X, Y & Z Read online

  For, having received a bundle of intercepted messages which were in some sort of numerical cipher, Kowalewski was not content with analysing the call signs and potential evidence of troop movements: he wanted to know what the messages were actually about. He determined to attack this puzzle and soon found that the cipher was a simple enough bigram substitution system with an overlaid twelve-digit key. And the messages were certainly worth reading.

  What the messages revealed was not just what the Reds were thinking, but their appreciation of the Whites. The threat to Poland was not just from the Bolsheviks: if the Russian imperialists regained power, they would want to get their old empire back, right up to the German border. But Kowalewski’s decrypts showed that General Deniikin, the White commander, was being threatened in his rear by the Reds. Being able to see both sides of the Russian civil war amazed the Polish chief-of-staff, General Rozwadowski. The gift was an eagle’s-eye view of the entire strategic situation. It wasn’t good news, but it was just what General Rozwadowski needed to know. From now on, monitoring and decryption were a priority.

  Jan Kowalewski looked like something of a bear, with his imposing physique, but people who got to know him valued him for his sense of humour and his extraordinary, intuitive mind. By putting Kowalewski in charge of Polish decryption a crucial step was taken that would lead to Poland becoming world leaders in the art of code-breaking. For Kowalewski’s first request was to ask for all volunteers who were mathematics professors to be assigned to his team. Before Kowalewski, code-breakers were supposed to be linguists and psychologists. But Kowalewski was an engineer and he was redesigning the profession of cryptology.

  The new, scientific approach soon showed its power. Over three days in early July 1920, crucial messages revealed new Russian operational orders given to coordinate the operations of Budyonny with the other Red Army forces invading Poland:

  An order for the Army of the South-Western Front … The 14th army, taking into account the tasks of the Cavalry Army, will break the resistance of the enemy on the line of the River Zenic and will use the full force of its assault team when carrying out a decisive offensive in the general direction of Tarnopol-Przemysl-Gorodok …

  [signed] The Commander of the South-Western Front YEGOROV

  Member of the front’s Revolutionary Council STALIN

  Chief of Staff of the South-Western Front PIETIN

  24/VII-1920

  Deciphered/checked against original: Kowalewski4

  These orders spelt out a new offensive:

  The enemy himself informed our headquarters precisely of his moral and material state, his strengths and losses, his movement, of victories attained and defeats sustained, of his intentions and orders, his headquarters’ stopping points, of the deployments of his divisions, brigades and regiments, etc … We were able to follow the whole operation of Budyonny’s Horse Army in the second half of August 1920 with simply incredible precision … On 19 August we monitored, and on 20 August read, an entire operational order from Tukhachevsky to Budyonny, in which Tukhachevsky states the tasks of all his armies.5

  The intelligence went straight to the Chief of Staff and some even into the hands of the Commander-in-Chief, Józef Piłsudski himself.

  It just got better and better. The radio men intercepted messages that told of the Bolsheviks’ order of battle, the dispositions of their forces, and even details of new ciphers they were going to use. They could overhear the arguments about the gap between Tukhachevsky’s army and Budyonny’s. The Red Army under Tukhachevsky was entirely stretched out: it might make sense for the Russians if they were going to encircle Warsaw, but it made them vulnerable, providing Piłsudski could move his troops into position. Piłsudski raced his troops through the gap, leaving Warsaw wide open, so he could encircle Tukhachevsky from the rear in a classic Napoleonic manoeuvre. Badly outnumbered, Piłsudski was taking an enormous gamble based on the intelligence supplied by Kowalewski.

  On 12 August 1920, Kowalewski’s unit picked up a message several pages long. It was obviously important, not just due to its length, but also because it was in a new cipher. Kowalewski’s team were at the top of their game: it took them just an hour to work out the new system and to decipher enough of the message to get its gist. The decisive attack on Warsaw was due on 14 August and there was only just time to react. Being strung out over hundreds of miles, communication between the Russian units was dependent on radio. The crucial goal was to keep Tukhachevsky where he was while Piłsudski completed his own manoeuvres. To shift the odds, if only fractionally, Piłsudski ordered that the Russian radio communications be jammed. It would cut off the supply of intelligence but cause delay and confusion among the enemy, and a couple of extra days was all that Piłsudski needed.6

  Miracle on the Vistula. A Russian telegram, countersigned by Stalin, and decoded by Jan Kowalewski, during the Russo-Polish War of 1920. (Instytut Piłsudskiego w Ameryce – Piłsudski Institute of America)

  The troops were moved. Spearheaded by General Władysław Sikorski, the great roll-up of Tukhachevsky’s army began. The surrounders were themselves exposed to locally superior numbers, which allowed them to be defeated in detail. A hundred thousand prisoners were taken by the Poles; in fear, the Red Cossacks galloped away into East Prussia, where they were interned by the Germans. And Tukhachevsky escaped with the remnants of his army back to Russia to face a grilling from Trotsky and Lenin.

  In Poland, these events were called the Miracle on the Vistula. In August 1914, three very different forms of government had existed on Polish territory. There were six currencies, four legal codes, two railway gauges, and countless languages (even if Russian and German were the only ‘official’ ones). Now, with the country unified and at peace, the nation could become a state and fly its own flag once more: a crowned eagle, with an exuberant tail, on a red and white field. Radio and cryptology had put Poland back on the map. The vital role of Kowalewski’s decryption team had been understood by the authorities and they, perhaps more than any other government in the world, appreciated the importance of supporting a modern approach to the profession. Thanks to the Polish radio men, Europe would remain free of the Bolshevik menace. For the time being.

  2

  ENTER THE KING

  SHEPHERD: Sir, there lies such secrets in this fardel and box which none must know but the king.

  William Shakespeare

  The Winter’s Tale, Act 4, scene 3

  Jan Kowalewski was a little older than Maksymilian Ciężki, so his impact on Poland’s existential struggle of 1920 had been greater. Ciężki himself had been in Poznań, still in radio. His friend Antoni Palluth had been closer to the action, in the front line for much of the conflict, but also in radio. After the war, Kowalewski, the man at the top, was in demand for many roles, political and diplomatic as well as those based in the shadowy world of intelligence. In 1922, he was seconded to Tokyo to help the Japanese improve their own codes and ciphers, to the irritation of an American code-breaker who then found his attacks on Japanese codes thwarted. It was necessary to find a replacement for Kowalewski as head of signals intelligence. So in 1923 – shortly after Maksymilian Ciężki was assigned to the Radio Intelligence Unit – Kowalewski handed over the reins to an agreeable successor, Franciszek Pokorny. At least Ciężki had been under Kowalewski long enough to be noted by the man whom all the authorities revered. For Ciężki to have his appraisal form marked ‘good’ by Kowalewski was no light matter.1

  Following the war, Maksymilian Ciężki had been pursuing a traditional military career. After his father’s death in 1920, although only 21, he’d found himself responsible not just for his own upkeep but also that of eight siblings, five of whom were girls. He needed a stable source of income. For a young man with his record, army life provided a straightforward opportunity for this. Ciężki was commissioned as a lieutenant, completed his interrupted secondary education, and was posted to various places around the country, always specialising in communications. No
w, settled into radio intelligence, he could also settle into family life, marrying Bolesława Klepczarek in 1924. That year, Ciężki’s first son, Zdzisław, was born, followed by Zbigniew in 1926 and Henryk in 1929. Life was relaxed. Ciężki could spend his spare time in the garden. He had a good job, working with good colleagues.

  Antoni Palluth’s path into radio intelligence was somewhat different. In the Bolshevik war he’d been posted to a radio intelligence unit, where he’d been introduced to traffic analysis and cryptanalysis as skills to add to his existing radio interception abilities. He had been on the reserve list since 1921 and was setting himself up in business. When he had returned to civilian life after his own part in the defence of Warsaw in 1920, Palluth had studied civil engineering at the Technical University in Warsaw. Palluth had not given up his interest in radio with the end of the war. Far from it, he’d become a radio ham, with his own call sign (TPVA), a subscription to a number of radio magazines and a love of sending crackly messages from his house through the ether to his friends. Being an amateur radio-ham was not just good fun. There was money to be made in radio too. The military signals branch of the Polish Army needed long-range short-wave radios; masts; interception equipment; transmitters; amplifiers; you name it. Palluth’s friends Ludomir and Leonard Stanisław Danilewicz (call sign TPAV) were thinking of going professional and their mutual friend Edward Fokczyński had already set up shop making walkie-talkies and radio equipment in a small workshop in the centre of Warsaw.2

  In this way, a new business was born. With Palluth and the two Danilewicz brothers providing financial capital and Fokczyński throwing in his premises and existing business goodwill, the new partnership took over a little factory in 43 Nowy Swiat. It was called AVA, by amalgamation of the Palluth and Danilewicz call signs, and the first orders for radio equipment came from the Polish Army’s signals intelligence section. Radios the size of a credit card holder were made for use by Polish agents in foreign territory: AVA technology was miniaturised, sophisticated, and extremely secret.

  Yet Antoni Palluth was no ordinary young entrepreneur. He was a paid-up member of military intelligence and his role as factory manager was a cover for a wider range of clandestine activities. Poland’s need to keep a close eye on Russia and Germany had not disappeared with the Treaties of Versailles and Riga: only an optimist would assume that either of those neighbours was comfortable with the new order. Foreign ciphers were every bit as important now as they had been during the war and Antoni Palluth was one of the secret team whose job was to find out what evils lay in the plans of the Germans. In the evenings, an officer would come round to the Palluth household with cryptological problems for Antoni to work on. For, since June 1924, as well as his ostensible day job for as a radio factory manager, he had been working for the Second Department – the intelligence section – of the Polish General Staff, and he held a post in Maksymilian Ciężki’s German section of the Biuro Szyfrów, the cipher bureau.

  Despite the help he was getting, Ciężki had a problem. Until 1928, the German section of the cipher bureau had been operating smoothly; indeed, one might say it had been going by the book. The book in question was by General Marcel Givierge, the head of French military cryptanalysis in the Great War, who had done the unthinkable and written up the cryptographic techniques of his era and published them for all to read. The book begins with the simplest form of cipher (in use since Caesar’s time) and details substitution systems, transposition systems, bigram methods, double-key substitutions, book codes and even some mechanical methods for coding. It also covered code-cracking techniques, such as frequency analysis for substitution ciphers, along with methods for finding key lengths. The toughest problems were known to the Poles as Doppelwürfelverfahren, or double dice, and these arose from a hand-based cipher system in current use by the German military. Double dice involved reshuffling the letters of a message according to a predefined scheme – a bit like a giant anagram – and then reshuffling them again using a different scheme. In Givierge’s book the double dice system was laid bare. That didn’t necessarily make it easy to crack: the code-breakers needed to find the two keys to the double-transposition system in order to tease the plain text out and this took sweat, concentration, and plenty of squared paper.

  Ciężki’s team could get results against Doppelwürfelverfahren, at least until 1928 when matters suddenly became a great deal tougher. In February 1926, the German Navy had started sending messages that were obviously not enciphered using a transposition method at all. You could tell this just by looking at them. With a transposition cipher, letters occur with the same frequency as in regular German, but the frequency of letters in these new messages had the purity of randomness. Some form of substitution was going on. Old-fashioned substitution ciphers had, over the centuries, yielded to frequency analysis, but not these new ones. There could be only one explanation for the perfect equality of the letter count in the intercepted signals. As foreseen by General Givierge, the substitution was being done by a machine. The affairs of the German Navy were of little interest to the German military section of Polish Army radio intelligence, but then, in 1928, the German Army started doing something similar.

  Polish intelligence had heard that the machine was called Enigma. But what was this thing, the Enigma machine? The cipher seemed resistant to all the usual methods cryptanalysts could bring to bear. First Ciężki and then Palluth had a go at breaking it. Then they called in their colleague, Wiktor Michałowski, to see what he could do. Michałowski had been born in 1895 and, like Ciężki, had been in the German Army and later participated in the Wielkopolska uprising as a communications officer. His subsequent career had been a bit more colourful than Ciężki’s: Michałowski had spent a period accompanying silent films in cinemas on a piano with his home-composed scores and also in trying his hand at business, with a pencil factory which had failed. In 1928, just as Ciężki was wrestling with the new mysterious machine cipher, Michałowski joined the Polish General Staff and was assigned to military intelligence. Coming from Wielkopolska, Michałowski was another natural German speaker and therefore an ideal fit for the German ciphers section under Maksymilian Ciężki.3

  In about 1926, the Second Department of the Polish General Staff invested in an Enigma machine. Enigmas were commercially available, and internationally many intelligence services were considering the advantages of mechanical ciphers over the old-fashioned code books and pencil-and-paper ciphers which had been used – and successfully broken – in the Great War. The Polish purpose in obtaining an Enigma machine was not for encrypting their own messages, however. They wanted to study the assembly and mechanism of the machine with the goal of breaking the German codes.

  At first glance the Enigma machine appeared to be like a typewriter, but instead of a roller with paper the machine had an array of small lamps: little torch bulbs illuminating the letters of the alphabet. When a key was pressed, one of the lamps lit up and the letter was always different from the letter on the key which had been pressed. The conversion of letters happened as a result of the wiring in the machine: pressing the key down activated an electrical circuit which flowed through three cylindrical rotors, then around a ‘reflector’ which turned the electricity back through the rotors again but on a different path, and thence to the lamp. If every signal behaved in the same way, the resulting cipher would be simple enough to break, as each letter pressed down would have a fixed letter that had lit up. But the beauty of the machine for those wanting to create a near unassailable encryption was that the path of the wiring changed with every single keystroke.

  The physical action of pushing down on an Enigma key caused the rightmost of the rotors to rotate by one place. Press a letter, T, say, once and a bulb might come on the B. Press T again and instead of B, you might light the U. There was no obvious continuity at all between the letter T and its various appearances in the coded message. Moreover, once during the full 360-degree rotation of the first rotor the second rotor would tu
rn on one place. And likewise the third rotor moved on too, impelled by the turning of the second. With this mechanism the path of the electrical wiring was different for each new letter in the message until, after 26×26×26 = 17,576 keystrokes, the three rotors returned to their initial position.

  Buying an Enigma machine helped the Poles not one jot.

  Together with Antoni Palluth and Leonard Danilewicz, Ciężki and Michałowski tried a pencil-and-paper method for attacking Enigma, mapping out the transformations done by the coding rotors in the machine on to strips of celluloid, which could be slid one against the other to copy the path taken by the current in the machine. But this was no good. The celluloid maps followed the wirings of the commercially

  available Enigma machine but yielded no results when applied to the German military traffic. The German military Enigma machine had been modified in unknown ways. The Enigma company’s marketing material said they’d make you rotors to order, so it was a fair guess that the German military were using different rotors; perhaps also they had changed the wiring in the reflector or the way the plugboard was connected up to the unit containing the rotors. And they could have added some sort of extra contraption to the machine to make the encipherment jumble even more complicated. Those changes were unknowns and something special would be needed to help understand the new type of box which the Germans were actually using for the impenetrable radio messages being received in Poznań.