Inside the Centre: The Life of J. Robert Oppenheimer

by Ray Monk

  The Serbers’ immediate neighbours in this duplex were Robert and Jane Wilson, who had recently arrived from Princeton. Wilson was there as head of the Cyclotron Group, part of the Experimental Physics Division led by Robert Bacher. His participation was crucial, not only because he was one of the leading experimental physicists in the field of neutron research, but also because he brought with him Princeton’s cyclotron, one of the very few accelerators that the new laboratory had at its disposal. The others were two Van de Graaff generators from Wisconsin, which were put at the disposal of the Electrostatic Generator Group led by J.H. Williams from the University of Minnesota, and Manley’s own Cockcroft–Walton accelerator, which accompanied him from Illinois and provided the data that Manley’s ‘D-D Source Group’ used to work out what material would form the best ‘tamper’.

  None of these machines was up and running until June 1943, which is when experimental physics at Los Alamos really started. Getting the machines to Los Alamos and then setting them up in an as-yet-uncompleted laboratory at the top of a mountain in a remote part of New Mexico was so difficult that it seemed to some of those charged with accomplishing it an almost insane plan. To get to Los Alamos from Santa Fe required crossing the Rio Grande at a place called Otowi, where there was what Serber has described as ‘a toy one-lane suspension bridge that looked as if it might be safe for two horses’. ‘It was hard to believe,’ writes Serber, ‘that all the construction trucks for Los Alamos had to cross that bridge and then climb 1,500 feet up a perilous switch-backed dirt road to the top of Los Alamos Mesa.’

  Among those who had their doubts about the wisdom of building the laboratory in such a place was John Manley, whose job it was to work with the army engineers on the design and construction of the laboratories that would house the accelerators. Manley recalls that he was particularly concerned with the long, narrow building that would house the two Van de Graaff generators and the Cockroft–Walton machine. As he told the engineers, the Van de Graaffs were extremely heavy machines that would need a good, strong foundation underneath them, while the Cockcroft–Walton was a tall vertical machine that required a basement. Given these requirements, Manley writes: ‘Cost and construction time could obviously be saved if they selected the terrain properly’ – that is, the building should be built over a slope, wih the Cockcroft–Walton machine in the lower, deeper part of the building. When he went to inspect the buildings, however, Manley discovered that, instead of making use of the sloping terrain, the engineers had needlessly created their own slope by actually digging a basement for the Cockcroft–Walton machine and then using the resulting debris to make the foundation for the Van de Graaffs. ‘That was my introduction to army engineering,’ Manley remarked.

  Part of the point of building the new laboratory in a remote part of New Mexico was to keep it from prying eyes, but, of course, in some ways it was far more conspicuous there than it would have been in a large centre of population. In a small town like Santa Fe, the arrival of dozens of strangers could not possibly go unnoticed. In fact, as locals were quick to see and remark upon, there were two distinct kinds of strangers descending on their town: first, there were the young bohemian-looking characters with open-necked shirts, who seemed polite, if a little unworldly; and then there were the besuited, slightly threatening men in fedora hats who invariably went around in twos and had a watchful, furtive demeanour. That the first group were scientists was rather less obvious than that the second group were security agents.

  Most of those agents would have been working for the army, rather than the FBI. In March 1943, the FBI was explicitly ordered by Major General Strong, the head of G-2 (the branch of the US army concerned with counter-intelligence), to close its file on Oppenheimer. Security issues relating to anyone – even civilians – working on a military project, Strong insisted, were the responsibility of the army. Astonishingly, the FBI was not officially informed about the Manhattan Project until April 1943, after it had learned of its existence through their surveillance of Communist Party leaders. Though under orders to confine themselves to civilians, FBI agents would inevitably often find themselves covering the same ground, even following some of the same people, as G-2 agents, and, despite the reluctance of military intelligence to confide in them, the FBI was generally quick to inform G-2 of anything that might concern them. Complementing these two security agencies, and sometimes causing further complications, was the Manhattan Project’s own security organisation, which, though officially part of G-2, was under the direct command of General Groves, and thus to some extent separate from it. To start with, this organisation consisted of just a few men, whose main job was to liaise with G-2 and the FBI, but by the autumn of 1943 it was large enough for Groves to insist that it took over all security responsibilities relating to the project. As Los Alamos, Oak Ridge and Hanford continued to expand, so did the Manhattan Project’s own security force, so that by the end of the war it had working for it nearly 500 ‘creeps’, as the agents came to be called.

  The man whom Groves chose to head the Manhattan Project’s own security force was John Lansdale, who, after a series of rapid promotions, was by this time a Lieutenant Colonel. Lansdale worked closely with Groves, who evidently shared Conant’s high opinion of him. Like Groves, he was based in Washington, but most of the people working for him were based on the West Coast. At Berkeley, for example, Lansdale set up a secret, disguised office, run by Lieutenant Lyall Johnson, which became the centre of a covert surveillance operation, keeping a watchful eye on the research scientists working at the Rad Lab.

  Before the Manhattan Project took complete responsibility for its own security, there was something of a turf war between John Lansdale and G-2’s head of counter-intelligence for the West Coast, Lieutenant Colonel Boris Pash. Pash was a formidable figure. Even by the standards of military security officers, he was passionately and belligerently anti-communist, his antagonism fuelled partly by his family history and his personal experience of fighting the Bolsheviks in Russia. He had been born in the US, but was from a Russian family (his father was a Russian Orthodox bishop, based in San Francisco) and had gone to Russia during the civil war that followed the revolution in order to fight alongside the White Army. Since America’s entry into the Second World War, Pash had been an enthusiastic and dedicated member of the US army’s counter-intelligence division, welcoming the opportunity to hunt down Soviet spies, among whom, he was convinced, was Oppenheimer himself – a conviction that remained with him throughout the war and beyond. When, after the war, newspapers reported the spying activities of the Soviet spy Klaus Fuchs, Pash was reported as remarking that he ‘would next be reading about Dr Oppenheimer’s involvement in such activities’. When he was asked at Oppenheimer’s security hearings in 1954 whether in 1943 he considered Oppenheimer a security risk, he replied straightforwardly: ‘Yes I did.’

  From his office in San Francisco, Pash orchestrated an intense security effort to keep Oppenheimer under surveillance: his phone was tapped, microphones were installed in his home, agents were employed to act as his chauffeur, and wherever Oppenheimer went, G-2 men followed. At Los Alamos, the G-2 man in charge of security was Captain Peer de Silva, who was as convinced as Pash that Oppenheimer was a security risk, and who was under orders from Pash to keep Oppenheimer under the closest possible scrutiny.

  It is unclear how aware Oppenheimer was of the intensity with which he was being scrutinised by security agents. It is said that the agents who served as his drivers were thwarted in their efforts to hear what he was saying to his fellow passengers because of his habit of winding the window down in order to create a wind noise that would drown out his conversation. This may have been a clever ploy to prevent his conversations from being heard by men he knew to be security agents, but it seems equally possible that, on the assumption that he thought his drivers were civilians, it was a perfectly sensible precaution.

  When he arrived at Los Alamos, Oppenheimer was in the extremely odd and vulnerabl
e position of having been appointed director of the most secret laboratory in the country while still not having the security clearance that would normally be a prerequisite for taking up even the most junior appointment in that laboratory. He and Groves seemed to take the view that he could start work before receiving his clearance, on the assumption that it would eventually be granted. Pash and de Silva were rather of the opinion that Oppenheimer’s request for clearance should be refused and that he should be removed from the bomb project as soon as possible.

  Two weeks after Oppenheimer moved to New Mexico, at a time when he and his colleagues were still making hurried preparations for the opening of the new laboratory, Pash received news from the FBI that would, he thought, finally convince the authorities that his suspicions of Oppenheimer were well founded. It concerned a conversation picked up by the ‘technical surveillance’ of Steve Nelson’s home between Nelson and a man known to the FBI at this time only as ‘Joe’ (which is how Nelson addressed him in the conversation). As the FBI would discover two months later, ‘Joe’ was Oppenheimer’s friend and ex-student Joe Weinberg.

  The conversation took place in the early hours of the morning on 30 March 1943. Weinberg had arrived at Nelson’s house the previous evening, telling Nelson’s wife, Margaret, that he had some important information to pass on – so important that he was prepared to wait several hours for Nelson to return home in order to discuss it with him. As Nelson discovered when he got home, the information Weinberg had was indeed of great importance and was bound to be of enormous interest to the Soviet Union: people engaged in the new weapon project (which, Weinberg thought at this time, would include himself) were about to be relocated to a remote spot where experiments on explosives could be conducted in secret. Clearly feeling nervous and (as he freely admitted to Nelson) ‘a little bit scared’, Weinberg spoke in a whisper when giving Nelson some technical details of the project. The classified information that he passed on to Nelson centred on developments at Oak Ridge, which Weinberg must have heard about from friends who worked at the Rad Lab. The FBI notes on the conversation at this point become a little sketchy – it was evidently difficult to hear what Weinberg was saying – but the gist is clear enough. A separation plant, Weinberg told Nelson, was already being built in Tennessee that was expected to employ thousands of people, the separation method being ‘preferably that of the magnetic spectrograph with electrical and magnetic focusing’. Towards the end of the conversation Weinberg discussed with Nelson how he might be able to provide information in the future via his sister, who lived in New York, and Nelson emphasised how important it was not to put anything in writing.

  The conversation left no doubt at all that Weinberg was willing, indeed eager, to play an important role in Soviet espionage. And though the FBI did not yet know who ‘Joe’ was, they did know that he was an ex-student of Oppenheimer’s. That much was clear from the conversation, in which Oppenheimer was mentioned several times, usually referred to as ‘the professor’. Pash clearly thought the mere fact that Oppenheimer was associated with two people plotting espionage would be enough to establish him as a security risk, but in fact the FBI notes of the conversation provide pretty good grounds for thinking that Oppenheimer was not a risk. Whenever ‘the professor’ came up in the conversation, either Nelson or Weinberg (or both) made some comment to the effect that he had cut his links to the Party and that he was emphatically not prepared to pass on secrets to the Soviet Union.

  At one point, Nelson remarked that Oppenheimer was ‘very much worried now and we make him feel uncomfortable’, to which Weinberg responded by saying that Oppenheimer kept him off the project because he was worried that he would ‘attract more attention’ and also because ‘he fears that I will propagandize’. Oppenheimer, Weinberg told Nelson, had ‘changed a bit . . . You won’t believe the change that has taken place.’ Nelson agreed, saying: ‘To my sorrow, his wife is influencing him in the wrong direction.’ Evidently on the basis of his recent farewell lunch with Oppenheimer, Nelson told Weinberg that Oppenheimer, encouraged by his wife, was keen to dissociate himself from his former colleagues in the Communist Party, because he did not want to threaten his central role in the important project to which he had been recruited.

  Even if it did not implicate Oppenheimer, this conversation between Weinberg and Nelson provided irrefutable evidence of a threat to the security of a top-secret military project, and, as such, it was taken very seriously indeed by the FBI, to whom ‘Joe’s’ information was as much a revelation as it was to Nelson. The FBI immediately delivered a transcript of the conversation to Colonel Pash, whose response was to fly to Washington to tell Groves and Lansdale that he had evidence of Oppenheimer’s involvement in espionage.

  Of course, what Pash had was fairly conclusive evidence that Oppenheimer – much to the disappointment of his former friends in the Communist Party – was not involved in espionage. Indeed, from the point of view of incriminating Oppenheimer, the conversation between Nelson and ‘Joe’ did not tell Groves and Lansdale anything new; it simply confirmed what they already knew about him, and what they had discussed with each other many times, namely that he had a history of close associations with communists. Lansdale later recalled that when he and Groves first looked through Oppenheimer’s FBI file (he could not remember exactly when this was, but thought it was while Los Alamos was still being built, so probably sometime in the first two months of 1943), Oppenheimer’s political history caused them ‘a great deal of concern’ and they discussed it at length. ‘General Groves’s view, as I recall,’ Lansdale said, ‘was (a) that Dr Oppenheimer was essential; (b) that in his judgment – and he had gotten to know Dr Oppenheimer very well by that time – he was loyal; and (c) we would clear him for this work whatever the reports said.’ So, on the question of Oppenheimer’s loyalty, General Groves had already firmly made up his mind, and he was a man who trusted his own judgement. Nothing short of incontrovertible evidence that Oppenheimer was a security risk would make him drop his conviction that Oppenheimer was the man to get the job done.

  However, it was now clear that the Soviets already knew much about the American atomic-bomb project (more, for example, than the FBI knew at the time) and that, unless the flow of information was stopped immediately, there was a strong possibility they would very soon know a good deal more. It thus became a matter of urgent importance to discover the identity of ‘Joe’ and prevent him having any further access to sensitive information. So seriously did G-2 take this that they immediately established a closer working relationship with the FBI. Thus, on 5 April 1943, General Strong met J. Edgar Hoover’s assistant, E.A. Tamm, to inform him officially of the existence of the Manhattan Project. The following day, Groves and Lansdale met with two representatives of the FBI to discuss ways in which the two security organisations might cooperate in order to establish ‘Joe’s’ identity and protect the project from Soviet espionage.

  A few days before those meetings the FBI had already gathered some counter-intelligence that was, they now realised, of immediate interest to G-2’s attempt to maintain the security of the Manhattan Project. In response to the conversation between Nelson and ‘Joe’, they had decided to keep Nelson under constant, twenty-four-hour surveillance, and on 1 April their agents had seen him walk to a corner shop, from where he phoned the Soviet consulate in San Francisco to arrange a meeting with Ivanov. When the meeting subsequently took place, on 6 April, FBI agents were there to observe it. Then, on 10 April, FBI agents watching Nelson’s house noted the arrival there of none other than Vasily Zubilin, the head of the NKVD espionage operation, who was based at the Soviet embassy in Washington. The microphones inside the house picked up a long conversation between the two about the structure of the Soviet espionage operation and the respective roles played within it by the American Communist Party and the NKVD (Nelson was worried that the former was being bypassed by the latter). Agents also heard Zubilin counting out large amounts of money to give to Nelson, who exclaimed: ‘J
esus, you count money like a banker.’ Presumably Nelson and Zubilin soon realised that US counter-intelligence was on to them, because this was the last time that either was recorded as having anything to do with espionage.

  While the security forces were trying to discover his identity, Weinberg managed to insinuate himself into a position in which he would have access to secret information. Sometime in April 1943, Oppenheimer, despite his earlier reservations about hiring Weinberg, employed him at the Rad Lab to work on some calculations that were part of the effort to improve the focusing of the beam of the Calutron. Of course, Oppenheimer knew nothing of the recorded conversation between Weinberg and Nelson. However, Pash and de Silva later cited Oppenheimer’s willingness to employ Weinberg as evidence of his complicity with Soviet espionage. Lansdale, on the other hand, when asked years later about this period, not only did not see anything suspicious in Oppenheimer’s behaviour, but went out of his way to praise Oppenheimer for being ‘very helpful’ in the attempt to impress upon his fellow scientists at Los Alamos the importance of maintaining strict security.

  ‘The scientists en masse presented an extremely difficult problem,’ Lansdale said, adding: ‘I hope my scientist friends will forgive me, but the very nature of them made things difficult.’ Scientists, by their very nature, like to share information, which put them somewhat at odds with the people whose job it was to ensure that information was not shared. From both sides there was, from the very beginning, mutual incomprehension. In the many reminiscences of Los Alamos written by scientists, the security arrangements are almost invariably regarded with a mixture of contempt and amusement. Robert Serber, for example, describes the initial attempts to secure the Los Alamos site as comically lax and amateurish. ‘Oppie,’ he writes, ‘wrote passes for us on University of California stationery which didn’t well survive being carried in hip pockets.’