess, as well as others, I doubt if this feeling still prevails.
No one but Fuchs and the Russians know what he told them. We do know, however, what he knew and consequently what he could have told them. He knew the general progress of atomic development up until the fall of 1942, when we stopped the almost complete interchange of information. He knew the general design of the gaseous diffusion plant at Oak Ridge and many of its details. He knew the details of design for both types of atomic weapons, the gun type and the implosion type. He knew our thinking about how these could be improved, and he knew of our studies and belief in the possibility of the H-bomb, not as it was later developed, but as the much more expensive and complex weapon we then envisaged.
It was, of course, utterly impossible to make a thorough investigation of the past history, loyalty, habits, and citizenship of the thousands and thousands of construction and operating employees at Oak Ridge, Hanford and elsewhere. In general, all were investigated to some degree, but for practical reasons, the degree varied widely. The investigations of those who would not have access to any classified information, such as truck drivers, cafeteria workers, and the like, were limited to a brief check of police and fingerprint files. The background of others who were in positions where they could obtain secret information was much more carefully examined. For a few, the investigation was most thorough and went back to their infancy.
In the restricted areas, each employee had to fill out a special personnel security questionnaire before he could enter the area or handle classified documents. Where there seemed to be any question about him he was thoroughly investigated.
All fingerprints were sent to the FBI. If a record was on file, it was returned to be compared with the arrest record given by the employee at the time he was hired. Most of these arrest records involved traffic offenses or drunkenness. Any employee who had failed to give the correct arrest record when he was hired, or had failed to give a complete record on serious charges, was interviewed. Depending on his attitude when questioned, the seriousness of his arrest record, the quality of his work, his absentee record, and the need for men of his particular ability, he was either retained or discharged. Many were discharged for falsifying their records of arrests on serious charges, and many were retained. Of those retained, most were used in the general areas and not in the restricted areas. No one was hired or kept on who had been convicted of rape, arson or narcotics charges. Such persons were felt to be unreliable because of their demonstrated weakness in moral fiber and their liability to blackmail.
Press security was the other side of the coin. Here we had the invaluable co-operation and assistance of the Office of Censorship under Byron Price and N. R. Howard, editor of the Cleveland News. Howard was succeeded by Jack Lockhart of the Scripps-Howard papers, and it was he who was our usual contact. With his wise advice this aspect of our security problem was well handled indeed.
The general principles governing our control of information were simple. First, nothing should be published that would in any way disclose vital information. Second, nothing should be published that might attract attention to any phase of the project. Third, it was particularly important to keep such matters out of any magazine or newspaper that was likely to be read by an enemy agent or by anyone whose knowledge of scientific progress would enable him to guess what was going on.
An item carried by an Amarillo, Texas, newspaper, for example, had nothing like the potential danger of an identical item in a New York newspaper or in a national weekly magazine. For the same reason we extended the ban even to the reprinting of relevant articles that had appeared abroad. The press was always a bit restive under this restraint, but we were concerned lest the republication of such articles in this country might lead to others and to speculation by astute reporters about the purpose of the MED. We were only too aware that the piecing together of bits of published information is a prime source of knowledge to every intelligence organization.
It was in order to prevent speculative articles as well as the publicizing of any of our efforts that the press and radio had been asked to avoid the use of certain words, such as “atomic energy.” Certain decoy words, such as “yttrium,” were included in the list to camouflage its real purpose. This was a step we did not want to take, for it automatically pointed out to the press that the government was interested. However, Howard insisted that we simply had to do it if press security was to be maintained. Most reluctantly we agreed. As it turned out, it was a very wise move and an absolutely essential one.
We wished, too, to avoid any widespread mention of such places as Hanford or Oak Ridge and all mention of Los Alamos, as well as any reference to the MED. We also did not want any mention of my name that might arouse the interest of a foreign agent in my activities. Yet to have banned all reference in the near-by papers to Oak Ridge or Hanford would have been neither practical nor desirable, for it would only have tended to attract attention locally. We did try to keep Los Alamos entirely out of the news, but the Knoxville papers were permitted to carry items—mostly in the nature of social notes—about employees and events at Oak Ridge, though nothing, of course, that would help the average reader determine the purpose of the project or its importance. The same leeway was given to the papers close to Hanford.
We did have several unfortunate security breaks, but none of them, so far as we could ever find out, attracted any particular interest. The one with the worst potential for damage was a radio program that discussed the possibilities of an atomic explosion. The script for this had been prepared for the regular news reporter on a network program; he himself had had nothing to do with writing it. Unfortunately, in order to meet his travel schedule, he delivered it from a small affiliated station, where apparently it had not been reviewed to make certain that it did not violate press censorship rules.
From all that we could ever discover, there had been no deliberate breach of security. The information on which the talk was based came from a scientist who was not connected with the project in any way but who evidently had an inkling of what was going on, gleaned, we thought, from some of the project’s scientists at the large laboratory in his city. The actual text was written for the reporter by a friend of the scientist. There was never any question in my mind but that the reporter delivered it in good faith. The failure of the radio station to stop it was attributable to plain carelessness.
Another incident that concerned us greatly was the appearance in a national magazine of an article hinting at the theory of implosion. While it did not violate any rules, it was most disturbing. A thorough investigation indicated that it resulted from the work of an alert and inquisitive reporter in another country.
There was one unfortunate happening not too long before the bombing, when a Congressman, in discussing an appropriations bill, commented on the importance of the Hanford Project. This item was picked out of the Congressional Record and was republished in a newspaper without any comment. I could never disabuse myself of the feeling that this newspaper did it with the deliberate intent of letting me know that our security prohibitions were not so effective as we thought.
After the war the excellent co-operation of the American press continued. Articles written by anyone with access to classified information were invariably cleared with my office. In one instance a newspaper learned of a rather large construction job in progress and wrote a series of articles about its most unusual character. Although the publisher of the paper completely disagreed with me, he canceled the articles before publication when I said that I thought it would be injurious to the best interests of the United States to publish them.
1 By the end of the war the MED’s force of “creeps,” as they became known, numbered 485.