Stealing the Atom Bomb: How Denial and Deception Armed Israel

by Roger Mattson

  About the time of his FBI interview, Shapiro told the New York Times, “We operated within rules and regulations. I have been told by responsible officials that our operations were pretty much in the ball park, in the range of other similar facilities.”511 His statement contradicted what AEC, NRC and ERDA officials told Congress over the preceding twelve years about the amount of HEU missing at Apollo compared to other facilities. His statement also contradicts what DOE’s declassified material accountability records later showed, i.e., while he was president of the company, NUMEC’s losses as a percentage of throughput and in absolute value exceeded those of all other U.S. commercial enterprises that ever processed HEU, by a wide margin.512

  Shapiro’s statement to the Times about NUMEC’s losses at Apollo being comparable to other plants of its type contradicted what AEC General Manager Hollingsworth wrote to JCAE Executive Director Conway on February 14, 1966. “NUMEC’s cumulative losses from time of plant start-up in 1957 have been higher than those determined by other companies having comparable operations.” Hollingsworth went on to say that the HEU lost on the WANL contract amounted to 6 percent of the total HEU received by NUMEC on that contract. He also provided Conway with a table of “The total cumulative losses of U-235 expressed as a percent of total cumulative receipts. . . .” for five commercial companies, including NUMEC. The losses of U-235 by the four other companies ranged from 0.28 to 0.61 percent of receipts and averaged 0.49 percent, while NUMEC’s loss was 1.21 percent.513 Furthermore, the Apollo plant preferentially lost HEU over LEU, as mentioned above in connection with the GAO audit in 1966.

  This was not the only time Shapiro made overly optimistic statements about operations at Apollo. He was quoted in the Washington Star in June 1977 saying that a statement by an anonymous source that NUMEC was “taking material from one batch and using it to account for losses in a prior batch until the losses became too big to cover up anymore” was “utterly ridiculous and totally untrue. Losses are inherent in the fuel fabrication business because of the highly complex process, which uses uranium in solid, gaseous, and liquid chemical forms and which results in a large amount of waste material at each step. If you’re cooking a small thing in a large pot, then you have material that sticks to the pot.”514 This claim was counter to what AEC found in its 1965/1966 inventory and investigation of the missing uranium at Apollo. Furthermore, Shapiro himself contradicted this statement a week later in his June 27 FBI interview where he said NUMEC blended materials and wastes among contracts. It appears the FBI agents were not attuned sufficiently to nuclear fuels processing to understand the importance of the difference in his statements or they did not read the newspaper. By paying attention to such contradictions, the FBI agents should have been able to probe Shapiro’s statements.

  Accordingly, with no one to question them, Shapiro’s statements had their intended effect. On July 21, FBI’s Washington Field Office reported, “In view of the results of this interview [of Shapiro] and all others conducted in this matter, the opinion of WFO remains unaltered . . . namely, that the lack of cover up of the NUMEC matter has been adequately demonstrated by the extensive investigations conducted” by AEC and GAO. “WFO feels that a polygraph examination of Shapiro is unwarranted on the basis of information developed to date. Consequently, this office considers that all logical leads have been covered and is now closing this case.”515

  ***

  On July 19, 1977 nearly two months after I briefed Udall on the work of the Conran task force and Rowden suggested he talk to the NSC, Udall wrote to Zbigniew Brzezinski, the president’s national security advisor, seeking a briefing by “appropriate officials.”516

  ***

  If Udall had followed up on Rowden’s suggestion to discuss NUMEC with Rickover, he might have received the same reception as the FBI. In May 1979, a hapless FBI agent telephoned “The Admiral” to request an interview to discuss NUMEC. The ensuing exchange, as recorded by the agent, was classic Rickover.517 The requesting agent was directly connected to the Admiral, who was told that the FBI wished to interview him. The Admiral asked why and when told it related to the NUMEC operation at Apollo, Pennsylvania, he became very agitated.

  The Admiral stated that he only knew what had been printed in the newspapers. He demanded to know what questions would be asked so that he could prepare answers. When the requesting agent attempted to set a date and time of interview, Admiral Rickover again shouted that he only knew what he read in the paper and that “I can’t add anything. I have nothing to say.” The admiral, continuing in an agitated state, shouted “You can send a bunch of FBI agents down here and I still won’t have anything to say.”

  Admiral Hyman Rickover

  Director of Naval Reactors

  1949-1981

  Although Admiral Rickover never refused to be interviewed by FBI agents, he refused to set a time and date and based upon his behavior, it did not appear that an interview could be arranged. Admiral Rickover was told by the requesting agent that it was clear no interview could be arranged. The admiral said, “Fine.” And forcefully hung the phone up.

  Approximately five minutes after ending the first telephone call, the requesting agent received a telephone call from Admiral Rickover’s secretary who connected Admiral Rickover to the agent. The admiral, in a calm, rational voice stated, “We sent all that information over to the Department of Energy (DOE) but you (the agent) can come over and review it.” The admiral was told that the FBI would like to come over and review any documents available, to which Admiral Rickover retorted “Well, why don’t you go over to DOE and look at them.” As the admiral was again becoming agitated, it appeared that any further conversation would be fruitless and the conversation was terminated.

  The special agent was jinxed: Seaborg also refused his invitation to be interviewed.

  ***

  Just nine days after Udall asked Brzezinski for a briefing on what the NSC knew about NUMEC, on July 28, 1977, the NSC staff asked CIA for a briefing on “Israel and MUF.”518 The following day, “Ted Shackley called me [John Marcum of the NSC staff] today on a secure line and provided the following responses to our inquiries of yesterday.” The next page and a half of the memo documenting what Shackley told Marcum were redacted by ISCAP when the document was released in March 2014. In the last half page of the letter, Marcum went on to say,

  I also asked Shackley to get us a rundown on the political aspects–e.g., when were the President and Congressional officials briefed on the Israeli weapons program, on the NUMEC connection, and what were their reactions. In December, Carter was briefed on the NUMEC problem as President-elect by Bush in Georgia. I have also heard sketchy accounts of briefings for Johnson and Nixon, but it would be useful to get these details in hand in case there is a Congressional inquiry later. We should discuss next steps on this issue and the MUF [ID] release. At this point, despite the FBI clean bill of health, I do not think the President has plausible deniability. The CIA case is persuasive, though not conclusive.

  Two lines of redaction follow these statements, thus ending the memo.

  ***

  Chairman Udall held a subcommittee hearing on Conran’s complaints on July 29. The Conran task force and the NRC commissioners knew that Udall would explore the NUMEC affair as part of the hearing, in addition to the other matters that Conran had raised in his open letter.

  A few weeks before the Udall hearing, Rowden’s term as commissioner expired and he left NRC, thus ending a nineteen-year government career with atomic energy law. His departure left the commission without a quorum. The two remaining commissioners were Victor Gilinsky and Richard Kennedy. Under this circumstance, the role of representing the agency at the Udall hearing fell to NRC’s Executive Director for Operations, Major General Lee V. Gossick (USAF retired).

  Gossick assembled the NRC staffers that were cognizant of NUMEC matters and Conran’s other concerns. We met on July 13, 20 and 28 to help him prepare for the Udall hearing. In the course of those meeti
ngs, the staff urged him to read a memorandum that the Conran task force had written about the ERDA and CIA briefings of the commission.519 Reading that memorandum was important because Gossick had not attended the ERDA, CIA and Conran task force briefings.

  Major General Lee V. Gossick

  U. S. Air Force

  1941-1973

  NRC Executive Director for Operations

  1975-1980

  That memorandum was specific about the commission’s decision in early 1976 to stop saying there was no evidence of a diversion of special nuclear material. The commission had concluded that it should qualify such statements because of the uncertainty surrounding NUMEC and because of measurement uncertainty inherent in handling large quantities of SNM. Victor Gilinsky also recalled, “I specifically told him about the Duckett briefing before his hearing—in the company of Kennedy and two others, one of whom claimed that Gossick was called away just as I was saying the key words.”520

  In spite of the advice about the commission’s use of qualifiers and uncertainties in describing the evidence of a diversion, General Gossick and Robert Fri, the Acting Administrator of ERDA, testified at the Udall hearing that there was no evidence of a diversion at Apollo. Jim Conran and senior staff of the Division of Safeguards and the Office of Nuclear Materials Safety and Safeguards also attended. Our heads wobbled when Gossick said there was no evidence of a diversion.

  ***

  The same day that he testified to Udall’s subcommittee, ERDA’s Acting Administrator Robert Fri, accompanied by Generals Giller and Starbird, received a briefing from CIA.521 Ted Shackley provided the briefing in Fri’s office. ERDA requested the briefing in connection with its pending August 4 release of the historical summary of strategic SNM inventory differences. In his briefing, Shackley provided a talking paper that Giller recalled “as a basic recompilation of the already-known background information and evidence [gathered by CIA] relating to the issue.” Although DOE, ERDA’s successor agency, redacted much of Giller, Starbird and Fri’s recollections of the briefing, Giller did mention some things not addressed in redacted summaries of other briefings that Shackley conducted in the summer of 1977. For example, Giller “emphasized his belief that, even with careful analysis, the possibility of poor information stemming from HUMINT [human intelligence] sources still remains.” He went on to say, apparently in reference to something Shackley said about the discovery of HEU in Israel, “the U.S. government has made authorized shipments of high-enriched SNM to Israel in the past, which were intended for the Israeli reactor program.”

  No documents have been found to show that the former ERDA officials passed along the new information provided by Shackley. Although much of Shackley’s briefing was redacted from the referenced DOE document, it is logical to assume that he told the ERDA officials the same thing he told the NSC staff on July 28, 1977.

  In September 2015, CIA released a five-page Talking Paper that may have been the one shown to the three ERDA officials on July 29, 1977. Unfortunately, the paper was so heavily redacted that it shows no date or setting for the message it delivered. Unredacted statements on page four of the Talking Paper are of interest in determining who knew what and when. It is clear from this disclosure that CIA did not tell FBI everything it knew about the possibility of a diversion to Israel.

  While this information [referring to 12 lines of redacted information immediately preceding] is of obvious importance in reaching an intelligence decision on the probability of diversion, it is not of any legal pertinence to the FBI’s criminal investigation of NUMEC. In our discussions with the FBI we have alluded to this information but we have not made the details available to the special agents from the Washington Field Office of the FBI who are working on the case.

  ***

  That same day, Congressman Dingell called the National Security Council to say he was “troubled about investigations of Materials Unaccounted For . . . particularly the NUMEC plant. . . . He understands both the FBI and the CIA have been involved.” Dingell asked that the NSC brief two of his staffers, Michael Ward and Donna Levigne.522

  ***

  On August 2, Ted Shackley, Carl Duckett (by then retired from CIA) and (name redacted), a representative of CIA’s Office of Legislative Counsel, briefed Frank M. Potter and Peter D. Stockton of the Subcommittee on Energy and Power of the House Committee on Interstate and Foreign Commerce.523 The Memorandum of Record prepared by Shackley said the Committee Chair, Congressman Dingell, requested the briefing. Shackley pointed out that he was neither a scientist nor a first-hand participant in the events that unfolded from 1968 to 1977 relative to the NUMEC case. He merely provided “supervision to the people who were conducting research on the Agency’s involvement in the NUMEC case.” Shackley said there were gaps in the records. “Mr. Duckett chimed in to say that he personally had been a participant in several key discussions around Washington on the NUMEC issue on which he had written no memoranda for the record, therefore, he could understand the frustrations that were being encountered by Mr. Shackley and his associates in their attempts to fully reconstruct the events which had transpired. . . .”

  The discussion then centered on Duckett’s recollections. Duckett said “CIA had been concerned about the nuclear weapons proliferation issue for a number of years (six lines redacted). As a result, CIA began to look at information which was available to it concerning possible diversion of uranium materials from NUMEC.” Reading between the lines and acknowledging that CIA efforts focus offshore, the Agency must have gathered some intelligence in Israel that there was enriched uranium in the Israeli weapons program that could not have been produced or shipped there. Hence, the Agency went looking for other sources and focused on NUMEC.

  Duckett told the Congressional staffers “as he recalled the situation, approximately 178 kilograms of uranium were missing from NUMEC.” The DOE redacted the next three lines from the memorandum before CIA released it in September 2015. In the next twenty or so lines of the memorandum, most of which are redacted, Duckett told the staffers that this was “his most telling point in his hypothesis.” Duckett said the “CIA was the recipient of FBI and AEC material on the alleged diversion and this data influenced the CIA estimating process on the key issue of did Israel have a nuclear weapon system.” Shackley confirmed that Helms wrote to the Attorney General in 1968 requesting an investigation because of the possibility that a diversion from NUMEC had taken place. Duckett told the staffers “he could say with certainty that CIA, as an institution, had not been involved in any kind of nuclear materials diversion operation.” Duckett then recounted the story Helms told him about the briefing of President Johnson in which Johnson told Helms not to tell anyone else about the Israeli bomb. Although this part of the memorandum is redacted, it is obvious from context that this was the subject of discussion. Shackley supported Duckett’s statement about no involvement of CIA in any diversion of weapons material. Neither Shackley nor Duckett offered evidence to support their conclusion.

  Shackley said he did not know if an FBI investigation of the NUMEC case was currently underway. In response to questions about who would have first-hand knowledge of past investigations of the NUMEC case, Duckett referred the Congressional staffers to George Murphy of the JCAE staff and Richard Kennedy then an NRC commissioner and a former NSC staffer. In response to a question about how to transport stolen uranium from NUMEC to Israel, Duckett gave a two-and-a-half-page response that CIA completely redacted when it released the memorandum. Shackley recorded in the Memorandum of Record that CIA stressed throughout the briefing it “did not have any facts which would stand up in court which could be used to conclusively prove that there was linkage between the alleged NUMEC diversion (six lines redacted).” Shackley probably had no knowledge of the unique uranium enrichment that the Portsmouth gaseous diffusion plant produced for the U.S. naval reactor fuel processed by NUMEC. There were some secrets that Admiral Rickover and the Naval Reactors Office of AEC/ERDA/DOE kept from nearly everyone u
ntil 2006 when the DOE released its history of uranium enrichment in the United States. In any event, the CIA redacted roughly half of this thirteen-page memorandum, making it impossible to know the qualifiers that Shackley placed on his conclusion about “conclusive proof.”

  ***

  On August 2, 1977 National Security Advisor Brzezinski wrote a memorandum to President Carter on the NUMEC affair.29 Brzezinski said he had been briefed by ERDA, FBI and CIA and summarized the “essential conclusions” for the president. He said that all but 56 kilograms30 of the missing material from

  NUMEC could be physically accounted for, but ERDA believes now (but has no evidence) that even this remaining 56 kilograms can be accounted for by operational losses, but this will be a very hotly contested conclusion. The ERDA report also reaches a very carefully guarded conclusion that no evidence of theft of significant amounts of material has been found.

  Two pages of narrative follow this excerpt, summarizing the two FBI investigations and the AEC investigation of NUMEC. The ISCAP redacted about half of that part of the memorandum when it released the document in March 2014.524

  In his memo for Carter, Brzezinski took at face value ERDA’s claim that the AEC had investigated the NUMEC affair. He also passed off the FBI’s findings as amounting to no more than confirmation that Zalman Shapiro had frequent contacts with Israeli officials, including a science attache “thought to be an intelligence officer,” and that Shapiro received unexplained VIP treatment in Israel. Brzezinski either did not know or he knew and did not tell Carter of Shapiro’s other contacts with Israeli intelligence and nuclear weapons officials.

  Brzezinski told Carter the FBI had just concluded its latest investigation and “was unable to uncover any evidence of theft, although the interviews included many current and former NUMEC employees.” In fact, the FBI investigation continued for two more years, and interviews of NUMEC employees revealed suspicious circumstances concerning NUMEC’s shipments to Israel, as described below.