On 18 March 1985, Welchman was informed by Andrew via Denniston that he was proposing to send the article to press on 1 July. Welchman had also decided to send Andrew a copy of the paper he had written about himself and Turing. He had given it to his friend Diana Lucy to take to England along with the final corrections to his paper and photographs of Enigma-related and other documents.21
With the paper ready for publication, Welchman enquired about contracts and fees. As it was 16,000 words, his agent Jane Wilson thought that he would be paid for his work. Much to his disappointment Andrew informed him that no payment could be made as it was appearing in an academic journal. He had not been aware of this and was surprised to learn that he would have no financial return for six months’ work. Andrew cited the example of David Kahn as an academic who published free but then earned money subsequently from book spin-offs. Welchman still had hopes of publishing the paper in book form but, for the moment, decided to issue it in Andrew’s journal. One positive piece of news was that the Penguin paperback could now be sold in the USA because American rights had reverted to Welchman when McGraw-Hill remaindered The Hut Six Story.
At the end of March 1985, Welchman and Teeny were able to have a two-week holiday in Tortola, British Virgin Islands. On returning to Newburyport, he became worried that Andrew had failed to clear the paper with GCHQ and the NSA. He had promised Twinn, Herivel, Chamberlain and Babbage that the paper would be cleared before publication and that they would be protected.
Welchman had been suffering from jaundice on his return from holiday in early April and arranged to see a liver specialist in Boston on the 11th. He thought he might need a gall bladder operation. Never one to remain idle for long, he was already working on ideas for a book that his old McGraw-Hill editor, Bruce Lee, wanted him to write. The subject would be the development of computers and information technology on both sides of the Atlantic up to around 1960.
On 7 May, he wrote a letter while a patient in the New England Deaconess Hospital and sent copies to Robin Denniston, Christopher Andrew, Jane Wilson, Bruce Lee, Cy Deavours and David Kahn. The news was not good. A seven-hour operation on 17 April had revealed several tumours After a few difficult weeks he felt he was making progress and hoping to be home in two or three weeks more. He had been plagued with health problems throughout his adult life and after several heart operations, two complete hip replacements and numerous bouts of debilitating flu, he and his family thought that this was just another problem that he would overcome.
In light of Welchman’s concerns about publishing the paper before getting clearance from the proper authorities, Andrew had submitted it to the D-Notice Committee at the Ministry of Defence. On 22 July he was able to write with the good news that the D-Notice Committee had cleared the article and he attached written confirmation from Major General Kay, Acting Secretary. Apparently, this had been achieved after protracted discussions between the committee and the ‘relevant departments’. This was interesting news as Welchman had received a letter from the Director of GCHQ, Sir Peter Marychurch,22 four days before he received Andrew’s letter. He had come home from hospital on 5 July and, while still very weak, was improving slowly. The letter was extraordinary for several reasons, the obvious one being that Welchman had in his possession written confirmation from the D-Notice Committee that his paper had been cleared by them. It was also hard to understand how the Director of GCHQ, the successor organization to GC&CS, could use such condescending language to a person of Welchman’s stature in the history of his organization.
Several days after writing to Welchman, Marychurch also wrote to Neil Webster, rescinding GCHQ’s previous approval for him to publish his personal memoir about his wartime work in traffic analysis. On 9 August, Andrew replied with an explanation which made the Marychurch letter even more reprehensible.
When I first submitted your article to the D-Notice Committee I received an acknowledgement from Major General Kay promising to get in touch with me again as soon as possible. Some weeks passed. I then had a letter from Major General Kay containing no comment on your article but saying that it was up to you to clear the article directly with your former department. This response clearly derived from Major General Kay’s consultations with GCHQ. I duly remonstrated with Major General Kay. He himself seemed unhappy with the bureaucratic reply he had passed on and agreed to have further talks with the ‘department concerned’. Following these discussions, he rang me to say on behalf of the D-Notice Committee that no deletions were necessary before your article went to press. He duly confirmed this in writing by the letter which I have sent you. Evidently, he had succeeded in wringing from the ‘department concerned’ an admission that there is of course nothing in your article which in any way threatens national security. I imagine that Sir Peter Marychurch composed his letter to you in the light of Major General Kay’s first response to your article before it had been cleared by the D-Notice Committee. Anyway, though Sir Peter may not like it, the fact is that I have cleared your article with the appropriate authority.
Andrew did not see how there could be any American objection to a paper cleared by the D-Notice Committee in the UK as it was exclusively restricted to British matters. He asked Welchman’s permission to publish ‘the whole sordid affair’ of his treatment in a future issue of the journal. On 22 August, Welchman wrote what appears to be his last letter on the matter. He gave Andrew permission to use the material about his case in any way he saw fit. He felt that it would be nice if the shabby way in which he had been treated could be publicized before he died.
Epilogue
Like that of many of his former BP colleagues, Welchman’s later life was defined by his wartime work. BP proved to be a unique experience and one that he would never be able to replicate. The Canadian Pat Bayly and the American William Bundy felt the same, despite distinguished careers after the war. In a letter to Welchman in 1983, Bayly observed that ‘In retrospect, the years I spent in your company were the happiest and most productive, years of my life.’
Inevitably, Welchman felt compelled to share some of that experience with the general public through The Hut Six Story. He felt that by explaining some of the technical detail that lay behind BP’s success, readers would have a better understanding of the magnitude of its achievement. He also believed that the lessons learned at BP could help the United States and Britain develop a more secure tactical communication system. When Welchman began to write his book the Cold War still had seventeen years to run.
Many of the early books about the BP’s wartime activities contained inaccuracies or missed crucial details from the story. This was even the case when Harry Hinsley published the first volume of his British Intelligence in the Second World War in 1979. Given that access to official technical documents was restricted to Hinsley’s team, there was much anticipation that a technical appendix would shed light on how the Enigma messages were actually broken. Alas, neither the long awaited appendix nor the body of the book gave any technical detail of either Polish or British methods.
Welchman vented his frustrations in a letter dated 10 July 1984 to Ralph Erskine, a leading historian of wartime codebreaking. The following are extracts from that letter:
Hinsley could have avoided his gross errors by talking to people who were involved in the birth of Ultra. Travis, Denniston, Knox, Turing, Jeffreys and Alexander were dead. But several key members of the early Hut 6 were still alive. For example, Babbage, Colman, Milner-Barry and myself. Perhaps Hinsley was not allowed to consult us.
Hinsley suggests to the uninformed reader that the breaking of Enigma keys was entirely due to the Bombes. This, as I tried to point out in my book, is quite untrue. Our success was the result of German errors and of excellent communication, collaboration and cooperation among the many specialized activities that were involved.
Indeed Hinsley ignores the importance of the high quality people who were recruited in the early days for the many different tasks that would be involved. If Travi
s and I had not been able to start building up the many-faceted Hut 6 organization and staff before the end of 1939, Ultra intelligence might never have come to bloom. Fortunately, we had built up an embryonic team of new-style cryptanalysts, with supporting activities operating round the clock, before the invasion of Denmark and Norway in April 1940.
Without the early demonstration of codebreaking success in Norway, based on the Polish manual methodology, we would have not won support for the major expansions that proved so necessary. If Turing and I had not come up so quickly with our ideas for a Bombe, the early success would have fizzled out. If I had not come up with an organizational plan before the war was three months old, the early success itself might not have been achieved. These startling facts are not brought out in Hinsley’s ‘Official History’.
You may well wonder, as I do, why on earth so much nonsense has been talked about the early days of Hut 6. The answer may be that very few people in GCHQ knew what was involved, and that the people who have written and talked, like Hinsley and Good, arrived late in the game and never worked in Hut 6. Even Calvocoressi, who headed the Air Force Section in Hut 3, arrived pretty late. Bundy, who did work in Hut 6, arrived even later. Our continuing success depended on being able to recruit large numbers of top quality men and women, but when they arrived they were rightly concerned with what they had to do; not with how things started.
*
At the end of the Second World War, a number of BP veterans returned to academic life and successful careers while others remained within GCHQ. The first five post-war directors of that organization, Edward Travis, Eric Jones, Clive Loehmis, Leonard (Joe) Hooper and Arthur (Bill) Bonsall had worked at BP during the war. While it is not appropriate to equate the experience of Welchman and others at BP to that of fighting men and women returning home after the war, the void in their lives must have been similar.
Welchman never seriously contemplated a return to Cambridge, having become disillusioned with academic life there. He had completed research which he believed would have earned him a PhD but had never got around to submitting it formally before putting his life on hold and reporting to BP on 4 September 1939. He thought that lack of a doctorate would have been a barrier to success at Cambridge if he had returned. While he did work at MIT after the war, he tried to discourage his daughter Ros years later from becoming an academic. She did not take his advice and in the end he was proud of her academic achievements.
In its early days, BP was like a research laboratory which eventually put in place an infrastructure capable of turning innovation in cryptanalysis and signals intelligence into real products and systems. Its research activities as described by its wartime participants, particularly in the early days, share many of the elements of what today’s IT world calls skunkworks projects.1 These are typically developed by a small and loosely structured group of people who research and develop a project using processes which encourage radical innovation. Skunkworks often operate with a high degree of autonomy and, unhampered by bureaucracy, are tasked with working on advanced or secret tasks. These projects are often undertaken with a tacit understanding that if the development is successful then the product will be designed later according to more structured processes. BP might well have been the world’s first such endeavour and more recent examples include CERN (the European Organization for Nuclear Research)2 and PARC (Palo Alto Research Center Incorporated).3
Of the technologies that emerged from BP, the bombes and Colossus best fit the approach described above: that is research and innovation leading to a real game-changing product which can be produced on scale. The bombes wedded Turing’s and Welchman’s innovative thinking to existing technologies already being produced by BTM in Letchworth. In January 1944, Tommy Flowers delivered to BP the world’s first electronic computer. Like the bombe, this development was based on the innovative research of people at BP such as Bill Tutte. However, its realization at the Post Office’s Dollis Hill research laboratory also required research and innovation since many of the components of the end product needed to be invented. Brian Randell, the man who subsequently told the world, in some detail, how Colossus actually worked, claimed that Flowers’s achievements were a major and highly original step towards post-war electronic computers in Britain.
Welchman would have initially agreed with Randell but, with his extensive experience of American post-war computing developments, he changed his mind. His view was that while Flowers did a magnificent job for a specific problem, developments in America had a much greater impact on post-war electronic computers, in Britain as well as in America. He believed that the BP experience was more of special-purpose equipment than computer experience. He also felt that the development of the modern computer started after the war by taking advantage of technology developed during it, in places such as the MIT Radiation Lab.
The success of BP was in many ways so astonishing that many German officials who survived the war refused to believe that it had actually happened. In 1983, Welchman’s nephew invited him to meet a work colleague in the American aerospace industry. His colleague was a German by the name of Karl Kober. A brilliant man with two doctoral degrees, Kober had been brought over by the young American space programme from a French prison after the war and turned into a space engineer. He had been a leading figure in Germany and here he was, now a respected NASA scientist, sitting across a kitchen table from one of the people who had made a significant contribution to Hitler’s downfall. The discussion was lively and quite remarkable. Kober completely denied even the possibility that Britain had ever broken the Enigma machine. According to Welchman’s nephew, Kober got quite heated and was adamant and not to be convinced otherwise. Welchman appeared to be quite calm and almost bemused to see his former enemy unwilling to accept the evidence before him.
Towards the end of 1983, while still recovering from a hip replacement operation, Welchman read Andrew Hodges’s biography of Alan Turing. He was moved to write a paper, titled ‘Ultra Revisited, A Tale of Two Contributors’, which contrasted his life to that of Turing. After the war, it was rumoured that both he and Turing were being considered for knighthoods. They were both then led to believe that they could not receive such an award for work carried out as Temporary Civil Servants at a salary of £600 per annum. He summarized his frustration as follows:
Then, not long ago, the Hodges book on Turing came my way and helped to crystallize my thinking. Before doing anything else, I decided to write about the comparison between his life and mine. The contributions to science that he might have made, if he had not been hounded to death, would have been far deeper than anything I could produce. What a waste! Yet, I too, in a more pedestrian way and over a longer period of time, have been able to contribute visionary ideas, many of which could still be valuable. But I find myself with frustrating problems.
I have tried over the years to take advantage of the expertise of others. But changes have been rapid and are likely to continue that way. Much of my acquired knowledge and perception of the future must already be out of date. I still believe that I have something of value to contribute to our military preparedness, but my contribution can bear fruit only if I can pass on my accumulated experience and ideas to an interdisciplinary team of younger men who will keep up to date.
Here again, I run into frustration. What I want to contribute is twofold. First, my evaluation of what is seriously wrong with our military R&D, particularly, but not exclusively, in the area of communications, intelligence and security. Second, my ideas on what could be done to improve our military preparedness. These are highly sensitive matters. We do not want to broadcast our weaknesses to potential enemies. Consequently, the contribution that I am uniquely qualified to make can be made only in an environment in which secrecy can be effectively maintained. Yet, just as Turing was prevented from making further contributions to cryptology because he was regarded as a ‘security risk’, so am I prevented from contributing to our future national security for the sam
e reason. This, also, seems to be a waste.
Ronald Lewin had written to various officials in the intelligence community in 1982 about their shabby treatment of Welchman. Other authors such as Nigel West and David Hooper would later refer to the matter in books published in 1986 and 1987 respectively. Former colleagues, however, kept quiet after Welchman’s book was published and, according to Diana Lucy, he felt a bit let down by them. However, he understood the pressure on anyone who had signed the Official Secrets Act to remain silent. West would go so far as to add the subtitle ‘Including the Persecution of Gordon Welchman’ to his 1986 book The SIGINT Secrets: The Signals Intelligence War, 1900 to Today. In the book, West gave his view of why GCHQ and the NSA had reacted in the way they did to the publication of The Hut Six Story:
Until his intervention, GCHQ and the NSA had perpetuated the myth that code breaking had won the war and the Enigma machine had been solved. Very few knew this to be a distortion of what had really happened. In consequence public attention had focussed on cryptography and been diverted away from the more sensitive matter of traffic analysis.
He would go on to conclude that the impact Welchman’s book had on the intelligence community in the USA and Britain was to compromise the carefully nurtured cover story of the Enigma’s vulnerability. Also included in an appendix was some of the correspondence between Welchman and the NSA.
Gordon Welchman Page 26