• • •
Like almost everyone else, the British had not really been paying attention. Britain had an Empire to protect and the Empire was protected by a navy and the navy was not threatened by Germany. Apart from a has-been backbencher called Winston Churchill, whose Cato-like croakings about Germany were thoroughly tedious, nobody in Britain wanted to get engaged in anything to do with defence in Europe.
So the British were not ready for a war, imminent or otherwise. Nor were they ready for a new alliance, particularly with the French. The Prime Minister, Neville Chamberlain, said that the French, ‘never can keep a secret for more than half an hour’ and thought that staff talks would recreate the insanity of 1914.14 The one reliable thing about 1914 had been the network of alliances, which with the oiled efficiency of Edwardian engineering had swung into action when the starting lever was pulled. By contrast, in 1938 the old alliances were cold and the non-aggression pacts shifting and unreliable. French objectives were as clear to the British as a glass of pastis and as likely to have unwanted consequences. They wanted to pull the British into another continental entanglement and the way they would do this was via intelligence talks, which would become joint staff talks and then joint operations. And as to the Poles, since the fiasco over the Curzon Line during the Russo-Polish War, the British had been so unenthusiastic about relations with Poland that they could almost be considered hostile. When Sir Austen Chamberlain, half-brother of prime minister Neville, had been foreign secretary in the 1920s, he had said, ‘for the Polish corridor, no British government ever will or ever can risk the bones of a British grenadier.’15 British involvement in military operations to thwart Germany, it would seem, was wholly out of the question.
At the Government Code & Cypher School in London, though, it was more obvious that Germany was a threat, but the German Armed Forces Enigma was a closed book. Perhaps, given the new atmosphere of cooperation with the French, there might be a possibility of getting something that would help with Enigma? On 9 September 1938, Captain John Tiltman of GC&CS prepared a shopping list:
Can the French be asked to give us all the information they have with regard to the following points?
Full details of all devices additional to or differing from the market model.
What is the Army practice with regard to the following:-
Fitting new drums.
Changing the order of drums.
Moving the tyres (bands) on the drums …16
Wilfred Dunderdale (cover-name Dolinoff) duly put the question to Colonel Rivet. The French had no hesitation in establishing cryptology cooperation with the British. It was an essential part of the plan, in fact, to develop a relationship that became so close that the British could not escape from it. The French opened the tap and a flood of information immediately began to flow across the Channel.
25 – Renseignement a/s des moyens de chiffrement utilisés par la Wehrmacht [Intelligence report on means of encipherment used by the Wehrmacht]
29 – Gebrauchsanleitung für die Chiffriermaschine ENIGMA [Guide to use of the Enigma cipher machine]
30 – Schlüsselanleitung für die Chiffriermaschine ENIGMA [Guide to settings of the Enigma cipher machine]
44 – a/s de la machine à chiffrer ENIGMA (Marine-Modell) [Concerning the Enigma cipher machine (Navy model)]
79 – Renseignement sur l’organisation de la Section du Chiffre du R.K.M. [Intelligence report on the organisation of the Cipher Section of the Reich Navy]
80 – Renseignement sur l’organisation du Service d’écoute de la Marine allemande [Intelligence report on the organisation of the Monitoring Service of the German Navy]17
Along with reports under these titles, came many, many more. Gustave Bertand later wrote that 925 intelligence reports were sent to MI6 and GC&CS between September 1938 and August 1939. For each delivery, he obtained a receipt, signed by Dunderdale in the name of ‘Dolinoff’.18 The British code-breakers were enthusiastic. The materials were highly praised: ‘of very great interest’; ‘of very considerable value to us’; ‘with these pinches [enemy material acquired by whatever means] came pictures and diagrams of the Steckered enigma, from which we were able to establish [what] previously had been only surmise’; ‘the value … was enormously enhanced when the French communicated to G.C. & C.S. in 1938 a series of documents on the German W/T networks’. The British, however, could not entirely stop themselves from being sniffy, also writing that ‘French cryptography does not appear to be of the highest class’.19
The British code-breakers called these reports from France ‘Scarlet Pimpernels’. The Pimpernels were typed using a purple ribbon and each of them was numbered, in red ink, in the top right-hand corner, in a characteristic French font. They were stamped TRÈS SECRET, also in red ink. No wonder the British thought that these billets doux evoked the secret communications with the French underground – sealed with a red monogram in the shape of a flower – as seen in the 1934 film of Baroness Orczy’s novel.
Tiltman’s ‘shopping list’ memo had been written days before Munich and only a week before the Germans made their changes to the Enigma operating procedures. If the British had been nowhere on German military Enigma before, they were even further in the dark now, despite the help they were getting from the Pimpernels. A face-to-face meeting with the French might provide even more information and it was also time for the British to trade something back in return for the Pimpernels. So on 21 October 1938, Alastair Denniston, the head of the Government Code & Cypher School, invited Bertrand to a meeting to take place in London in early November. The agenda: a British officer would give his theory of how the Germans communicated the set-up of their Enigma machines, along with a method of finding that set-up, as used before the 15 September change of procedure; what, technically, the September changes actually were; and the method of solving the commercial Enigma.20 The British also agreed to supply copies of the intercepted German signals which they had picked up. They also had Tiltman’s list of outstanding questions.
Not surprisingly, the French were unable to answer any of the questions which Knox – for it was he who was now the British officer dealing with Engima – had posed. But Bertrand knew who could. On 5 December 1938, Bertrand wrote to Denniston, saying that ‘he had found someone who believed he could get access to an army Enigma machine’ and suggesting that Denniston send across a questionnaire.21 Bertrand never revealed the identity of his source and did not even describe French intelligence’s relationship with Hans-Thilo Schmidt until 1972. But Bertrand was happy to mention ‘the agent’ again in his next letter to Denniston two weeks later. Rather than wait over a month for the possibility of new material arriving from Hans-Thilo Schmidt (the next meeting between Bertrand and his crucial source of intelligence on Engima did not, in fact, take place until 29 January 1939). Bertrand had another way of helping things along. Namely, to introduce the British to the Polish code-breakers.
Dear Colonel [wrote Rivet to Menzies on 14 December], After his meeting in London, B hinted that a collaboration with Warsaw on technical areas could offer common advantages, and asked what you thought: you indicated your consent – leaving him to put things in hand – with the assurance that the sensitivities of each party would be paramount. Accordingly, we thought to bring together representatives of each Service in Paris, with a view to putting forward the results of current researches on radio traffic enciphered with the ENIGMA machine used by the WEHRMACHT – thinking that this occasion could be the prelude of a deeper collaboration, both in peace and in war …22
The vital phrase in Rivet’s letter to the British was the one about each Service being subject to its own rules about secrecy (‘avec l’assurance que la susceptibilité de chacun serait placée au-dessus de tout’). Sharing raw materials like intercepts and the fruits of intelligence where the sources can be disguised, is one thing. Sharing know-how which might be turned against you is something completely different. Although the French had no particular code-break
ing insights to impart, they were fully aware of how difficult the tri-party collaboration idea could be for their partners. And those partners were not even proper allies: the Franco-Polish pact had been on ice for years, while the British were still disengaged, notwithstanding the unofficial Pimpernel deal brokered by Menzies.
From the French perspective, the game was more subtle. Bertrand was finding out rather more than it might seem. Ostensibly, as French code-breaking activity was weak, French intelligence thrived on purchases. And what better currency in which to purchase intelligence than intelligence itself? The idea of bringing the Poles and the British together, through the agency of the French, would create a perfect arrangement, where the French, as brokers, could see what the British and the Poles were willing to share. In this context, Bertrand put two apparently innocent suggestions forward to his old friend and Polish equivalent, Gwido Langer. Time was pressing, so there seemed to be two options to speed things up. Option 1: let the Germans know that their traffic was being read and thereby get them to change the system. Option 2: if there was some hope of a breakthrough, get everyone round the table and share their knowledge.23 Langer reacted as predicted. From the Polish perspective, Option 1 was out of the question – it would, at one blow, destroy years of success. But by agreeing to Option 2, Langer was in effect signalling (despite the word ‘hope’) that the Poles had a reason why they ruled out Option 1. Bertrand had found something out of great significance: the Poles must have made some progress with the machine. Now the stakes were higher. If the three countries could co-operate, perhaps they could break Enigma. The question now was whether either of the groups of guests would show their cards.
The founders. Jan Kowalewski (seated, right), who founded the Polish code-breaking service during the Russo-Polish War, together with his successor as head of the service Franciszek Pokorny (seated left), Maksymilian Ciężki (standing right), other members of the service and a lieutenant from the Japanese Imperial Navy, in 1925. (Barbara Ciężka)
Jadwiga and Antoni Palluth. The AVA company, run by cryptanalyst and engineer Antoni, built the machines which broke Enigma in the 1930s; Jadwiga faced down the Gestapo during the war. (Krystyna Palluth-Tunicka)
Wiktor Michałowski. Photographed in First World War German Army uniform alongside members of his family, Michałowski was one of the first to try to break the Enigma cipher. (Piotr Michałowski)
Palac Saski 1. Marshal Józef Piłsudski takes the salute outside the headquarters of the Polish intelligence service on Independence Day 1926. (The Józef Piłsudski Institute, 238–246 King Street, London)
L’homme à tout faire. Rex, the French intelligence service’s spy handler and passport trafficker, whose mesmeric gaze kept the agents producing the precious goods. (Christie Books)
The magician. Gustave Bertrand pictured in 1934, shortly after the crucial handover of stolen Enigma documents. (© Service historique de la Défense, CHA Vincennes, GR 14YD 755)
An expensive holiday. Hans-Thilo Schmidt relaxing with his wife; the holidays and other habits were funded by the French in payment for stolen secrets. (Gisela Schmidt)
Secrets from the bathroom. A page from the Enigma operating manual, showing the special features of the Wehrmacht model, as photographed by Gustave Bertrand (© Service historique de la Défense, CHA Vincennes, DE 2016 ZB 25/5)
The code-breakers: (Top left) Marian Rejewski; (top right) Marian and Irina Rejewski in Warsaw in 1935; (below left) Jerzy Różicki; (below right) Henryk Zygalski. (Janina Sylwestrzak/The Enigma Press)
Dilly Knox, the British code-breaker who made a successful attack on the no-plugboard Enigma before the war.
The Commander. Alastair Denniston, who had to keep Knox from upsetting the fragile X-Y-Z co-operation. (Reproduced with kind permission of the Director, GCHQ)
007. Wilfred Dunderdale, aka ‘Biffy’, was the liaison officer behind Anglo-French intelligence.
Bomba. A replica, on display at Bletchley Park, of the Polish Bomba, the first machine designed to find Enigma settings. (© Andy Stagg)
Bunker in the woods. The unassuming building in the woods at Pyry, beneath which the code-breakers worked and where the July 1939 revelations took place, photographed during the German occupation of Poland.
Hotel Bristol. The British delegation to the Pyry conference stayed in Warsaw’s poshest hotel. (Author’s collection)
Pimpernel No. 52 (© Service historique de la Défense, CHA Vincennes, DE 2016 ZB 25/5)
Enigma reborn. A reconstituted Enigma machine, made in France to the Polish design based on Marian Rejewski’s reverse-engineering. (The Józef Piłsudski Institute, 238–246 King Street, London)
Palac Saski 2. Führer Adolf Hitler takes the salute in October 1939, presumably unaware that the Enigma machine had been reverse-engineered in the building behind him.
The Château de Vignolles. Located near Paris, the elegant château was the home of the exiled Polish code-breakers until the invasion of France in 1940. (Author’s collection)
Before the deluge. Left to right: Gwido Langer, Gustave Bertrand, and Kenneth ‘Pinky’ Macfarlan in the sunshine of the phoney war.
Team photo. Michałowski (1), Smoleński (2), Szachno (3), Paszkowski (4), Ciężki (5), Sylwester Palluth (6), Gaca (7), Langer (8), Antoni Palluth (9), Bertrand (10), Marie Bertrand (11), Graliński (12), Captain Honoré Louis (13), Różycki (14), Rejewski (15), Fokczyński (16), Zygalski (17), photographed in 1941. The others are unidentified. (Anna Zygalska-Cannon)
Fake ID. Maksymilian Ciężki becomes ‘Maximilien Muller’, born in Strasbourg, for his safe conduct to Algiers. (The Józef Piłsudski Institute, 238–246 King Street, London)
5
HOW THEY BROUGHT THE GOOD NEWS FROM GHENT TO AIX
I sprang to the stirrup, and Joris, and he;
I galloped, Dirck galloped, we galloped all three;
‘Good speed!’ cried the watch, as the gate-bolts undrew;
‘Speed!’ echoed the walls to us galloping through;
Behind shut the postern, the lights sank to rest,
And into the midnight we galloped abreast.
R.B. Browning
‘How They Brought the Good News from Ghent to Aix’
In 1938, Antoni Palluth began to receive some strange commissions in his capacity as partner in the AVA radio and precision electrical equipment business. One of the strangest was to build, on behalf of the Biuro Szyfrów, a sex machine.
The machine had six wheel-like objects, arranged in two sets of three. The machine also had a bank of switches and lightbulbs and a variable resistor to ensure that the current did not blow the bulbs if only one or two were lit up. Once again, the wheels were synthetic Enigma rotors, but this machine was not a duplicate Enigma machine.
The machine was nicknamed the ‘sex machine’ because its job was to find females.1 Roughly 40 per cent of the time, a repeated letter showed up in the encrypted version of the six-letter indicator sequence which the Germans used to tell the recipient of a radio message how to set up the Enigma machine for the incoming signal. These repeats were called females as a play on words, since in Polish they were called samiczki for ‘the same’, while samica means ‘female’.
Officially the device was called the cyclometer, because its job was to find the length of cycles (trails of letters) that could be traced from message ‘indicators’ enciphered using the same rotors orientated in the same starting-position. As early as 1932, Marian Rejewski had noticed that indicators had a characteristic cycle-length depending on the combination of rotors being used. From the catalogue of characteristics – one for each of the 17,576 possible orientations of three rotors, multiplied by six for the six different ways you could put three rotors into the machine – it was possible to narrow down, even to identify exactly, which three rotors were in use. The cyclometer was designed to speed up the process for matching cycle-lengths to a rotor setting by lighting up a number of bulbs corresponding to the length of the cycle. Single-letter cyc
les were particularly prized, because they could be seen in a single intercepted message, giving the code-breakers a quick way to create a short-list of the possible rotor settings. Single-letter cycles equate to ‘females’, and so in time cyclometers came to be called sex machines.*
Palluth received an even odder commission later in the year. This time he was asked to make a ‘bomb’ and this time the peculiar name was not even a nickname. The Biuro Szyfrów was not getting into explosives and AVA was not an armaments factory: the odd name was something the code-breakers had given to a new cryptological device.
We will probably never know for sure how the cryptological bomb (in Polish, bomba) came to be given that name. There are various theories.2 One is that it made a ticking noise, as it had rotors, just like Enigma rotors, which were driven around all 17,576 possible positions by a central motor and it stopped when it found a possible solution to the problem of how the Germans had set up their Enigma machines for the day. Another is that the machine was designed while the Poles went out, as was their habit, for an afternoon dessert. They enjoyed chocolate-coated ice cream, a famous black spherical confection designed in France to look like the explosive and thus called a bombe. A further possibility is that bomba is a colloquial exclamation when something really good is done. Maybe all three are right. The code-breakers themselves seem to have treated it as something of a secret. Różycki got blamed for naming the machine and Rejewski said it was called bomb ‘for lack of a better idea’.
X, Y & Z Page 9
