• Email
  • Single Page
  • Print

Were the Atomic Scientists Spies?

A number of other wartime Soviet intelligence efforts to obtain information and documents on the Manhattan Project are known to have taken place. Among the people used by the Soviets were a scientist named Clarence Hiskey at the Metallurgical Laboratory in Chicago, where the first reactor designs were produced; a soldier stationed at Los Alamos, David Greenglass, who passed crude drawings of plutonium bomb design on to his sister and brother-in-law, Ethel and Julius Rosenberg; and the British scientist Alan Nunn May, attached to a joint American-British heavy-water project in Canada, who provided the Soviets with an actual sample of U-235. A young colleague of Enrico Fermi in Rome in the 1930s, the Italian physicist Bruno Pontecorvo, defected to the Soviet Union in 1950 and security officials concluded he had probably been working for Soviet intelligence during the war as well. Best known, of course, is Oppenheimer himself, who was the target of an effort directed by a Communist Party organizer in San Francisco, Steve Nelson, who sent confederates seeking information in the early days of the Manhattan Project. Oppenheimer rebuffed another approach from his friend Haakon Chevalier, but failed to give a full and candid account of it to security officers, thereby arousing a storm of suspicion which never died.

On a train trip from Cheyenne, Wyoming, to Chicago in September 1943 Oppenheimer confessed to General Leslie Groves, military director of the Manhattan Project, that “he had probably belonged to every Communist front organization on the West Coast.”6 Oppenheimer was married to a former Communist, was the brother of a former Communist, had conducted a love affair with a Communist, and had admittedly been approached by Communists, sent by Communists, seeking classified data about the project. But Oppenheimer insisted to Groves that he was not a Communist himself and could be trusted, and Groves believed him. Groves’s chief security officer on the west coast, Colonel Boris Pash, emphatically didn’t believe him, and he refused to grant Oppenheimer a security clearance until Groves issued a written order for him to do so. Historians are in general agreement that Oppenheimer was stripped of his security clearance in 1954 primarily for his failure to support the H-bomb program enthusiastically, not because he couldn’t be trusted to keep secrets.

But the record of Soviet intelligence efforts to find out about the bomb includes a recent, still unresolved, claim. In 1992 the aged Anatoli Yatskov, who has since died, claimed in a series of interviews with Michael Dobbs of The Washington Post that there was another major Soviet spy at Los Alamos, a physicist given the code name of Perseus. Yatskov claimed that Perseus was in contact with Soviet intelligence in 1942, more than a year before Fuchs’s arrival, that the courier who contacted Perseus in New Mexico was Lona Cohen, a figure of importance in other Soviet espionage cases; and that Perseus was still alive.7 Precisely what secrets Perseus stole and when he stole them Yatskov did not say. This story set off a frenzied search for Perseus by historians and journalists, and resulted in some unpleasant moments as spotlights of suspicion were directed at aging scientists living in quiet retirement. But Yatskov’s claim was too vague, and provided too few clues, and Perseus remains without a name, if indeed he existed at all. This, briefly described, is the roster of known Soviet intelligence agents or assets to which we must now add the names of Oppenheimer, Fermi, Szilard, and Niels Bohr, if the authors of Special Tasks can be believed.

The lives of intelligence officers involved in espionage are built around cases—either the recruitment and running, or the discovery and exposure, of agents engaged in secret work. These cases are always highly particular, and the file of even the most routine case can run to hundreds or thousands of pages. Until they are claimed by death or Alzheimer’s disease, intelligence officers can usually recount their cases in painstaking, voluminous detail—just how and when the agent was spotted and recruited, what he obtained, how he was handled and looked after, when and how the case was terminated, every stage of it accompanied at the time by minute analysis and much conjecture. Years are sometimes spent in supervising a single case; every detail involving the assignment of tasks and making contact may be hashed out in committee meetings lasting late into the night. Intelligence officers joining an operation already in progress may spend weeks just reading into the case. They learn their cases backward and forward; success or failure is usually the result of details anticipated or overlooked. When an intelligence officer feels free to recount an old case the narrative can last for hours.

What distinguishes the account of atomic espionage presented in Special Tasks is its complete lack of the establishing and supporting details that are the signature of genuine espionage cases. When charges are made almost nothing is offered by way of circumstance, and in the very few cases where details are cited they are irrelevant, misleading, or blatantly wrong. “It is in the record,” the book says, “that on several occasions they [Oppenheimer, Fermi, Szilard, and Szilard’s secretary] agreed to share information on nuclear weapons with Soviet scientists.” This sounds vaguely like espionage, but no prosecutor could frame an indictment without knowing what sort of agreement was reached, with whom, on which occasions, for sharing what information. If Special Tasks is to be taken seriously it must provide such particulars; without them the most far-reaching charges evaporate, and we are spared the spectacle of a libel case only by the Anglo-Saxon common law precept that one cannot libel the dead. “Giving secrets to the Russians” is not espionage; knowingly passing a copy of a classified document or classified information to an unauthorized person on or about a certain date is espionage.

Only two charges brought in Special Tasks even approach this standard for espionage, one against Oppenheimer in 1941, and a second against Bohr in 1945. The first claims that Oppenheimer told a Soviet diplomat, Gregory Kheifetz, over lunch in California in December 1941 about a letter to President Roosevelt from Albert Einstein urging a research effort to study the feasibility of making atomic bombs. It is possible but far from certain that Oppenheimer knew about this letter, written in the summer of 1940. In any event the letter contained no secrets, and was not itself an official secret. Oppenheimer had no official position with any secret program in December 1941, and the Manhattan Project did not yet exist. Lest this seem a pettifogging defense I ought to add that I do not believe the lunch ever took place, or that the authors of Special Tasks can provide a lucid taped or videotaped claim by the elder Sudoplatov dating or describing it.

The rest of the charges against Oppenheimer in Sudoplatov’s book tend to evaporate on scrutiny. The principal ones are as follows: (1) that he deliberately recruited Fuchs to work at Los Alamos, but the huge record on the British mission establishes beyond doubt that Oppenheimer had nothing to do with bringing Fuchs either to America or to Los Alamos; (2) that he allowed Fuchs to persuade him to oppose the building of the hydrogen bomb, but the question did not come up until three years later; (3) that he allowed himself to be talked into unspecific acts of treason by his wife, Kitty, who was herself under the influence of Elizabeth Zarubin, wife of the chief Soviet intelligence officer in Washington, Vassili Zarubin; but Kitty was in Berkeley, Los Alamos, or Pittsburgh throughout the war, and there is no evidence Mrs. Zarubin left Washington until she left for good in 1944; (4) that he deliberately made secret documents available (by leaving them out on his desk at night, according to Leona Schecter on the MacNeil/Lehrer show) to a Soviet spy (“mole”) he had himself placed in the Los Alamos laboratory, but no evidence including the name of the “mole” is presented to support this.

Fermi and Szilard are also charged with having arranged, in secret concert with Oppenheimer, to give secret documents to young moles they themselves placed in their several laboratories. It is separately charged that Szilard’s “secretary” was working for the Soviets. In fact Szilard had no secretary, but called upon the stenographic pool at Chicago’s Met Lab. The authors of Special Tasks appear to believe that Szilard worked at Los Alamos—he was never there during the war—and was close to Oppenheimer; in fact they met only once, in May 1945, according to William Lanouette, Szilard’s biographer, and they did so in the office of General Groves.

These moles all remain unidentified, and nothing is said of when they went to work or what documents they spirited away. Leona Schecter’s claim that Oppenheimer and his colleagues left documents out on their desks at night betrays a deep ignorance of the security measures prevailing throughout the Manhattan Project, which included nightly checks to see that doors and safes were locked, trash baskets were empty, and desk tops were clear. Sudoplatov as an intelligence officer would have known that any attempt to leave documents out at night would have instantly attracted the attention of security officers. The basis for Mrs. Schecter’s extremely implausible claim is never made clear.

Other errors are just as glaring. Special Tasks alleges that Gregory Kheifetz reported in December 1941 that Oppenheimer told him he “and his colleagues were planning to move from Berkeley, California, to a new site to conduct research in nuclear weapons.” In fact, the first proposal to send scientists to a remote laboratory site came nearly a year later, and the move itself did not occur until the spring of 1943. Kheifetz simply could not have sent such a report in late 1941. How then did it get into the book?

An even more troubling claim concerns Bruno Pontecorvo, the young Italian physicist who had worked with Fermi in Rome in the 1930s. Only hours after Fermi succeeded in creating the world’s first self-sustaining chain reaction in Chicago on December 2, 1942, Pontecorvo is said to have reported the news to his Soviet case officer. The report was “a prearranged telephone message saying, ‘The Italian sailor reached the new world.’ ” This story is suspect, to say the least. In the first place Pontecorvo was not in Chicago at the time of the experiment, but working as an acoustics expert with an oil-drilling rig in Oklahoma. But the account raises deeper questions of veracity as well. According to one of the best-known stories in the history of the Manhattan Project, Arthur Compton, director of the Met Lab where Fermi’s experiment was conducted, telephoned James Conant, president of Harvard University, after the reactor became self-sustaining to report the success in a transparent code, which he said he invented on the spur of the moment:

  1. 6

    In the Matter of J. Robert Oppenheimer (US Government Printing Office, 1954; MIT Press, 1971), p. 59.

  2. 7

    Michael Dobbs, “How Soviets Stole US Atom Secrets,” Washington Post, October 4, 1992. Yatskov’s claim that Lona Cohen was the contact for Perseus in New Mexico may have a bearing on the Rosenberg case; Harry Gold, in his confession to the FBI, claimed that Yatskov asked him to contact David Greenglass in Albuquerque in June 1945 because Greenglass’s regular contact was unavailable. No comprehensive general history of the Soviet atomic espionage efforts exists, but accounts of the different episodes may be found in Robert Lamphere, The FBI-KGB War: A Special Agent’s Story (Random House, 1986); Philip Stern, The Oppenheimer Case: Security on Trial (Harper and Row, 1969); and David J. Dallin, Soviet Espionage (Yale University Press, 1955).

  • Email
  • Single Page
  • Print