X, Y & Z

by Dermot Turing

  Antoni Palluth and Edward Fokczyński – both veterans of the AVA factory – now joined the group at PC Bruno, together with Henryk Paszkowski, a junior cipher officer. More Poles joined in February and March, devoting their efforts against Russian ciphers. Since the arrival of the Zygalski sheets from London, the Poles concentrating on German messages were having to work very hard. They were finding the Enigma settings more rapidly: in the two months after the move to PC Bruno, the average time to find the settings fell from over eighteen days to five. In some cases the Enigma team could work out the settings just two days after the associated messages were sent. As they caught up with the older traffic, the workload went up. Bletchley Park solved about fifty Enigma settings between mid January and the end of March; for PC Bruno the number was about forty in the same period.12 Most importantly, every month, they were now reading thousands of messages, some of which might reveal still-current facts not yet known to the Allies.

  With many coded messages, when the fun of finding the secret key is over, there can be a sense of anti-climax. The disappointment when the content of an Enigma-enciphered telegram turns out to be nothing more than, say, congratulations on the promotion of some unknown officer to the rank of major, can easily be imagined. Despite an intense effort, the content of messages deciphered during February and March 1940 was of limited strategic worth. Bertrand’s boss, Colonel Louis Rivet, thought the increasing volume of material about the Luftwaffe showed promise, and some feedback from air intelligence in London on the scraps they were sent was mildly encouraging.13 At least the Allies were getting a bit better at handling the traffic. If the phoney war turned into a fighting war and Enigma messages contained tactical information, they would need to be.

  Gustave Bertrand ran PC Bruno according to slightly strange rules. For one thing, the working environment was a bit weird. Équipe Z – Langer’s Z team – were put into an artist’s studio at the very top of the château, where the walls were daubed with attempts at landscape painting by its former occupant. The furnishings, however, were very basic. And then there was segregation: each nationality was required to eat separately from the others. In addition to the Poles, Bertrand was accommodating a group of Spanish refugees, Équipe D, who were tackling the codes of the Franco régime (and its ally the Italian fascists) with success, so that meant there were three dining rooms, for no obvious reason that the code-breakers could see.14

  In addition to the Spanish residents, there was a British liaison officer; we are not told whether he was expected to eat alone or with one of the other groups. He was Captain Kenneth MacFarlan, nicknamed Pinky (apparently on account of his complexion). MacFarlan was not cleared to share in the Enigma secret: as a relatively junior army liaison officer his job was to speak fluent French, the language of liaison, but otherwise he was clueless. Bertrand was keeping Pinky at arm’s length. The French re-nicknamed Pinky the Le Navet Royale [the royal turnip]. Pinky was not amused.15

  Gwido Langer soon worked out what was going on. It was divide and rule:

  Bertrand’s attitude towards the English was very French. He was most displeased with the fact that in 1939 the English were simply handed over the machine and the technique of breaking the codes, whereas he had spent a long time trying to obtain the machine. He then said that if the machine hadn’t cost the English a lot, they must pay a lot for it now. However, when the English submitted a proposal, according to which some of them were to come to France to help, Bertrand fought this threat of ‘British colonization’ with passion.16

  If the British became fully engaged at Bruno, the French would become mere suppliers of intercepts and no more.

  The Poles weren’t exactly locked up, despite Bertrand’s arrangements. Distraction came in various forms. There were excursions to whatever hostelries the locality could offer and plenty of laughs. A luggage label, in immaculate condition (which proves that it was never used), is in the possession of the Zygalski family. The label was supposed to be tagged to the buttonhole of Henryk and it gives directions to the reader of what to do with its drunken and unconscious wearer (including disinfection). At a special Christmas dinner on 24 December 1939, there were novelty cigarettes whose smoke turned into snowflakes; sugar lumps which contained plastic cockroaches which revealed themselves as the code-breakers drank their coffee; biscuits with rubber bands inside so that the bitten-off part sprang out of the eater’s mouth; and, the pièce de résistance, a ‘Verdun bomb’. A veritable tribute to the work of the code-breakers, the Verdun bomb was a round dark chocolate shell. When its fuse was lit, it actually exploded, showering the company with sweets and other goodies.

  Marian Rejewski later confessed to having exposed a terrible secret of PC Bruno.

  Once, in the hollow of a tree growing near the sluice gates to the pond in the park, I found a crank which opened the sluice. I turned the crank and opened the sluice just a bit, and I left. Next morning, all the meadows in the vicinity were flooded, and a big pile of empty wine bottles had appeared at the bottom of the pond. It turned out that Major Ciężki would take wine bottles, emptied in secret night-time drinking sessions with Lieutenant Colonel Langer, under his coat and throw them into the pond.17

  Gustave Bertrand was promoted to commandant (major) on 25 March 1940. This helped with an anomaly. The idea of Langer, a lieutenant colonel, and Ciężki, a major with three years’ seniority already, being subordinated to a mere captain, was strange enough.18 Langer believed his unit ought to have been incorporated into the Polish Army which had been reassembling under the authority of Sikorski.19 For his part, Bertrand was trying not only to prevent that, but also to establish his authority with the Poles and in his dealings with the British.

  Periodically the British would come across to check how the relationship was going. At a meeting in February, led by Denniston’s number two, Commander Edward Travis, there was a longish agenda, covering the progress the British were making with their special machinery, Russian codes, and more.20 The British wanted to know how the Poles were getting on with their theoretical attack on Enigma, but Langer had to report that the team was already working flat out – not going to bed until 2 a.m. – on deciphering, translating, and reporting. There was no time for research. It was a contrast with the other country house at Bletchley, with its separate teams for theory, decipherment and translation, and intelligence.

  The park at Bletchley was becoming less and less like a stately home with gardens. Its residents knew that with express trains only 100 yards away, belching filthy coal smoke and whistling as they passed the Bletchley junction, comparison with the country house idyll of the Château de Vignolles was laughable. Even before the professors of September 1939 arrived, carpenters had moved in and sturdy wooden outbuildings were being knocked and hammered into place all over the lawns. Soon the maze would be pulled up by its roots and the rose garden would go. Eventually the organisation would outgrow the mansion, its stable-yard cottages and the wooden huts. Bigger, more permanent buildings would be needed. The builders kept the whole site in a state of activity, increasing the sense of urgency.

  The Bombe designed by Alan Turing had taken several months to develop and even then nobody knew whether it would work. On 18 March 1940, a prototype arrived in the back of a lorry from Letchworth. It was heavy and ugly. On its bronze-coloured front face, there were rows of what looked like little dartboards picked out in brass. The dartboards were set out in columns of three, dozens of them. Each dartboard had at its bullseye a spindle, on which a small, squat cylinder would be set. These cylindrical drums would behave like Enigma rotors, each trundling round the 17,576 possible positions of the three-rotor Enigma machine. Round the back, red plaited cables connected up the sets of three drums to test a crib. It was Britain’s most secret technical development. It would solve the Enigma problem once and for all. But there was just one glitch: it didn’t work. Or rather, it worked a bit too well. The idea of Alan Turing’s Bombe was that it would stop – literally, the drums wou
ld stop revolving – when it had found a possible setting which could cause the observed encipherment. The issue was that the machine stopped too often, finding too many plausible solutions. The method only worked efficiently if you could connect up the cables to make not just one circuit around the sets of drums, but as many circuits as possible. Having multiple electrical circuits meant that far fewer false positives would be thrown out for further testing. Until the Bletchley Park technicians could find a solution to this, the machine wasn’t going to be of much help, and it wasn’t going to win a war. To decipher Enigma, they would need to go on using Zygalski’s sheets.

  • • •

  Gustave Bertrand understood that the solution to Enigma depended on the ability of his joint Franco-Polish team to have access to the technical equipment they had relied on before the war. The search was on for a factory which could reproduce in secret, in France, what the AVA team had achieved (in spite of Colonel Sadowski’s suspicions), in Poland. Gwido Langer had saved one of his team’s duplicate Enigma machines from the graveyard near Włodzimierz; with the machine previously donated to Bertrand in July, that made two to work with. Tucked away in PC Bruno, the technicians of Équipe Z were dismantling Langer’s machine, to enable the French to make more copies.

  The Polish Enigma machine was not a straightforward copy of a German military one. In 1939, nobody in the Allied intelligence services had yet seen a real version of the Wehrmacht Enigma, except in the photos stolen by Schmidt and copied in European bathrooms. So the Polish machine differed in appearance: its plugboard was at the back and its keyboard was arranged in alphabetical, rather than standard QWERTZU, order. But these differences were superficial. The machine worked in exactly the same way as the German machine. It was a triumph of reverse-engineering.

  By April 1940, Bertrand had found his engineering company in the shape of the Établissements Édouard Bélin in Paris and the order was placed. For a total price of 240,000 francs (something around £1,350 at then prevailing rates) he ordered twenty-five cipher machines of type ‘A’ and three cyclometer machines of type ‘B’. The intention, in due course, was to increase the number of type ‘A’ machines – the Enigma analogues – to forty. The manufacturers reckoned they would need four months to make the first twenty-five.21

  In light of what the French intelligence team had just learned, four months would prove to be a very long time.

  • • •

  The entry in the diary of Colonel Louis Rivet, the head of French Secret Intelligence, for 18 March 1940 is unexciting. He had lunch with Amédée Bussière, head of the Surêté Nationale, the French equivalent of the FBI. And he went to see General Louis Koeltz, to discuss ‘various matters’. How dull. But one should never treat the diaries of spies as documents that tell the complete story. Henri Navarre was a career army officer who had been assigned to the German section of the Deuxième Bureau in 1938, with Louis Rivet as his immediate boss. On 17 March 1940 Navarre had returned from a meeting in Switzerland; then there had been that discussion of ‘various matters’, and by the next day, 19 March, with Navarre in tow, Rivet insisted that he needed to see General Maurice Gamelin, the commander-in-chief, to impart intelligence of major importance. Navarre’s trip to Switzerland had been an affair of the utmost secrecy. Even the self-important Bertrand was kept in the dark, despite the fact that the meeting had been with Bertrand’s personal contact, none other than Hans-Thilo Schmidt.22

  Notwithstanding the state of war, Schmidt had come to meet Navarre in the bar of the Hotel Eden in Lugano. Hotel Eden had a gorgeous view of the lake, framed by the snow-tipped mountains. The sun was shining and Schmidt needed to talk. He was on his way back from Rome. But that wasn’t the reason for the meeting. Schmidt’s brother, now returned from invading Poland, had been promoted again and was the general in command of the 39th German Armoured Division, which was lined up along the River Moselle and facing Luxembourg and Belgium. The tank general had learned, during a lunch with no less a personage than Adolf Hitler, the tactical outline of something called Fall Gelb. The person setting out the plan was General Erich von Manstein. Where Fall Weiss (Operation WHITE) had been the invasion of Poland, Fall Gelb (Operation YELLOW) was a forthcoming attack on the West. Manstein had explained that the direction of attack by the Panzers would be through the Ardennes, bypassing the French defences of the Maginot line.

  Hans-Thilo Schmidt foresaw that when Operation YELLOW was put in train, it could be the end of his special relationship with the French. His business partner had already been called up and the soap-factory, which had done so well as a front for his ill-gotten funds, could no longer serve. It was going to be wound up. Schmidt was becoming a hunted man. An official investigation into a leak had been initiated and all communications with foreigners were being watched. But this wasn’t what scared Schmidt. His concern was that he knew the French had revealed the existence of a highly placed German intelligence agent to Langer and the Poles. What if the Poles had left something behind which revealed who the source was? Moreover, he had learned that the Sicherheitsdienst, the intelligence agency of the SS and the Nazi Party, had found proof in Warsaw that the Poles had reconstituted the Enigma machine.

  In 1940, Schmidt knew about the Franco-Polish liaison and suspected that the Poles had reconstituted the Wehrmacht Enigma. ‘I hope that M. Barsac [Bertrand] and M. Lemoine [Rex] haven’t given anything to the Poles which could put me in the frame,’ he said to Navarre. Thus, with the war about to enter a new phase, it was with a sense of finality that Hans-Thilo Schmidt asked Navarre to pass his particular regards to Bertrand, and, additionally, to Rex, as he took his leave on 10 March 1940.

  • • •

  The intelligence supplied by Schmidt was beyond any price, completely impossible to value. Other agents had warned of an offensive on the Western Front to begin on 11 April 1940; then another agent produced an Abwehr questionnaire requesting of its own spies minutely detailed information about bridges, railways, roads and watercourses in the Ardennes sector.23 The intelligence was contradictory and did not imply a new concentration of German forces. John Colville, who spent much of World War Two serving at 10 Downing Street, recorded in his diary the views of his superiors: attacks on the Netherlands and Belgium were threatened in November 1939 and again in January 1940, but nothing had come of these.24 No surprise, then, that the French high command was becoming increasingly sceptical about reports of imminent attack by the Germans. Anyhow, an attack through the Ardennes was impracticable owing to the difficulties of the terrain, not to mention the violation of Belgian neutrality that would be involved. ‘The accumulation of evidence that an attack is imminent is formidable … and yet I cannot convince myself that it is coming,’ fussed Prime Minister Neville Chamberlain.25 But coming it was.

  The Germans introduced a new Enigma network on 10 April 1940; this seemed to be focused around Norway and Denmark. The network was labelled ‘Yellow’ by the traffic analysts at Bletchley, who assigned the names of colours to each network for ease of sorting. Each network used its own basic set-up for its machines (ring settings, plugboard, choice of rotors) so each network had to be attacked separately. Bletchley Park broke ‘Yellow’ within five days using Zygalski’s sheets.26 The material was ‘voluminous and highly operational’, if sometimes hard going and difficult to understand owing to the copious use of military jargon, abbreviations and acronyms and the lack of context. With effort, it disclosed information about army, air force and naval movements. Better still, the code-breakers were working much faster. Compared with the months of February and March, the volume of material generated in April was significantly greater. Enigma intelligence was coming of age.

  On 7 April 1940, German ships had already set sail. On the following day, the Polish submarine ORP Orzeł sank a German transport off the coast of Norway. German troops landed in Norway on 9 April and on the same day Denmark was occupied. An Anglo-French task force attempted a counter-invasion the following week but within two weeks they
had been driven off. It was an unmitigated disaster for the Allies. The famous Royal Navy had been unable to prevent a seaborne landing by the Germans, which had cut off the Anglo-French troops. Then, on 9 May 1940, German forces marched into Luxembourg and on 10 May into Belgium and the Netherlands. Prime Minister Chamberlain was ousted the same day, his failed policy of appeasement now compounded by a bungled operation in which British troops had died for no reason. The Battle of France was about to begin and at last there was the opportunity for signals intelligence to play its part.

  Except that everyone knew signals intelligence would have no role to play. The experts knew that radio silence is imposed in major operations. Certainly the major decisions would be communicated over landlines where possible. And, in any case, the French would be defending their country from the Germans by standing firm behind the Maginot Line.

  The Maginot Line was a formidable defence, certainly impenetrable. It stretched all the way north from Switzerland. When it reached the Belgian border, however, it became … Belgium. French territory from the Belgian border was, to cite an acid remark of John Colville, ‘only defended by an unimpressive ditch’, because there was no need to defend France from the Belgians. Behind the ditch was posted the small and subordinate British Expeditionary Force, who had spent the past six months trying to turn the ditch into a defensible proposition. In defending France from Germany, from previous experience one might imagine that a German invasion would probably come via Belgium, especially if something as inconvenient as a Maginot Line were blocking the other way.

  When General Rudolf Schmidt’s panzers fired up their motors their route was, sure enough, the Belgian one. That put paid to one piece of received wisdom. The next piece of received wisdom to crash was the idea of radio silence. The transmitters were buzzing with orders and reports from the front. The volume of traffic grew beyond any expected measure. During the fateful Norway campaign, PC Bruno had handled 768 messages over four weeks; during the first three weeks of the Battle of France, 3,074.27