I find it inconceivable that any information I could reveal about our methods over forty years ago could be of the smallest value to a potential enemy today or in the future – but that would be no defence if I were arrested for breach of the Official Secrets Act on stepping onto British soil.4
Peter Twinn, one of the first to join Knox’s team, had been in correspondence with a series of Directors of GCHQ for years seeking, without success, permission to write the history of the solution of the Enigma. He had been somewhat disappointed to see Welchman’s book describing the breaking of the Army and Air Force Enigma traffic and Andrew Hodges’s biography of Turing, describing the breaking of the naval traffic, appear in print. Twinn felt that he could have filled in considerable detail about GC&CS’s relations with the Polish cryptanalysts from February to September 1939. He could also have written the whole of the Abwehr Enigma story, which was quite different from that of the German military Enigma. He asked GCHQ why Welchman and Hodges had been able to publish their accounts. The answer he received was that as Welchman was now an American citizen, they could not stop him and Hodges had never signed the Official Secrets Act and they did not know his sources. Twinn would not publish his own brief account until 1993, in the same book as Stuart Milner-Barry, Hinsley and Stripp’s The Codebreakers.
Welchman had thought long and hard about the implications of publishing his book without the approval of GCHQ or the NSA. All advice he received from former colleagues who still had close connections with GCHQ, indicated that, even by 1982, it was unlikely to give him approval to publish. Inquiries within MITRE and elsewhere in the American intelligence community failed to come up with a way to approach the NSA which would lead to anything other than indefinite delay. Welchman revealed his thoughts while working with Neill Rosenfeld from McGraw-Hill in a private memo to himself:
Finally, as I said, I would like if possible to avoid bad feeling with GCCS and NSA. Above all, I cannot risk trouble with MITRE. My book must in no way run against the position of MITRE with military funding. Nor must it result in MITRE being told to get rid of me. (I am too old to get established elsewhere.)
In the end he had to make the decision himself. He believed that the ideas that occurred to him at BP during the first months of the war could not possibly be regarded as classified material in the United States. He passionately believed that publishing his book was doing something that was important for the future national security of the US. The alternative was submitting his manuscript to GCHQ and an almost certain veto. It seemed quite clear that, as a US citizen, the future security of his country must come first. So, late in 1980, he decided to go ahead with publication.
Following its publication in the USA in February 1982 a copy of The Hut Six Story was sent to Rear Admiral Ash of the D-Notice Committee in the UK. Ash had one meeting with Welchman’s British publishers, Allen Lane, at which he expressed concern, but he made no official comment. Welchman never received any formal complaint from GCHQ about his book.
Shortly after publication, Welchman received a letter from William Stevenson, author of the bestselling book A Man Called Intrepid. Stevenson’s book about Sir William Stephenson had been a great success and even spawned a miniseries starring the well-known actor David Niven as Stephenson. Welchman had originally found the book an interesting source of information, but eventually became disenchanted with its many inaccuracies.5 In his letter, Stevenson thanked Welchman for acknowledging Stephenson’s accomplishments and expressed a concern about ‘the vigorous efforts to diminish his role, especially in the field of communications’. On 23 July Welchman received a telegram from Sir William Stephenson himself offering congratulations on his book.
After arranging for autographed copies of his book to be sent to his son, two daughters and other family members, as well as the long list of people whom he had consulted during its writing, Welchman was planning to develop his ideas further at MITRE. He also had several new writing projects in mind but, before committing to them, he wanted to spend a bit more time with friends and family. However, at the age of seventy-six, he had no intention of giving up just yet. His love of music had led him and a friend, Dick French, to give monthly concerts at three local nursing homes. This was mostly classical music that they recorded on cassette but they usually ended with something light-hearted like ‘Yes, We Have No Bananas’.6
On 20 April, Welchman was visited at his home by two USAF special agents and an NSA representative. Much to his surprise, he was told that the ideas that had occurred to him in the first three months of the Second World War were still regarded as classified in the USA. They wanted to interview him formally to find out why he had published this information without first obtaining permission to do so. The next day he met the Corporate Security Officer of MITRE and learned that the top security man in the US Department of Defense had written an unpleasant letter to the president of MITRE complaining that one of his employees had published classified information. He also learned, both from his lawyer and from the MITRE security man, that there was an old agreement under which any cryptological information that was classified in Britain was automatically regarded as classified in the USA.
The interview was held on 22 April, with his lawyer John Stevens present, and Welchman gave an on the record account of how the book came to be written and published. It was a very pleasant discussion and he had the strong feeling that, by the end, all three interrogators were in sympathy with his contention that he had acted in what he believed, with good reason, to be in the best interests of US national security. One of them even asked him to autograph his copy of the book. After the interview, Stevens felt quite confident that there was very little likelihood that the US Government would pursue criminal charges against him. His own impression was that this enquiry had been triggered by a request from GCHQ based on the old international agreement. They were told to expect a decision on the matter by 6 May.
On 29 April, Welchman’s security clearance was rescinded and he could no longer enter MITRE’s premises without an escort, which severely interrupted his work. He began preparing statements that might be needed if he had to defend myself. From what his interviewers had said and from the fact that McGraw-Hill had not been approached by the Department of Defense, it seemed that, if any action was to be taken, it would be against him personally. There seemed to be no intention of interfering with publication of the book.
The investigation by American special agents was under an international agreement that British ideas on cryptologic security could be enforced in the United States. This document, 18 USC §798, had been drafted in the late 1940s and, if convicted, Welchman could face a fine of not more than $10,000 or imprisonment for not more than ten years, or both. It was part of a US code titled 18 USC Chapter 37 – Espionage and Censorship. Section §798 covered ‘Disclosure of Classified Information’ and prosecution was possible because he had published material ‘concerning the nature, preparation, or use of any code, cipher, or cryptographic system of the United States or any foreign government’.
When Welchman had joined MITRE in 1962 he had signed a statement to the effect that he had been given an initial security briefing. He felt quite certain, however, that this briefing had not included any mention of 18 USC §798. As his work for MITRE had never required a special clearance for cryptological matters, there would have been no need to do so. When he had made his decision to publish The Hut Six Story, it was without knowing that in the field of cryptography, information regarded as classified in Britain would automatically be regarded as classified in the United States
At the end of his first interview, Welchman was told to be on the lookout for enemy agents who might try to extract from him additional information on matters discussed in his book. In view of his long experience of handling secret information, it seemed highly unlikely that he would be caught out in this way. To be on the safe side, he decided to ask for a definition of what the authorities were afraid he might reveal. In response to
this request, the Chief of the Information Security Division, Central Security Service, National Security Agency, M. J. Levin, sent MITRE a one-page document on 21 May.
18 May 1982
Guidance for Mr.Gordon Welchman
If questioned or interviewed by the media or others relative to the information contained in The Hut Six Story:
You may discuss:
– the history, organization, geographic setting of Bletchley Park;
– the identities of persons involved in the wartime effort at Bletchley;
– the ‘Enigma’ machine in broad generalities;
– the effect of the ‘breaking’ of the ‘Enigma’ on the war effort;
– cryptologic collaboration between December 7, 1941 and May 8, 1945;
– the significance of communications in military or other operations.
You may not discuss:
– technical details of the structure, logic or operation of the ‘Enigma’ machine, or other similar machines;
– weaknesses of German or other communications systems and the methods of exploitation of such weaknesses;
– methodologies which cryptanalysts use to successfully exploit code or cipher systems, except for those which have been officially declassified by the United States;
– methodologies used to provide protection to U.S. or Allied communications;
– details of any cryptanalytic or cryptographic methods, except for those officially declassified by the United States.
Initially, Welchman thought that it was the NSA who were the ‘troublemakers’ and concerned about the last section of his book. This dealt with lessons which military forces of the day could learn from the BP story. Ironically, senior staff at MITRE had encouraged him to write this part of The Hut Six Story! Most of the cryptological material in the book was already in the public domain in some form or other, apart from his description of menus for the bombe and the diagonal board. He had invented the diagonal board at a crucial stage in BP’s battle against the Enigma machine and his account of it was the first to appear in print. If this was the cause of concern by the intelligence people, it struck him as highly ironic. Before he had become ill Alastair Denniston had called him into his office, congratulated him on his achievements, particularly the diagonal board, and assured him that he would be rewarded. Denniston’s successor, Edward Travis, had also promised that he would be rewarded after the war. He was approached by Travis in 1948, who said that he was very worried about Turing, as well as Welchman, and that he was determined to do his best for them. Unfortunately, Travis got ill at that point and the matter was left in the hands of a deputy, whom Welchman only knew by the name ‘Willy’. The man arranged a meeting at which Hugh Foss and Harold Fletcher were to persuade him that he did not deserve any reward. Travis had recommended that his claim be based on the diagonal board. The original drawing that he had shown to Turing was produced, and it clearly demonstrated that the idea was his. However, ‘Willy’ told them that he would not be given a reward for two reasons. First, an award had already been given to Harold Keen for inventing the diagonal board. Second, his idea was the sort of thing that could be expected from a Foreign Office employee at his salary level, which was £600 a year. Foss had pointed out that this idea was by no means his only contribution, but it had no effect. Later on, Welchman asked to see his original drawing again, but was told that it had disappeared.7
May 6 came and went and the promised decision by the US intelligence services did not materialize. What really upset Welchman was that throughout the investigation, there had been no mention of the value to the US of what he had done during the war, what he had done since and what he could still do. After being interviewed again on 28 June, he decided to document the testimony that he had given and completed a paper called ‘The Story of My Story’ on 30 July.
Previously, on 1 June 1978 he had sent B. J. Workman, his line manager at MITRE, two memos: ‘Wartime Experiences and General Background’ and ‘Tentative Ideas for Further Study’. He decided to send copies to MITRE’s Corporate Security Manager, Robert J. Roberto, to show the origins of his MITRE research and the relevance of his wartime work to it. Levin at the NSA wrote again on June 30, saying that his guidance did not apply to a government-approved classified environment. He claimed that it had not been intended in any way to inhibit his contacts or discussions with appropriately cleared MITRE personnel, subject to whatever ground rules MITRE might apply. In view of this official statement from the NSA, it was hard to see why he was still prevented from continuing his work as a MITRE consultant. In exasperation, he wrote to Levin again and cited an incident which showed up the absurdity of his situation.
In January 1975, Welchman had received a letter from Brian Randell. Randell was researching the history of computers and had attached a list of questions which he hoped Welchman could answer. One aspect involved a study of the role played by Alan Turing in the development of the stored-program concept.8 Randell’s problem was that this facet of Turing’s work seemed to be entwined with his work at BP during the war. He had no interest in accessing classified information about the cryptanalysis work undertaken, but just wanted to clarify the early history of electronic computers and computer-like devices. In his reply, Welchman acknowledged that he had collaborated with Turing on several technological developments at BP during the war. In his role as Assistant Director for Mechanization, he knew about Turing’s wartime contacts in the US and he had made a trip himself for the purposes of technological liaison. However, he told Randell that he was still not allowed to give a complete account of what he, Turing and their American counterparts were doing at that time. Welchman had to decline further invitations from Randell to contribute to his work apart from providing some detail about his postwar career. Randell went on to write a significant paper which, for the first time, described in some detail, the development of the world’s first electronic computer, Colossus and some aspects of its role at BP.9
Randell wrote again in November with the news that there had been significant changes in UK Government policy concerning the classification of Colossus. GCHQ had made available a set of photographs with explanatory captions and arranged for him to interview Max Newman, whose section at BP, the ‘Newmanry’, had commissioned Colossus. Randell had also been able to interview Tommy Flowers, the lead engineer and creator of Colossus. He had, however, been told that: ‘The details of the use made of Colossus, or such details of its logical design as might reveal the processes it automated, are still regarded as confidential.’
Once again Welchman had to say that he was unable to help Randell because of the ‘increasingly ridiculous clouds of secrecy which still shroud my wartime activities’. In the end, Randell restricted himself to a brief description of Welchman’s post-war work. A number of years later, shortly after Welchman had published his book, Randell was visiting the US. They arranged to meet at the opening of the Digital Computer Museum in Marlboro, Massachusetts, on 10 June 1982. As Welchman said in his letter to Levin:
Early in June, while attending a function at the Digital Computer Museum near here, I was examining the two German Enigmas that they have on public display for anyone to play with. Someone, knowing that I had written ‘The Hut Six Story’, asked me what the plugs in front of the machine were for. I had to explain that I am not allowed to discuss details of the Enigma, even though the answers to simple questions such as this can be found in my book and elsewhere in open literature. Brian Randell, who has written about Colossus, was there and answered the question with more detail than I could have supplied. He has his own German Enigma, a 1940 model, at his home in England and has taken it apart, whereas I had not seen an Enigma since the war, and even then was not too familiar with details. (The description of the Enigma in my book was based on my memory and I now find that it is incorrect in some matters of detail.) I had to cancel a seminar at the Museum, scheduled for June 27, because I was expected to make some remarks on their Enigmas.
In early August 1982, Welchman showed his paper ‘The Story of My Story’ to Bob Everett, President of MITRE, in the hope that he might be able and willing to set up a development overview team to work in the area of communications, intelligence and security. He soon came to realize that this was quite impossible. Any team of highly-qualified people would come up with suggestions that would not be welcomed by the establishment, and MITRE’s very existence depended on pleasing its sponsors. It then dawned on him that the same difficulty would apply to other organizations engaged in military research, and where else would one find people with the necessary background?
In late August, the head of the MITRE department for which he had been working asked him to complete a task that he had begun before he lost his security clearance. The work was unclassified and could be done at home. Its purpose was to draw lessons from his research and development project from what had recently become known about intelligence in the Second World War. The resulting paper ‘Intelligence Aspects of the Joint Surveillance and Target Attack Radar System (JSTARS)’, drew on Lewin’s Ultra Goes to War and The American Magic, Kahn’s Hitler’s Spies, Hinsley’s British Intelligence in the Second World War Volumes I and II and Calvocoressi’s Top Secret Ultra. He worked hard to complete the task before his consultancy contract expired at the end of September. With the work completed on schedule, MITRE Security Services showed its thanks by writing to him on 15 October to remind him that the Department of Defense required that terminated employees should return all picture badges. He had now abandoned all hope of working with a team in MITRE, or in some other establishment engaged in military research and development
Once again, he was on his own and, as six months had passed since the NSA began its investigation, he decided to start offensive action. In October he began working on two documents that he intended to use in an approach to the United States Congress. The first, dated 30 November was entitled ‘The Urgent Need for Development Overview Teams Reporting to Congress’ and its central argument was:
Gordon Welchman Page 23