Most Secret War

by R. V. Jones

  Actually, I had only spoken the truth: it certainly was the best piece of technical observation on the part of a ground agent I had seen, even taking into account the magnificent work of the Belgians—the difference was that the observer was obviously a highly qualified physicist or engineer who could both draw diagrams and write out mathematics accurately. Dansey had merely sent for me because he knew that I had connections with the Air Staff, and might therefore be able to find out where the letter had originated. I drew a deep breath and said ‘I wrote that letter—but before you explode let me tell you why I wrote it!’ I then told him the story, and when I had finished he walked round and round his room saying ‘THE CHEATS, THE CHEATS, THE CHEATS!’ And then ‘As a matter of fact, he was our agent anyway!’

  Again, I thought that the matter was at an end: but not so. Some months later Hesketh-Pineapple telephoned me to say that he had thought it would be nice if I could meet the man who had written such good reports, so S.O.E. had pulled him out of France and flown him to London. In fact, they could not have done a more embarrassing thing. In the first place, it was much more useful to us to have a good observer on the other side of the Channel than kicking his heels in London. We might well have to spend time looking after him, if only as a matter of courtesy, when we were already very short-handed.

  S.O.E. explained that they intended to send him back within a few days, and I then had the pleasure of meeting him. He turned out to be Yves Rocard, Professor of Physics at the Sorbonne, a fine physicist and a very gallant man. S.O.E. in fact failed to get him back within the short time that they had hoped, with the result that it would then have been too dangerous for him to return, so he spent the rest of the war up to D-Day in England. Fortunately he was not the kind of man to remain inactive, and he joined the Free French Navy and became their Director of Scientific Research. The Gestapo were after him in Paris, but in letters to his wife he led them to think that he had fled to Corsica and they ultimately lost interest. He was imperturbable. When, in February 1944, the Luftwaffe staged some minor raids on London, he had a room on the fourth or fifth floor of a house in Queen’s Gate. This was struck by a bomb and most of the house destroyed. He was in bed at the time and, fortunately, the bed remained perched, as it were, on a shelf formed by the surviving fragment of the floor. He felt all round his bed in the dark and found that most of the floor was missing but, deciding that he was at least safe where he was, went to sleep again, and awoke to find a crowd in Queen’s Gate pointing to the spectacle he presented. What is more, although he was frequently in my office, one of the only two non-British nationals who so gained our confidence, he never mentioned to us that he had been bombed. We heard it after the war from his son Michel who, to his father’s embarrassment, turned out to be ultra Left-Wing, and figured prominently in the French student riots of 1968. He even put himself up as a Presidential candidate against General de Gaulle.

  Yves Rocard was as staunch a friend as Britain could ever find. After the war the German cryptographers had fled from their headquarters at Treunbritzen and had ended up in the relatively small sector of Germany that was allocated for French control. Our own cryptographers at Bletchley were very anxious to talk to their German counterparts, particularly to investigate how far the latter had been successful in breaking our codes. A formal diplomatic approach to the French had failed completely. I therefore offered to see what I could do, and told Rocard of our problem. Within twenty-four hours he was back with full permission from the French for our men to enter their zone and interrogate the Germans—and the only request the French made was that if we found anything that affected the security of France, would we please let them know. We were all delighted when Rocard received the C.B.E. for his services to us.

  He told us that on his return to Paris he noticed that various of his friends treated him curiously, sometimes even to the extent of seeming to wish to stay on the other side of any convenient table. It turned out that Madame Rocard was faced with the problem of explaining his disappearance when S.O.E. failed to get him back, and she managed to convince everybody that she had not been unduly surprised because he had shown signs of going off his head, and that she could only assume he was now completely mad. She succeeded in convincing not only the Germans but his close friends.

  CHAPTER THIRTY-ONE

  The Kammhuber Line

  GENERAL JOSEF Kammhuber had been appointed by Goering on 17th July 1940 to take command of German nightfighters. His first concept was a great belt of searchlights deployed to the west of Germany proper, through which our bombers would have to fly to reach their targets. Warned by the Freyas on the coast, the German nightfighters would orbit suitably placed beacons in front of the searchlight belt ready to swoop on any bomber caught in the lights. The setting-up in 1941 of what came to be called the ‘Main Belt’ was of course soon reported by our bombers, and the question for us was: how did it work?

  Our own experience strongly suggested that searchlights were unnecessary for nightfighting if a proper radar technique was available, and the apparent German dependence on lights was therefore something of a puzzle—all the more so because the 40 and 60 kilometre night-fighting circles we had already found suggested that searchlights need not be involved. Our bombers crews’ reports repeatedly suggested that the accuracy of initial aiming of the lights was due to their being radar-controlled, along the lines of our own ‘S.L.C.’ (Search Light Control) radar, in which a special radar set was directly attached to an individual light, but we could find no true German equivalent. On the contrary, we had encountered various mentions of searchlights in Enigma, but none associating them with a radar control. Two direct lines of attack could be made on the puzzle, for if the lights were radar controlled the obvious equipment would be the Würzburg or something derived from it. We could therefore fly special aircraft carrying observers with suitable receivers for the 53 centimetre transmissions of the Würzburgs over the Main Belt; we could also attempt to photograph the searchlights and see whether any had a radar equipment associated with it.

  The first line of attack was already available, in the R.A.F. unit which had made the original flight to discover the Knickebein beams in 1940, for later in the Battle of the Beams the same personnel had taken to flying down the beams in order to bomb the transmitters, and thus to making radio and radar observations over France. Specialist observers from the Telecommunications Research Establishment had volunteered for the flights, and had received R.A.F. Commissions in case they were lost on operations.

  This indeed happened to one of the first observers, Howard Cundall. Cundall had been flying over France in the early hours of 4th November 1941, when a stray A-A shell fragment damaged an engine, and the crew had to bail out. The loss of a crew would have been a serious enough matter, anyway, but Cundall knew a good deal about our own radar, including the new centimetric devices which were coming along, and might thus be a valuable prisoner for the Germans to capture. The crew all landed without undue damage, and they remained at large for some days. After several adventures Cundall reached the coast near Mont St. Michel on 18th November, and seized a rowing boat in which he tried to hoist a make-shift sail and get back to England. He was seen by the Germans and captured. All the other members of the crew were caught except for the second pilot, who succeeded in escaping to Spain. This was fortunate because the Germans thought that they had captured a full crew, and therefore did not suspect the nature of Cundall’s duties. While in Stalag Luft 3 he not only concealed his knowledge of our new radar devices, but built a radio transmitter with which he opened contact between the camp and London, maintaining it even during the long march eastwards as the Germans pulled back the prisoners in the late stages of the war. In this way he provided information from captured air crew regarding their experiences with the German night defences, and thus aided our offensive even from behind the barbed wire.

  An even earlier observer on radar listening flights was Eric Ackermann, who in two years made more than 90
flights, including more than 40 on which bombs were dropped against the beam stations. Most of his later flights were over the Main Belt, listening for Würzburg transmissions. Not only did he hear these transmissions very frequently over the Main Belt, but he also concluded that some of the lights were radar-controlled.

  Besides the radar transmissions themselves, we had the possibility of listening to the instructions transmitted by radio telephony to the German nightfighters from their ground controls. In these instructions there was frequent reference to something called ‘Kleine Schraube’ (Little Screw), which we ultimately realized was nothing more than a code-name for the radio beacon which the nightfighter was to orbit.

  The other thrust of our attack, the location and photography of searchlights and radar stations, drew first blood in March 1942 when we photographed the Giant Würzburgs near St. Trond (p. 227) following a report from the Belgian Resistance. The photographs partly solved our puzzle, because the searchlights around one of the Giants were presumably controlled by it: these could be ‘master lights’ which showed other searchlights where the target was. We had already encouraged the Resistance workers to report as many searchlights positions as possible, with the intention of asking P.R.U. to photograph the positions reported.

  The Belgians had become tremendously active, producing an enormous amount of information of all kinds which they took across France to Lisbon. One of the main routes was the express that ran from, I think, Lille to Lyons, where the fireman acted as courier. He hid the reports under the coal, so that all he had to do if the train was searched was to shovel the whole lot into the fire box. It seemed more difficult to get the information back to Britain: Jempson told me that at one stage there were 15 hundredweights of reports waiting in Lisbon for a plane back to England.

  Many years afterwards I met some of our Belgian helpers at a Human Rights Conference in Brussels in 1970, and I was remarking to them that it seemed to me that half Belgium had been working for us, even to small children sitting by the wayside carefully noting the numbers of every German vehicle that passed. In fact espionage seemed so compulsive to the Belgians that I could not help wondering whether if we had occupied them instead of the Germans, they would have been just as professional in helping the Germans as they had been for us. One of their leaders said ‘You are quite right. You must remember that Belgium has had a long history of being occupied and of finding ways of living with and getting the better of whoever was occupying us’ and he went on to say that the tradition was so strong that it seemed almost automatic for them to get organized once the Germans had taken control.

  But just as the information began to come back from Belgium the searchlights disappeared, and our reconnaissance photographs showed nothing but empty emplacements. What had happened was that Hitler had personally ordered all the searchlights back to Germany, where they could make a show of defending the towns. Bomber Command was now beginning to hit Germany despite Goering’s assurance that no bomber would ever appear over Berlin, and Hitler wished to show the German people that something was being done to defend them. So on 5th May 1942 the searchlights were withdrawn back to Germany, reversing the move by which Kammhuber had originally brought them forward into the Main Belt. By 20th May, when P.R.U. went out to reconnoitre, they had all gone.

  A few weeks after this disappointment I went into Jempson’s room to see whether anything more of interest had come in. In his own way he was a professional who reckoned that his task had ceased once he had got the information, and that it was not his job to worry about whether any use was made of it. He waved towards his top shelf, and told me that there might be a map ‘up there’, indicating several dishevelled piles of documents. I asked whether he would mind if I looked through them, and after a few minutes I found the map: it was a breath-catching moment for, as I unfolded it, I saw that it must show the deployment of a whole searchlight regiment covering the entire southern half of Belgium. I like to think that one of our Belgian friends, daunted by the prospect of cycling round the countryside laboriously plotting the sites of individual searchlights, thought that it would be simpler to break into the headquarters of the Regimental Commander and remove his map, for this is what it was. I warmly thanked Jempson and took it back to show Charles Frank. We could now see all the searchlight positions on 90 kilometres of the Main Belt, together with the various levels of headquarters. Everything seemed to be organized on the basis of a military trinitarianism. There were three lights to a Zug (squad), three Zugs one behind the other to a Batterie, three Batteries side by side to an Abteilung (battalion) and three Abteilungen to the Regiment. Each Abteilung covered a front of 30 kilometres, each Batterie taking 10 kilometres with the three Zugs 5 kilometres one behind the other and about 3 kilometres laterally between individual searchlights. In front of each Abteilung was a more elaborate sign, and fortunately one of them coincided with the Giant Würzburg station that we knew of near St. Trond. It therefore followed that the other signs at 30 kilometres on either side of it, one each in front of the flanking Abteilungen were similar Giant Würzburg stations. One section of the map showing a complete Abteilung is reproduced at Plate 16.

  The intended mode of operation was now clear. The Freya associated with the Würzburgs would first pick up an incoming bomber and direct one of the Würzburgs to it. The other Würzburg would be watching the German nightfighter which would have been ‘scrambled’ and instructed to orbit a radio and visual beacon (the ‘Little Screw’) some kilometres in front. As the bomber approached, the ground controller would begin to direct the fighter on to a course of interception, while the Würzburg following the bomber would direct his surrounding three searchlights towards the bomber. If these succeeded in picking the bomber up, it would now be visible to the battalion of searchlights behind, which would now light up, and try to hold the bomber in view as it crossed the belt so that the nightfighter could attack it. This explained why radar control of searchlights had seemed relatively scarce—for the three searchlights actually directed by radar, there were another 27 behind which depended on visual pick-up from the first three.

  I took the map to the Photographic Reconnaissance Unit at Benson to show the pilots, so they would know what it was I now wanted when my official request for sorties came through. Over the next fortnight they photographed many of the positions indicated by the map. It did not matter that the searchlights had been withdrawn, because the radar stations had been left in place, and we knew, as did no doubt the Germans, that the radar stations themselves were sufficient for interception, at least if backed up by an airborne radar system for the night-fighters. The stealing of the map by agent ‘Tégal’ was therefore of very great value.

  Fig. 11. Sketch of the German nightfighter control headquarters at Sautour, Belgium, made by a Belgium agent. The poor quality of the reproduction is due to the original having been transmitted to London on microfilm

  An interesting difference now emerged between the Giant Würzburg stations in the Main Belt and those such as Domburg on the coast. With the latter there was always a T-shaped hut, but when we photographed the Abteilung command posts in the Main Belt we found the T-huts some kilometres back from the Würzburgs among the searchlights. The T-hut was where the ground controller was situated, and it was evidently considered better for him to be among the lights when these were available. The next question was: what apparatus did he have at his disposal? More than one Belgian agent succeeded in getting into a T-hut and describing for us the ‘Seeburg Table’, on which the positions of the bomber and fighter were shown as spots of coloured light, red for the bomber and blue for the fighter, projected from underneath on to the ground glass screen forming the flat surface of the table. The projectors were moved by operators who received telephonic instructions from the two Giants. There was no cathode ray presentation such as we ourselves employed, and the plotting system was therefore ponderous and liable to human error.

  Although I have said that the Belgians were almost automaticall
y inclined to espionage, it would be entirely false to imply that they were without feeling. Later in the summer another report came through on microfilm from agent VNAR 2 of ‘Service Marc’. It ended with the following:

  (1) In view of the apparent current interest in similar installations, it seems to us odd that no attempts have yet been made to destroy them, especially since (a) except for No. 2, they are not powerfully protected, and (b) they were installed with difficulty in 4 months and it would not be at all easy to replace them. The Germans’ interest in them is clearly shown by the extremely strict way in which they guard the approaches, which has several times resulted in our being fired at by sentries, fortunately with more zeal than accuracy. The Jauche installation is particularly easy to spot and to attack from the air.

  (2) As far as our work is concerned, it would be helpful if we knew to what extent you and the British services are interested. We have been working so long in the dark that any reaction from London about our work would be welcome to such obscure workers as ourselves. We hope this will not be resented since, whatever may happen you can rely on our entire devotion and on the sacrifice of our lives.

  Fig. 12. The state of British knowledge of German nightfighter control stations in early September 1942. Six positions in the ‘Kammhuber Line’ were known in Belgium, and a seventh on Walcheren in the Scheldt Estuary. The parallel lines running south from Schleswig Holstein to Belgium were known to delineate the Kammhuber Line; the much larger area from Scandinavia to the Swiss border was designated as a general air defence region