I was glad to be out of London. My room in Pimlico had been pretty terrifying in the blitz and I had spent many nights in the basement at Dolphin Square. Fortunately I left before the million to one chance bomb came through the narrow ventilator shaft and killed everyone in that basement.
In Harpenden I soon managed to find a terribly suburban furnished house to which I brought my wife and baby. We stayed there till the following summer, when the unit moved to Beaumanor, a manor house at Woodhouse between Loughborough and Leicester. After three months living in Beaumanor, I got an idyllic furnished cottage on the outskirts of the neighbouring village of Woodhouse Eaves owing to the kindness of Judith Whitfield, Elizabeth Roscoe and Iva Dundas, who had found it for themselves but gave it up for the sake of my baby and my pregnant wife. Woodhouse Eaves turned out to be the country seat of a branch of my wife’s family, the Heygates. My daughter was born there and my family stayed there happily till the spring of 1943 when they moved temporarily back to my parents’ home in Cheltenham. I myself moved to Bletchley with the unit in the summer of 1942, but managed to visit my family fairly often both at Woodhouse Eaves and in Cheltenham.
The story of the work at Harpenden and Beaumanor is essentially one of expansion, increasing knowledge of German armed forces communications and improving methods of analysis and quality of work. The organisation of the unit was and continued from now on to be in two main parts - the log readers and the relatively small fusion room in which the log readers’ summaries were put together with information from decodes. We became experienced in identifying ‘stars’ and their stations from a combination of indications - frequency, number of stations, type of traffic or chat, working routine - quite apart from the content of decodes. Direction finding too, though not of great precision on high frequencies at long range, could be useful in deciding between alternative identifications or in indicating large movements. The practical results of our work, however, were limited by our physical separation from Bletchley Park, our lack of knowledge of its needs and its lack of knowledge of what we had to offer.
The first thing that Crankshaw did when we got to Harpenden was to call the log readers together and on his own initiative let them into the secret that Enigma, the German machine cipher, was being broken. Tozer was angry with Crankshaw about this, but said pragmatically, ‘Well, it’s done!’ Crankshaw told me that he felt he had to do this. He just could not work with people who did not know in broad outline what he was doing. There was however still much ambivalence about the security position. Even those who had been told knew little or nothing about the nature of the Enigma cipher, the methods used for breaking it, the extent of present success or the contribution which their own work might make to such success. Moreover new recruits to the unit were not told anything. This was as well as not every face fitted and some moved on. For instance, Aubrey Jones and a friend of his were posted to Harpenden, took an instant dislike to the unit, refused to talk to anyone and walked about the garden at the Warren like a pair of Italian carabinieri. They moved on within a week.
On our move to Beaumanor, Crankshaw left us, first, I think, for Bletchley and then to join the British Military Mission to Russia. Hamish Blair Cunynghame, who was already in charge of the log reading operation, took over general charge of the unit’s work, though a regular officer, Colonel Thompson, was for a time put in over his head. This led to some conflict. Thompson wanted to make the unit more military and keep it separate from Bletchley. But Hamish got his way. He visited Bletchley and talked with Gordon Welchman and Stuart Milner-Barry, the top men of Hut 6, which was responsible for breaking German army and air force Enigma messages. Welchman and Milner-Barry came down to Beaumanor, looked around and arranged for some of their cryptographers to spend some weeks with us as log readers. These scouts reported back that they had learned much which was useful to them and which explained things that had puzzled them; and they praised the quality and value of the work at Beaumanor, which few people at Bletchley appreciated. Moreover they felt that the security situation was a bit vague and anomalous and that it would be much better from this point of view too if the unit was moved to work within the Bletchley Park perimeter and was properly inducted into the Ultra secret. Welchman and Milner-Barry concurred; Welchman in particular had an instinct that our move to Bletchley would prove important.
In the summer of 1942 we accordingly moved to Bletchley and absorbed Lithgow’s party, the small radio intelligence party already at Bletchley and engaged in building up the communications picture from decodes. Lithgow himself moved to London to play his part in the radio intelligence operation against the Japanese.
I do not want to give the impression that our work at Harpenden and Beaumanor was valuable only as a preparation for our work at Bletchley. We did build up the communications picture; we did supply some leads to the cryptographers; and we did help to provide indications of major new strategic intentions such as the German abandonment of the invasion of Britain, their offensives against Russia and the Balkans, and the start of Rommel’s North African campaign, and in two instances at least we made contributions of great importance.
The first was in regard to the German signals regiment associated with navigational beams for bombers and experimental work on advanced weapons, such as flying bombs, rockets and atomic bombs. The radio operators on this ‘star’ were expert technicians with a very high speed of Morse operation and a contempt for ordinary security. They passed to each other Enigma messages in their own series of Enigma settings and an extraordinary jumble of Q code, their own improvised codes and abbreviations and even en clair. Hut 6 had broken some days of their Enigma messages and the decodes were being studied by Professor Norman, the bouncy little professor with the jaunty walk and twirling walking stick. He found the decodes by themselves rather puzzling, but when he put them together with our log readers’ summaries of the total communication on the star, he began to grasp what was going on, to work out the significance of all the chat and thus substantially to supplement and interpret the decodes. Sometimes too he found leads, e.g. the revelation in chat of part of the cipher setting, which led to further Enigma breaks. All this was a substantial log reader contribution in an important area and started when we were at Harpenden.
The other vital contribution was made when we were at Beaumanor and may well have affected the whole course of the war.
Enigma messages were sent in five letter groups. The first five letter group was from mid-1940 until 1944 a dummy group, the last three letters of which served as an indication of the setting which was being used. There were different settings every day and on each day for different groups and purposes. Each setting could be recognised by the use of any one of four different combinations of three letters allocated to it. These three letter labels were known to us as discriminants.
The intercept station operator included the dummy first group in the message preambles, which he put on the log. So the log readers in their summaries could distinguish different Enigma settings. For working purposes we referred to the different keys, i.e. the different series of daily settings for different groups and purposes, by codenames, largely the names of colours.
In 1940 the main keys were Red (the main air force key, the Luftwaffe-maschinen-schlussel; Violet (the air force administrative key, the luftgaumaschinenschlussel) and Brown (the key of the experimental signals regiment). There was also a practice key, which we called ‘Blue’.
Just before Rommel started his attack in Africa, new army keys began to appear, but it was some time before Hut 6 managed to break them and in the meantime Rommel had pushed forward and was rolling up the British army. In addition to the army keys there appeared an army-air force liaison key (fliegerverbindungschlussel) which we called Scorpion and separate air force and air force administrative keys for the area, which we called Light Blue and Primrose.
At Beaumanor we decided to put one of our junior officers, a certain Rodney Bax (later a judge), to do research on the pattern of discriminants. He sat in an old fashioned bathroom on the wide lid of the ancient private seat while his assistants including Staff Sergeant Sugar (later publicity officer of the Co-operative Union) worked on a board put over the bath. After some weeks of fruitless research, he came upon the interesting fact that the current Scorpion was repeating the Primrose discriminants of a couple of months ago in a scrambled order of days. Blair Cunynghame told Hut 6, who promptly checked. Yes, Scorpion was repeating the old Primrose settings, which had all been broken. The settings were cabled to Cairo, which could now intercept and read Scorpion at once without the delays caused by the need to break and the overburdened top secret communication links with England. It could not have been more opportune. It was just before Rommel’s final push towards Cairo and the main contents of Scorpion consisted of orders to bomber unit (sent to army units for information) and position reports on the German front line (so that the air force should not bomb its own troops). I was told that when the Stukas (fourteen, I think) were sent in to soften up the British defences before the final ground attack, the Hurricanes were waiting for them and shot them all down.