The ether was open. Into it all nations could discharge messages of the highest importance. Equally important was the complicating factor that all nations might, if they wished, receive or stop these messages. We were all involved in the problem of safeguarding our own information, of discovering and nullifying that of the enemy. It was not enough merely to prevent the latter from giving messages to its own forces and allies. It was vital that we should receive those messages and turn them to our own purpose.i
Parallel with the open conflict that raged between 1939 and 1945 there were other, hidden wars, and what they all had in common was that they were wars of communication, in which success depended on a flow of concealed and closely guarded information. Sometimes this meant a smuggled written message, at others a secretly transmitted wireless signal, or weeks and months of eavesdropping on the radio traffic of the enemy.
Many thousands of people took part in these secret wars. Some trained for long periods to carry out sophisticated campaigns of espionage; others committed a single impulsive or premeditated act of defiance. One of these hidden conflicts was the struggle of underground forces against the occupying Germans or Japanese. Another was the battle to obtain secrets, or carry out sabotage, by the spies who served both sides. A third was the war waged by armies of clerks, typists, linguists, analysts and assorted academics to discover the intentions – and weaknesses – of the enemy by breaking its codes. In Britain, it was only in 1974 that the publication of Group Captain FW Winterbotham’s book The Ultra Secret revealed the huge significance of this work and the extent to which it had contributed to victory.
The war imposed the necessity of secrecy not just on official and military personnel but on people who found themselves displaced, imprisoned or in some way unable to express their feelings freely. One example was the deliberate damaging of Axis war materiel made by forced labour in German-controlled countries. Alfred Spickett, a wireless officer in the British Merchant Navy, recalled an attack on his vessel by enemy aircraft in which little damage was done:
What none of us realized at the time was that we had in fact been hit by two aerial torpedoes. Very fortunately for us, both had failed to explode.
Anyone reading this might think it odd that both torpedoes had failed to explode. I must admit I did myself. A possible explanation, given later by a naval bomb expert in Rosyth, was that they were getting quite a number of torpedoes and bombs which failed to explode, and they were sure it was due to sabotage in a number of factories in German ‘occupied’ countries. I remember this same chap telling me that in one of the bombs they had subsequently dismantled (which had been dropped near London) they had found a hand-written note in the front section of the bomb which made it clear that it had been made in the Skoda works in Czechoslovakia. Written in English, the note had gone on to say: ‘This is the best we can do to help you.’ These particular factory workers were successfully interfering with the mechanism in both bombs and torpedoes, thus preventing them from exploding.
While some put their lives at constant risk in enemy territory, others laboured at routine deskwork, far from scenes of conflict or danger. The most significant of these were the men and women who sought to break the codes and ciphers of the Axis powers.
Communications, whether by radio or letter, had to be shrouded in secrecy, with use of passwords, code words, euphemisms and gibberish to baffle enemy eavesdroppers. This was characteristic of all ‘secret wars’ and was much in evidence by the start of the conflict in 1939. A story published a year earlier in the United States, when many Americans knew little about the shadow lengthening across Europe, made use of a code as its central theme. Address Unknown, a short novel by Katherine Kressman-Taylor, became a bestseller in the US. It greatly increased awareness of the evils of Nazism and has since come to be regarded as a highly important historical document (the Nazis themselves placed it on their list of banned literature). The book is set in the years in which Hitler consolidated power. A Jewish art dealer in America takes revenge for betrayal on his ex-partner, a German living in Munich. By the simple device of sending frequent letters – knowing that they will be read by the Gestapo – whose meaningless contents suggest a developing plot, the dealer is able to have his former friend ruined and arrested. One of them reads:
February 15, 1934
Dear Martin,
Seven inches of rainfall here in 18 days. What a season! A shipment of 1,500 brushes should reach the Berlin branch of your Young Painters’ League by the weekend. This will allow time for practice before the big exhibition. American patrons will help with all the artists’ supplies that can be provided, but you must make the final arrangements. We are too far out of touch with the European market and you are in a position to gauge the extent of support such a showing would arouse in Germany. Prepare these for distribution by March 24th: Rubens 12 by 77, blue; Giotto 1 by 317, green and white; Poussin 20 by 90, red and white.
Young Blum left last Friday with the Picasso specifications. He will leave oils in Hamburg and Leipzig and will then place himself at your disposal.
Success to you!
Eisenstein.
Once war began, millions of people made use of personal codes to keep in touch with their friends and relations in circumstances where security considerations, or capture, might rob them of freedom of expression. Lieutenant GP Darling, RN, whose usual form of address in letters to his parents was ‘My Dearest Ma & Pa’, sent them a set of instructions when he was called to active duty. He was concerned not so much with letting them know he was alive and well as with passing on to his superiors a report of what had happened, plus confirmation that sensitive materials aboard his vessel had been disposed of. After personal news and messages came this:
Now for something serious. I must make provision for being captured by the enemy. As soon as possible I will write to you and the manner of my writing will give you the following information. Forward the decoded version to Vice-Admiral Submarines, Northways, Swiss Cottage, London. If in doubt get in touch with Reggie Drake at Blockhouse.
(1) ‘My Dearest Mother & Daddy’ = Confidential books and asches iidestroyed.
(2) ‘My Dearest Mother’ = Confidential books destroyed.
(3) ‘My Dearest Father’ = Asche destroyed.
(4) ‘Best Love, Godfrey’ = Sunk by depth charges.
(5) ‘Love, Godfrey’ = Sunk by mine on surface.
(6) ‘Best love from your loving son, Godfrey’ = Sunk by torpedo on surface.
(7) ‘Love from your loving son, Godfrey’ = Rammed.
Keep this locked away and keep it to yourselves please.
Understandably, there were cases where cryptic meanings were ascribed to what were actually straightforward words or phrases. One such incident took place in 1943 in Burma, the result of a misunderstood abbreviation. The regional headquarters of OSS (the Office of Strategic Services, which was running clandestine American operations) in India received a radio message asking for supplies from one of its officers, who was in command of a remote unit of guerrillas recruited from the local Kachin people. Included in this were the letters CMA. These baffled the operator on the receiving end, and eventually it was decided that they stood for ‘Citation for Military Assistance’, presumably suggesting that an award be created to acknowledge the tribesmen’s loyalty and support. This was duly done, the result being a silver medal adorned with the American eagle and hung from a green silk ribbon decorated with embroidered peacocks. A number of these were dropped with the rest of the supplies and were handed out to the guerrillas, who greatly appreciated them. Only later did it become clear that CMA had stood for ‘comma’.
Commercial radio and BBC services offered possibilities for secret communication. Agents could, for instance, listen for pre-arranged signals in the form of a particular piece of music. It was for this reason that the United States Government forbade radio stations to play any music requested by listeners for the duration of the war. When broadcasting to occupied Europe, the BBC made regular use of the first bars from Beethoven’s Fifth Symphony as an ‘opening theme’ to its nightly broadcasts. Initially this habit puzzled both the German security services and the resistance. Since the music was written by one of their country’s greatest composers, the Germans could scarcely object to it; they had failed to realize that the first four notes sounded like a Morse code signal: dot-dot-dot-dash, representing the letter V, for victory. It was a subtle indication that, no matter what Axis propaganda was telling the local population, the Allies were winning the war.
The most famous use of radio code words was the communication between the European resistance movements and Britain. Agents in France, the Low Countries and Scandinavia would send wireless signals in code (often at great personal risk, for the Germans used location-finding equipment to track them down) and listen to the nightly broadcasts of the BBC from Bush House in London for a reply. Twice every evening a batch of announcements, disguised as and including personal messages, was broadcast in the appropriate language. Many of these sounded absurd. Some of them were deliberately intended to confuse and irritate listening German counter-intelligence operatives, while others had clear meanings for specific groups. A British officer, CW Kemson, who served with a unit of French maquisards in 1944, described what happened when one of its leaders was wounded and required medicines from England:
His physical state was very poor and it worried them. They could only communicate with London at predetermined times and they had to wait until the 24th to ask for urgent supplies. This took place at 9.30 a.m.
Michel sent the message in the code transcribed, and after ‘contact’ they were overjoyed when Gineste received confirmation signifying that an important message was to follow. The same day at 7 p.m. and 9 p.m. the BBC sent among the personal messages ‘The height is at the corner.’ At 11 p.m. a drop was received at Boulieu, and several hours later George was out of danger.
The message to London in code had read as follows:
Send Boulieu serum anti gangrene and anti tetanus for Maxime and other wounded awaiting medical equipment STOP column of about 270 vehicles transport to Figeac. Remain in contact with HQ STOP.
The information exchanged in these announcements was critical to the success of the D-Day operation. It informed the Allies of the strength of opposition they might expect in Normandy, and enabled the resistance to learn where the attack was coming, so that they could sabotage strategic transport routes or tie down German troops.
Radio could, of course, be used to deceive as well as inform. Hitler and many of his generals believed that an Anglo-American invasion, which they anticipated in April or May 1944, would be launched across the Straits of Dover in the neighbourhood of Calais, since this was the narrowest part of the English Channel, and that an attack anywhere else would merely be a feint to draw the defenders away. The Allies had no intention of assaulting this heavily protected region but, knowing that their radio communications were listened to by the enemy, set out to create an entire phantom army that would be based in south-east England in apparent readiness to make the move that Hitler was expecting.
This meant the invention of fictitious divisions, all with numbers, commanding officers and even specially designed insignia. It also meant the invention of radio call signs for each unit, and regular sending of Morse signals between them, so that enemy listeners could use these to track the ‘movements’ of the different formations. It required an endless exchange of radio messages between non-existent units – the relaying of orders, the requesting of supplies, the movement of immense columns of vehicles and the creation of camps, supply dumps and headquarters. From October 1943, German listeners began to hear, amid this storm of chatter, references to a body called FUSAG (First United States Army Group) and to the divisions and army groups attached to it. From these they deduced that 34 American divisions were massing in Kent and Essex (in fact 11 of them did not exist) and estimated that a total of 79 Allied divisions would be ranged against them in the invasion. The Allies, in fact, had only 47.
Radio was not the only means of conveying this sense that a mighty army was waiting to strike. Nevertheless nothing breathed life into the ghostly echelons of First Army Group more effectively than its radio traffic. For days and even weeks after the landings in Normandy, the Germans continued to waste time and resources on this force. On 9 June ‘Cato,’ a trusted (but ‘turned’) Axis agent, sent a message from England saying:
The present operation, though a large-scale assault, is diversionary in character.
Later in the month a situation report by German Intelligence noted that:
Not a single unit of the 1st United States Army Group, which comprises around 25 large formations north and south of the Thames, has so far been committed.
Clearly, continuous and detailed access to the enemy’s communications was a vital factor in winning the war. It not only revealed the intentions, strengths and weaknesses of the Axis powers, but their reaction to Allied measures, initiatives and deceptions. Their communications were, naturally, sent in code for reasons of security, and it was therefore a major task of Allied Intelligence to gain access to these.
The reading and interpretation of ciphers and coded intelligence was extremely difficult, for it was not simply a matter of breaking a single code and then sitting back to listen in. While the Germans and Italians used a cipher machine invented in 1918 (and first used for banking security) for sending secret messages, there were many variations of the codes they employed. Germany’s army, navy and air force, for instance, as well as its foreign office, each used different configurations. In addition, these were changed at frequent intervals as a matter of routine. It was also necessary that they keep track of a huge flow of signals dealing with every aspect of combat, supply and administration. Group Captain Fred Winterbotham, who coordinated the Special Liaison Units throughout the war, remarked:
At the height of the conflict the German war machine was sending well over two thousand signals a day on the air. It will be recognised therefore, that when, from time to time, we were able to intercept a number of signals and break the cipher, their contents covered a very wide field.
The Allied codebreakers, working in the Government Code and Cipher School at Bletchley Park and at other sites, therefore had constantly to attempt to reopen sources of information that had been closed to them. In some cases, they never succeeded in breaking a code; in most, however, they were able to unravel the mysteries time and again.
Those who worked on the codes might see no more than a single ‘piece of the jigsaw’ and have no understanding of how the paper in their hand fitted in with overall developments. The information they assembled was code-named Ultra (for ultra-secret). It was analysed and passed to commanders in the field through military personnel called SLUs (Special Liaison Units) who were attached to the various armies and lived with the troops in the theatres of war. Though Ultra was obtained by British intelligence, much of the information was shared with the United States Government, at Churchill’s request, even before America entered the conflict. Once the two nations were officially allies, SLUs were attached to US forces and Ultra was made available to senior American officers on exactly the same basis as to British.
For those who were engaged in the cryptographic war at secret locations in Britain, their distance from the fighting and the often numbing monotony of their work did not prevent them from sometimes sharing the triumphs and tragedies of their comrades overseas, for they could be closely following Allied progress through the signals they deciphered. Mrs EJE Openshaw, a member of the British ‘Wrens’ (WRNS, the Women’s Royal Naval Service) who, though based on land, was subject to the shipboard routine of naval ‘watches’, described the view of the war that she developed through her work:
My work in Derby House (Liverpool) was in the Communications Department, where I worked, with many others, as a Cipher Officer. Cipher messages, broadly speaking, were ‘Secret’ or ‘Top Secret’, and could only be dealt with by officers, and codes were ‘Confidential’, and dealt with by ratings. We were divided into four Watches which ran from 0800–1300, 1300–1800, 1800–midnight, midnight–0800. This was vital work, as all warships at sea received their orders in code or cipher, in Morse Code by wireless telegraphy. It was usually very busy on duty, sometimes hectic, during attacks on convoys, very exciting, and quite horrifying, as we knew the events at sea as they were happening. We knew about the loss of HMS Hood, and [that there were] only three survivors. We knew of the hunt for, and sinking of, the Bismarck, and were ecstatic. The Bismarck had sunk the Hood. When you are in a war for survival, you can only rejoice when the enemy suffers losses. You have a totally different perspective from pundits writing history years later.
Tension, monotony and occasional euphoria, the characteristic feelings of millions of troops involved in the ‘shooting war’, were experienced in equal measure by those, at Bletchley and elsewhere, whose contribution to the war effort was to listen to enemy signals. The fact that their role could never be revealed gave them a sense of shared purpose and pride, which perhaps exceeded that of any other unit.
Espionage is as old as humankind and it has proved as invaluable a source of information in peace as in war. It was always useful to know the intentions of neighbours and rivals – and never more than amid the tangle of European alliances that preceded the Great War. Diplomatic correspondence, which was normally in some form of secret writing, was therefore subject to scrutiny, and those with an ability to work out its meanings were in demand by both foreign ministries and war offices. Diplomatic messages were usually encoded, then enciphered to provide an additional layer of secrecy, and this practice continued with regard to military communications once World War I began. By 1914 the armies and navies of many nations included intelligence specialists devoted to the deciphering of signals. Hugh Hoy, who was one of these, remarked in his book 40 O.B. published in 1932:
Obviously none of the Powers would use the wireless for conveying vitally secret messages without disguising these as cunningly as possible. It was essential then that we should be able to read the enemy codes, constructed of the most baffling combinations of letters and numbers that the wit of man could devise. It was this necessity that led to the formation of a special department at the Admiralty, a department that in many ways may now be considered to have been the hub of the mechanism of the Great War.
As in World War II, the secrecy of this work was so well preserved that many participants never admitted their involvement in it, as Hoy remembered:
Its existence was not made publicly known until a few years ago. One of our most trusted naval officers who was at the Admiralty during the whole course of the war, told me that he learned of 40 O.B.’s existence some years after the armistice. One officer who worked in 40 O.B. carried his secret to the grave with him, and not very long ago his widow learned from me for the first time of the responsible work of her husband, whom she had understood to be doing some vague ‘clerking job’ at the Admiralty.
Room 40, or ‘40 O.B.’ (Old Building), at the Admiralty in Whitehall, was an address that became as famous in codebreaking circles as Scotland Yard in police lore. The Admiralty provided the most effective service of this sort in Britain, and the cryptographic department of British Naval Intelligence was directed from Room 40 by Admiral Sir Reginald Hall. Imaginative, innovative and dynamic, Hall was assisted by the brilliant codebreaker Sir Alfred Ewing. Together they created a highly efficient department that was to play a major role in the defeat of the Central Powers. One of their notions, prompted by a shortage of manpower, was to employ women as cryptographers. This would not be the only respect in which the codebreakers of Hitler’s war would be following a precedent set in the previous conflict.
Until the invention of wireless, the major means of international communication had been the telegraph cable. At the start of World War I Britain, with control of the seas, lost no time in denying this to the enemy. Hoy, working in Naval Intelligence, remembered that:
As soon as war was declared in August 1914, there were strange happenings in the deep waters which the vessels of the British Fleet rode. From the waves there slowly rose great snake-like monsters, thick with slime and seaweed growths, responding reluctantly to the grapnels which dragged them to the surface and beyond and laid their bulk athwart the deck of a boat, soon to be returned, severed and useless, to the depths. They were Germany’s cables by which she maintained direct communication with the rest of the world. Thus the British navy struck the first blow at the enemy’s war machinery. As far as telegraphs were concerned Germany was now isolated. She had two sources of communication left to her – cables via neutral countries, and wireless. Nor could she retaliate, and our British cables functioned throughout the War with very little interference.
In 1914 wireless had become, for the first time, a weapon of war. As a widely used means of communication that could be listened to by anyone, it offered unprecedented opportunities for learning of the enemy’s plans and movements. Within days of the outbreak of World War I, the army had set up a department to monitor foreign telegraph traffic, while the Admiralty established a series of radio listening posts throughout the country. These were administered by the General Post Office (they were, in a sense, a natural outgrowth of the government’s centuries-old practice of reading the mail of dubious foreigners) and on their staffs were academics and others with the relevant language abilities. One of them was Alistair Denniston, who would later take charge of the Government Code and Cipher School and would direct the work at Bletchley Park in World War II. Another member of staff, the professorial EWB Gill, remarked on an aspect of the work that would be noticed by codebreakers in both wars: the rigidity of German official thinking – and training – made their procedures and patterns of thought easy to predict and to follow. He said:
Nobody could desire more admirable opponents than the Germans for this class of work. The orderly Teutonic mind was especially suited for devising schemes which any child could unravel.
Even without access to the secret codes of an enemy, a great deal could be learned from studying his signals. Anyone who tapped out Morse or telegraph messages would have an individual style, known as a ‘fist’, as distinctive as handwriting. With practice and an attentive ear, it could be identified by listeners, who could trace its place of origin. If it originated from a general’s headquarters his movements could then be followed, since his signaller would naturally travel with him. Because units usually gave an individual ‘call sign’ when sending messages, their positions could also be established.
Admiral Hall’s staff concentrated on cracking German codes, but he realized, as would his successors in the next war, that obtaining an enemy code book would save a great deal of time and trouble. He was extremely fortunate: in two incidents he was able to acquire, without effort, precisely what his department needed.
On 20 August 1914 – within weeks of the start of hostilities – the body of a German sailor was washed up on the Gulf of Finland’s Russian coast. He was the wireless officer from the cruiser Magdeburg, which had been sunk by Russian warships, and in his arms was a code book that revealed the secrets of the High Seas Fleet Naval Code. The Russian navy, lacking a sophisticated intelligence service of its own, passed on this priceless treasure to its British counterpart. The second gift from fate was another code book. This was found in December 1914 in the nets of a British trawler in the North Sea. These books were undeniably useful, but they were not enough. As all the belligerent powers used different codes for each of their armed forces, government departments and diplomats, and as codes were in any case altered continually to protect secrecy, Hall had to seek out information wherever it could be found. One of his great successes was the recovery of a code book left behind by the German Vice-Consul to Persia, who had had to flee when caught attempting to sabotage an oil pipeline. This gave access to important diplomatic codes throughout the Middle East, a vital and sensitive theatre of war.
He was particularly keen to get hold of naval codes, and issued orders that any wrecked enemy craft within reach was to be carefully searched. This produced no immediate result, but he hit upon another idea. He was aware that the German Admiralty had cracked several British codes, and that eavesdroppers were therefore listening to his own radio traffic. They were interested in the signals of British minesweepers, which reported whenever they had cleared a channel through the fields of mines that German submarines had sown. On hearing these reports, a U-boat would be despatched to plant more mines and thus endanger British vessels.
Hall knew that one German boat, UC44, was engaged in this activity off Waterford in Ireland. He therefore had a signal sent announcing the sweeping of a channel through the minefield there. UC44 was immediately ordered by Berlin to return and sow more mines. The trap was then sprung; the boat entered the area (in which no ‘sweeping’ had taken place), hit one of its own mines and sank. Hall had deliberately engineered this sinking in shallow coastal waters so that he could send down a team of divers, and his ingenuity was handsomely rewarded when they brought up the very prize for which he had been hoping: a copy of a newly introduced naval code book. The seeking of the enemy’s secrets from captured or damaged submarines, which was to be vital to the gathering of intelligence in the next war, thus had an important precedent.
Hall’s department had realized that the German navy was enciphering its messages as well as coding them, but with the German code books in their hands his staff were soon able to read enemy signals within an average of six to eight hours. Once codebreakers came to understand the mindset of their enemy counterparts, and acquired familiarity with the phrases of greeting, reporting and routine enquiry, they were able to work speedily through encoded radio messages, even when the cipher key was changed every week or, as soon became the case, every day. For more than three years after the finding of the first code book, British Naval Intelligence was able to monitor daily the messages of its German counterpart.
In 1916 Hall’s department carried out an act of deceit that, once again, was to set a precedent for a larger-scale operation in World War II; in this case, he provided a forerunner for the creation of First US Army Group and its build-up for an invasion of France. That year had witnessed the hugely costly British offensive on the Somme, while the French had paid dearly in lives for the successful defence of Verdun. Something had to be done to relieve pressure on these Allied armies and Hall decided that, if the Germans could be convinced that a new attack was imminent elsewhere, they would withdraw troops from the trenches of Picardy to meet it.
He had messages transmitted that suggested a gathering of forces in southern England for an attack across the Channel. Much of the British Army was already fighting in France and Flanders, but in their thrust through Belgium in 1914 the Germans had seized and held most of the country’s North Sea coast, and it was this that Hall pretended would be the target. He invented a ‘North Belgian Front’, to which his operators made frequent reference in their signals, but he went even further. He planted stories in the British press that hinted at a forthcoming major operation, and he had British spies throughout the world instructed to drop hints or make throwaway remarks in the restaurants and hotel lobbies of neutral countries with a view to starting a groundswell of rumour.
He succeeded. Though his efforts did not end the stalemate on the Western Front, he did cause sufficient concern among the German High Command to have thousands of soldiers transferred to the Belgian coast to repel an invasion, with the frittering away of manpower and resources that followed the building and equipping of defences.
It was the decrypters of Room 40 who read what has become known to history as the ‘Zimmermann Telegram’. At the beginning of 1917 the US was still neutral. Germany intended, in February of that year, to begin a policy of ‘unrestricted submarine warfare’ in the Atlantic, which would seriously threaten this neutrality because it would mean that American ships could become the target of U-boat attacks. Should the US enter the war as a result, Germany sought an ally in the Western Hemisphere that could bring pressure to bear on it. Mexico, which shared a long frontier with the US, offered excellent possibilities, and Germany’s foreign minister, Arthur Zimmermann, sent a coded message to his ambassador there. In return for joining the Central Powers, Mexico was to be promised large areas of the south-western US, which it had lost in the 19th century. Zimmermann instructed the ambassador to approach the Mexican president with an offer:
We make Mexico a proposal of alliance on the following terms:
Generous financial support and an undertaking on our part that Mexico is to reconquer the lost territory in Texas, New Mexico and Arizona. You will inform the President of the above most secretly as soon as the outbreak of war with USA is certain. Please call the President’s attention to the fact that the ruthless employment of our submarines now offers the prospect of compelling England in a few months to make peace.
This was sent by Germany’s Foreign Office cipher, which had long since been cracked by Hall’s staff. Britain had been handed a hugely important opportunity to destroy German credibility, and the Admiral had every intention of using it. Like his successors in the next war, who sought to make use of Ultra information but had to conceal its origin, Hall had to decide how the contents of the telegram could be made public without revealing how they had been obtained. After some consideration, he succeeded in arranging an ‘interception’ of the cable in the US itself.
Its publication provoked an uproar and, instead of denial, Germany compounded its humiliation by confirming that the message was genuine but arguing that it was justified by circumstances. Two months later, America was at war with the Central Powers. The Zimmermann Telegram was therefore perhaps the most important enemy signal to be decoded in either World War.
Germany and her allies were not without able cryptographers of their own, as can be seen from their success in reading British naval signals. They claimed after the war to have comprehensively broken the codes of every country except Britain and America – ironically the two most vital ones. Especially proficient was the Austro-Hungarian Army’s cryptographic service, which had been established by an officer called General Andreas Figl.
The end of World War I did not bring an era of peace. The defeated Central Powers were obliged to reduce the size of their armed forces drastically, and their intelligence services were disbanded. Nevertheless, the revolution in Russia now threatened the whole of Europe with Bolshevism. All the Western nations, in a dress rehearsal for the Cold War 25 years later, became concerned with monitoring communications both from and within the Soviet Union. No country, however, succeeded in breaking the USSR’s diplomatic code. Britain was able to use the proximity to the Soviet frontier of its Indian Empire to glean information. RT Jenks, an army signaller of a later vintage, recalled:
I was told by an old and venerable friend that as a young soldier in northern India in 1921 they were intercepting Russian traffic, which was rushed to civilisation by mule!
Throughout the Great War, Britain’s cryptographic and intelligence services had gained a considerable reputation, despite the fact that they had been unable to make a coordinated effort owing to intense rivalry between navy and army. In the interwar era, not only the new balance of power but Britain’s expanded responsibilities overseas underlined the need for continued intelligence gathering. There was now a permanent official organization, the Government Code and Cipher School (GC&CS), whose function was to protect the security of codes used by Whitehall departments as well as monitoring those used by other powers. ‘Y’ Section, the army’s specialist unit for telecommunications, was also established at this time. Jenks, as one of its members, described how:
The basis on which ‘Y’ Section was formed was defined as: ‘The craft of estimating enemy strengths, intentions and locations from the pattern, volume and nature of his wireless communications.’ This does not necessarily involve the actual content of the traffic, although that would obviously be a bonus. The ‘Y’ Service was apparently started in Palestine about 1923. This later became 2 Wireless Regiment and was followed by 1 Wireless Regiment. These have probably the longest continuous histories in the Royal Corps of Signals.
As well as posting specialist military units in Britain’s overseas territories, plans began to be formulated for combating wireless espionage at home in the event of a future conflict. Drawing on the experiences of 1914–18, the government set out to establish a network of radio communications centres. Attention was also paid to searching out similar centres used by the enemy through ‘direction finding’ – the tracking of enemy radio, or radar, signals by using directional antennae. Plans were also made for the recruiting of suitable staff in the event of war. According to Jenks:
The need for the interception of illicit wireless transmissions was recognised in the late 1920s but came to nothing. However, in 1933 it was agreed that the General Post Office should act as the agent for the War Office for the manning, maintenance and technical operation of the service later to become known as RSS (Radio Security Service). The Postmaster General being the regulating and licensing authority for all wireless telegraphy matters. In December 1937 it was agreed that the GPO should build and equip three fixed Intercept and Direction Finding stations. The first was operational in 1938, and with the clouds of war gathering fast, approval was given for the establishment of a network of fixed and mobile stations, and also of an auxiliary ‘observer’ corps of licensed amateur radio enthusiasts (hams, as they were known among themselves).
While Britain retained its prewar and wartime security organizations, and continued to develop them so that by 1939 it had the largest signals intelligence establishment of any nation in the world, Germany had had to begin again from scratch. Though the army’s communications intelligence system had been dismantled, the expertise of some of its members lived on. In Germany a number of private armies, called Freikorps, came into being to preserve order amid the chaos of the early Weimar Republic and to prevent the Communists from seizing power. The members of these were almost all ex-soldiers, and of a conservative, authoritarian bent; many thousands of them later joined Hitler’s own force, the Sturmabteilung (SA). An intelligence unit was set up within one Berlin Freikorps to monitor and analyse the Communist threat.
When the Nazis came to power in 1933, their desire for communications intelligence became acute. Intent on creating a one-party state, the National Socialists sought to consolidate their position by silencing and eliminating all opposition. This was facilitated by spying on the German people, not least through extensive tapping of telephones. One body that carried out this and other electronic surveillance was called the Forschungsamt, or Research Office. It was the brainchild of Gottfried Schapper, a long-time Nazi who during the war had been in charge of the army’s General Headquarters Radio Station. He thus brought to the National Socialist Government the expertise of a military communications specialist and cryptographer. Because it was Goering whom he persuaded of the necessity for such a department, it was to Goering’s fiefdom, the Luftwaffe, that the Forschungsamt was attached.
The organization consisted of six departments. Those that specialized in foreign relations processed a vast amount of intelligence derived from agents’ reports, foreign newspapers and wireless broadcasts, radio traffic ‘in clear’ and coded messages, of which by the outbreak of war they were handling over 2,000 a month. The department called ‘Bureau IV’ dealt with all communications in code. The years that followed Hitler’s coming to power were characterized by growing assertiveness and steady rearmament. It was highly important for the German Government to know the reactions of its neighbours. Bureau IV therefore concentrated on decrypting diplomatic messages. Its staff of nearly 250 were fairly successful in this, breaking about 70 per cent of the codes used by neighbouring powers, and using this knowledge to tap the phones of embassies, individual diplomats and foreign journalists. Their task was facilitated by Germany’s central position in Europe, for many important international telephone lines ran through the country and could be intercepted. During the crisis over Czechoslovakia in 1938, this source gave Hitler invaluable knowledge regarding the extent to which France and Britain were willing to support Czech independence, effectively telling him in advance that he would be allowed to have his way in the negotiations. The following year, Poland was in a similarly vulnerable situation when German demands were discussed between Warsaw and the West. Though the attitude of Britain and France was now more inflexible, Hitler decided to risk invasion.
Like so much else in the Third Reich, intelligence gathering was not carried out by a single, centralized body. While the Forschungsamt worked for the Luftwaffe at decrypting diplomatic signals, the Foreign Office naturally saw itself as more entitled to know what was going on in this field, and maintained its own cryptographic unit to carry out the same type of work. There was also the governmental organization for espionage and intelligence, the Reich Security Service (RSA), and the Abwehr, a military secret service headed by the naval officer Admiral Wilhelm Canaris. Though largely efficient in themselves (army signals intelligence units were highly effective and made an important contribution to the success of the blitzkrieg campaigns) these organizations suffered from a mutual jealousy and distrust that precluded any pooling of knowledge or coordinated efforts. There had initially been some degree of cooperation, but by 1939 the various bodies had retreated into an atmosphere of petulant reciprocal dislike. This meant, for instance, that the navy and air force refused to share information about Allied radar, and that each service had to repeat the labour, and the mistakes, of the other. Under these conditions, while there would be some major successes, Germany’s overall effort was doomed from the beginning.
In terms of cryptography, the RSA did manage to achieve a small victory. One of its higher officials discovered that Andreas Figl, the legendary Austrian codebreaker, had been arrested following Hitler’s takeover of his country in 1938. The official had Figl released, brought to Berlin and assigned to the Schutzstaffel (the SS) as a cryptography instructor.
All of these organizations, like any concerned with military, diplomatic or political security, used codes and ciphers to conceal the contents of their signals. The practice of cryptography (the word means literally ‘secret writing’) is very long-established: Julius Caesar, for example, developed a basic cipher that is still in use. The difference between codes and ciphers was defined in a wartime training manual, as follows:
A code is a method of concealing a message in such a way as to make it appear innocent.
A cipher is a method of converting a message into symbols which do not appear innocent, and which have no meaning to a person not possessing the key.
In a code, words or numbers are substituted for plain language – for example, ‘blue’ could mean north, ‘12’ could mean an advancing army, ‘rainbow’ could mean ‘this Friday’ – and no one can know what these represent unless they have been let into the secret, or have gained access to the necessary code books. Ciphers, on the other hand, involve the repositioning of letters as a means of hiding a message, and the systems by which this is done can be worked out by various means. Indeed for those with the requisite ability, to pit themselves against a difficult one can be an enjoyable challenge.
The simplest cipher, such as Caesar’s, substitutes one letter for another by simply moving it one place in the alphabet, so that P becomes Q, and so on. This is called a monoalphabetic cipher. A refinement, developed many centuries later, is the polyalphabetic version. This complicates the cipher considerably by using not one alphabet but 26, in each of which the letters are moved one place to the left and those that have been displaced begin again. For instance, the first alphabet runs from A to Z; the second is B to Z + A, the third is C to Z + A + B, and so on until the last alphabet runs from Z to Y. To add to the perplexity of anyone trying to make sense of it, the letters are split into ‘groups’ of five or six. This means that no words can be identified by length or shape. To muddy the waters still further, a message may be split in half and the second part of it sent first.
Cryptography is a highly mathematical art. Those with a talent for maths and a feeling for statistics are well equipped to work out its mysteries. In addition, skill in playing chess or doing crosswords (and it is often the same people who have all these abilities) makes for a good cryptanalyst, and explains why mathematicians were in such demand by the GC&CS once World War II began. Gordon Welchman, one of the Bletchley academics, admitted:
For my part, I quite shamelessly recruited friends and former students. Stuart Milner-Barry had been in my year at Trinity College, Cambridge, studying classics while I studied mathematics. He was not enjoying being a stockbroker, and was persuaded to join me at Bletchley Park. Stuart in turn recruited his friend Hugh Alexander, who had been a mathematician at King’s College, Cambridge, and was then Director of Research in the John Lewis Partnership, a large group of department stores. They brought us unusual distinction in chess: Alexander was the British Chess Champion, while Milner-Barry had often played for England and was chess correspondent for the Times.
To help them break a cipher, cryptanalysts look for the repetition of letters. Every language contains particular letters that are more common than others, and studying these frequency patterns is the beginning of understanding. There are also configurations of letters that occur often, as well as combinations that are extremely rare. When dealing with a particular language, ‘frequency tables’ compiled by language scholars will indicate what these are. Those attempting to read enciphered military messages know that there are types of phrase, known as ‘probable words’, that are highly likely to be used: greetings, ways of reporting, or of referring to units, or of signing off. The carelessness of operators who neglect standard security procedures, or otherwise take shortcuts in their haste to deliver a message, enable a listening enemy to work out code words. Furthermore, vital words can be worked out where the contents of a signal can be guessed at – for example, if the enemy is short of men and ammunition it can be assumed that they will be sending requests for them. In both World Wars, Allied analysts found that rigidly structured and inflexible German military thinking made it easier to guess the content of messages.
In the 19th century a simple ‘machine’ for enciphering was invented by Commandant Bazeries, a French army officer. The device he created was easily carried and assembled. It consisted of a wooden frame over which were fitted a series of discs (he used 20), so that they formed a cylinder. Each disc was numbered on its side, and its rim was inscribed with 26 letters. They were arranged on the frame in a particular order chosen by the sender, and the result looked much like a rolling pin covered in lines of letters. The message was spelled out and then the discs were spun until the message was concealed. The receiver would also rotate the discs until he found the message, identifiable as being the only set of letters that spelled out actual words.
This was the basis for the later cipher machines. Those used in offices became larger, faster and more complex. From the 1890s, when the typewriter was invented, they began to be equipped with keyboards. Ten years later, when electricity had become a common power source for governments and businesses, it enabled them to perform speedy and complicated tasks and thus to render ciphers more difficult to break.
The cylinder with its set of lettered discs remained the basis of cipher machines, but the rotors could now move automatically and at varying speeds. The ‘settings’ were changed at regular intervals to ensure that outsiders who broke the cipher would not be able to read messages for long. They might be altered once a month, once a week, or once a day. Those who communicated using the cipher needed to know in advance what the settings would be for the coming days and weeks. This information, usually given in code books or other secret documents and closely guarded, became in wartime an important target of espionage or military operations. Such was the value of capturing these documents that considerable resources would be expended on making the attempt – and casualties suffered in the course of these actions were deemed worthwhile.
Another important innovation was the ‘one-time pad’. This was invented during World War I by an American army officer, Major Joseph Mauborgne. The principle was that numbers were given to all the letters of the alphabet, which were then put into a random sequence, to which the letters were added. When the message was sent, its recipient also had this random sequence of numbers and deciphered the words by simple subtraction. Once the process had been completed, the operators at both ends disposed of the sequence so that it could never be repeated. This system was extremely secure. Cryptographers were extensively trained in its use throughout World War II, and it was implemented for sending messages of particular importance or sensitivity. It took so long to operate, however, that it was simply not practical for the vast amount of signals traffic generated by an average unit, vessel, office or headquarters. Group Captain Winterbotham explained in more detail how it worked – and admitted its drawbacks:
In order to make the message secret, additional groups of figures known only to the sender and receiver must be added so as to make the final groups in the signal untranslatable by any of the party.
The safest way to do this is for both the sender and receiver to have a sort of tear-off writing pad, on each sheet of which are four columns of four digits printed absolutely at random.
The sender indicates the page, the column and the line where the message is to start in the first group of the signal, thus 1348 would mean page 13, column 4, line 8. Now if the next three groups on the pad are 4431, 7628 and 5016 and these are added to the ones already quoted, we find that the message reads 1348, 9904, 8470, 9609 which means ‘To the Commanding Officer, the Division will move on Monday.’
Once used the whole page of the pad is torn off and destroyed. This is known as the one-time pad system and was at that time the only known absolutely safe cipher. If, for instance, the cipher groups are in a non-destructible book form and are used over and over again, in time an enemy will work out where the groups occur in the book and be able to read the signals. This unfortunately occurred in our own naval ciphers during the war.
The one-time pad method is, however, a long and very cumbersome method to use on any very large scale. All the printing presses in Germany could hardly have coped with the numbers of tear-off pads required. It was therefore likely that Germany would turn to a mechanical system which could be quick and easy to operate, a system of so changing the letters of the words in the signal by progressive proliferation that only the receiver who knew the key to the system could set his own machine to unscramble the letters back to their original meaning.
The Enigma was to become the world’s best-known cipher machine. It was invented in Germany in 1918 but did not initially appeal to any of the usual governmental or military bodies. It was intended for banks, as a means of keeping secure the details of monetary transactions, and there was initially nothing secret about its existence or its use. It is sometimes suggested that in this role the machine was not successful and that it passed out of use before being rediscovered by the army. In fact, Enigma machines remained important in the financial professions, and not just in Germany. Years later, once World War II had begun and the Bletchley codebreakers were at work, one of them, IJ Good, found that an acquaintance of his was perfectly familiar with Enigma:
Although I did not know it at first, the original Enigma was an unclassified machine (for enciphering plain language). It had been used by banks. Curiously enough I first learned this when I was billeted close to Bletchley. There was a retired banker living in the hotel, and once he startled me by describing the Enigma which he had used in his bank. I probably said ‘fascinating’ and raised one eyebrow. Of course, I told him nothing of the work I was doing at the office!
Enigma machines could actually be purchased in the 1920s and Dillwyn (‘Dilly’) Knox, another Cambridge academic who was later to be one of Bletchley’s luminaries, bought one in Vienna. It was examined by the Government Code and Cipher School, in the shape of Hugh Foss (who was also to play a major role at ‘BP’), but it was decided that Enigma was not especially difficult to break, and the system was therefore not considered worth using. It was adopted, however, by the Germans, whose navy acquired it in 1926, followed by the army in 1928. The military authorities in Berlin also deemed it too easy to decipher, so the machines went through a series of modifications over the next few years – the last of them in 1937, after which the Germans considered it as secure as a bank vault. Electromagnetically operated, the machine was designed for enciphering radio messages, and it threw out seemingly random groups of numbers; it had the considerable advantage that, if one of these were intercepted, it could not be decrypted unless the interceptor had an identical machine. Germany’s opponents knew this as a result of information from a German civil servant, Hans Thilo Schmidt, and it became their primary objective either to steal an adapted military Enigma machine or to build an exact copy of it.
In size and shape the Enigma machine resembled a typewriter and was housed in an unprepossessing, square wooden box. It had a typewriter keyboard but above this, on another board, was a set of lights corresponding with the letters of the alphabet. Inside there were three wheels, each of which had the alphabet marked on it twice – on the inside and on the outside, and each outside letter was electronically linked to an inside one. Pressing a key lit up a bulb on the board as that letter was enciphered, and turned the first of the wheels by one position. After a set number of turns of the first wheel, the second would move by one position, then the third. IJ Good described how the basic Enigma worked:
The main ingredient of the Enigma machine is a so-called rotor or hebern wheel. It is capable of rotating and it is wired so that the input alphabet is permuted [arranged] to give an output alphabet. The original Enigma machine had three rotors in it together with a reflector so that the plain language letter would come through three rotors, and then get reflected and come back through the same three rotors by a different route. Thus, for any fixed position of the rotors the original input alphabet would go through a succession of seven simple substitutions. Hence no letter could be enciphered as itself. But the whole effect of the machine was not merely a simple substitution since the wheels stopped in a certain way each time a letter was enciphered.
The German military versions were more complicated, for they used different internal wiring, plus the addition of a ‘steckerboard’. This, like the rest of the machine, was ugly and unimpressive to look at. It was a plugboard with holes into which an operator could put jack plugs to connect yet more pairs of letters. This provided a further layer of encipherment, increasing the range of possible settings from merely several million to nearly 160 trillion. Once this level of security had been attained, a certain complacency set in among German military and governmental circles.
Work on deciphering Enigma began in 1929, when Poland, a country that had good reason to be concerned with German military plans, assigned cryptanalysts to work in on it in earnest. A machine had arrived in Warsaw at that time, sent by accident from Germany and held by Polish Customs. The German Embassy urgently requested that it be returned whence it came, and the Polish authorities complied – after secretly opening the box and having experts examine the contents. Polish Intelligence later made attempts to read Enigma messages, but failed because they had seen only the civilian and not the military machine.
Poland was, during the interwar period, a close ally of France, and it was through the Deuxième Bureau, the French secret service, that the next step forward was made. In 1931 it was approached by German civil servant Hans Thilo Schmidt with an offer to sell some important documents. Schmidt was an official in the Cipher Office of the Defence Ministry in Berlin and was motivated by a simple desire to make money. He sold the French two documents which, it transpired, related to Enigma. They showed how to encipher, though not how to decipher. Nevertheless, they offered a beginning. Photographs of the documents, shown to the Polish experts, confirmed that the machine they had seen was in military use. Further settings were provided after subsequent meetings with Schmidt, but the best way of understanding the machine was to build a prototype. Marian Rejewski, one of the ablest of the engineers working on the project, was eventually able to do this after receiving the settings for September and October 1932. Having discovered the layout of the wiring that controlled two of the wheels, he was able to calculate the workings of the third.
Once they were listening to Enigma, they made a welcome discovery. German operators of the military version were instructed that, when sending signals, they must twice encipher the ‘message setting’ (the position of the wheels when the first letter was enciphered) and then send it. This made it easy for Polish Intelligence to break the code on a continuing basis. Rejewski and his colleagues were also able to arrange a set of six Enigmas that operated in concert, wired together. A group of this sort became known as a Bomba. With one for each of the six existing wheel orders, they could often work fast enough to discover the day’s settings within about a hundred minutes.
For the remainder of the 1930s, and throughout the war years, the German operators of Enigma and their enemies engaged in a battle of wits. As methods were devised for making the machines more secure, the Polish codebreakers found new ways of breaking them. The Germans had originally altered their wheel order every three months, but from 1936 this was done monthly and then daily. The Bomba was rendered considerably less effective when more holes were introduced to the plugboards on Enigma, allowing far more jack plugs to be used and thus increasing the number of possible settings. A subsequent introduction of two further wheels expanded its capabilities yet further. A significant victory for the Poles, however, was their development of ‘perforated sheets’.
Rejewski’s colleague Henryk Zygalski had noticed that the indicators (the six-letter groups tapped out twice by the message sender) had some similarities. The groups had certain letters continually in common: the first and fourth, second and fifth or third and sixth. These were dubbed ‘females’. Since they could not be produced by the German senders, each one represented a setting (and altogether these were to account for nearly 40 per cent of the total) that could not be used and could therefore be eliminated from calculations.
Holes were punched in cards in places where a ‘female’ was expected to occur. When a whole series of cards had been treated in this way, they were piled on a glass table that was lit from below. Wherever a space showed through the entire stack, it indicated the wheel positions and wheel orders for that day. They would then be manually tested for confirmation. This practice was effective but laborious and therefore very slow, and it fell seriously behind when the Germans began using more wheels.
Most of this work was concealed from their allies by Polish Intelligence, and only in 1939 did they admit to the French and British that Enigma traffic had been read so extensively and for so long. The news caused a good deal of annoyance, not least because Schmidt, the German traitor, had gone on providing Enigma settings over the years at some risk to himself and to those dealing with him. Nevertheless, Polish expertise with the machine was invaluable, and both Paris and London were keen to add as quickly as possible to their own knowledge. ‘Dilly’ Knox went to Warsaw in the summer and obtained details of the machine’s wiring.
Another source had become available through Poland and this was to be of considerable significance. Group Captain Winterbotham, who would be in charge of disseminating ‘Ultra’ information during the coming war, recalled the chain of events that brought further secrets to the West:
In 1938 a young Polish mechanic had been employed in a factory in eastern Germany which was making what he rightly judged to be some sort of secret signalling machine. As a Pole, he was not very fond of the Germans anyway and, being an intelligent observer, he took careful note of the various parts that he and his fellow workmen were making. I expect it was after one of the security checks which were made by the Gestapo on all high security factories that they discovered his nationality. He was sacked and sent back to Poland. His keen observation had done him some good, and he got in touch with our man in Warsaw.
In due course the young Pole was persuaded to leave Warsaw and was secretly smuggled out under a false passport with the help of the British Secret Service; he was then installed in Paris, where he was given a workshop. With the help of a carpenter, he began to make a wooden mock-up of the machine he had been working on in Germany.
Because Enigma had been available for two decades, it was quickly recognized by specialists:
There had been a number of cipher machines invented over the years and our own backroom boys had records of most of them. It didn’t take them long to identify the mock-up as some sort of improved mechanical cipher machine called Enigma. The name Enigma had been given to the machine by the German manufacturers. The Pole had been told not to attempt to make his wooden model to scale. In fact, the bigger the better, because he could then more easily incorporate any details he could remember.
The result was rather like the top half of an upright piano, but it was big enough to tell us that it would be essential to get hold of an actual machine if we were to stand any chance of trying to break into its method of operation. We set about working out a scheme with our friends in Poland. We knew where the factory was and all about its security methods, and there were still some Poles working there under German names. However, the Polish Secret service thought the scheme might well stand a better chance of success if we gave them the money and they did the job. They knew the terrain and the people much better than we did, so we gladly agreed. It was [Commander Alistair] Denniston himself who went to Poland and, triumphantly, but in the utmost secrecy, brought back the complete, new, electronically operated Enigma cipher machine which we knew was being produced in its thousands and was destined to carry all the signal traffic of the great war machine.
Sir John Slessor, Marshal of the Royal Air Force, was later to comment on the immeasurable benefits of this act of larceny:
There was nothing very remarkable about the act of cracking enemy ciphers as such, but with the secret abstraction from Poland of the theoretically unbreakable German cipher machine, which gave birth to Ultra, that art took on a completely new dimension; and surely no other act in the history of officially sponsored skulduggery ever had comparably fruitful results.
By the time Hitler’s troops invaded Poland, the French and British each had two Enigma machines. The race to break the German codes had been won, but only just in time.
In the event, Enigma was not as formidable in the eyes of Germany’s rivals as might have been expected, and by the outbreak of war in 1939 a good deal of progress had been made toward understanding it. Alan Turing, a young Cambridge student of mathematics and the most powerful of the great minds that set about understanding Enigma, wrote a treatise on the machine. He described the early work in creating methods for decipherment:
The Poles found the keys for the 8th of May 1937, and as they found that the wheel order and the turnovers were the same as for the end of April they rightly assumed that the wheel order and Ringstellung [ring position] had remained the same during the end of April and the beginning of May. This made it easier for them to find the keys for other days at the beginning of May and they actually found the Stecker [plug] for the 2nd, 3rd, 4th, 5th and 8th, and read about 100 messages. The indicators and window positions of four messages for the 5th were:
| Indicator | Window start | |
| K F J X | E W T W | P C V |
| S Y L G | E W T W | B Z V |
| J M H O | U V Q G | M E M |
| J M F E | F E V C | M Y K |
The repetition of the EW combined with the repetition of V suggests that the fifth and sixth letters describe the third letter of the window position, and similarly one is led to believe that the first two letters of the indicator represent the second. Presumably this effect is somehow produced by means of a table of bigramme equivalents of letters, but it cannot be done simply by replacing the letters of the window position with one of their bigramme equivalents, and then putting in a dummy bigramme, for in this case the window position corresponding to JMFE FEVC would have to be say MYY instead of MYK. Probably some encipherment is involved somewhere. The two most natural alternatives are i) The letters of the window position are replaced by some bigramme equivalents and then the whole enciphered at some ‘Grundstellung’ [initial position], or ii) The window position is enciphered at the Grundstellung, and the resulting letters replaced by bigramme equivalents. The second of these alternatives was made far more probable by the following indicators occurring on the 2nd May:
| E X D P | I V J O | V C P |
| X X E X | J L W A | N U M |
| R C X X | J L W A | N U M |
With this second alternative we can deduce from the first two indicators that the bigrammes EX and XX have the same value, and this is confirmed from the second and third, where XX and EX occur in the second position instead of the first.
It so happened that the change of indicating system had not been very well made, and a certain torpedo boat, with the call-sign AFA, had not yet been provided with the bigramme tables. This boat sent a message in another cipher explaining this on the 1st May, and it was arranged that traffic with AFA was to take place according to the old system until May 4, when the bigramme tables would be supplied. Sufficient traffic passed on May 2, 3 to and from AFA for the Grunstellung used to be found, the Stecker having already been found from the … messages. It was natural to assume that the Grundstellung used by AFA was the Grundstellung to be used with the correct method of indication, and as soon as we noticed the two indicators mentioned above we tried this out and found it to be the case.
There actually turned out to be some more complications. There were two Grunstellungen at least instead of one. One of them was called the Allgemeine [general] and the other the Offiziere [officer] Grunstellung. This made it extremely difficult to find either Grundstellung. The Poles pointed out another possibility, viz. that the trigrammes were still probably not chosen at random. They suggested that probably the window positions enciphered at the Grunstellung, rather than the window positions themselves, were taken off the restricted list.
In Nov. 1939 a prisoner told us that the … digits of the numbers were [now] spelt out in full [by the German navy]. When we heard this we examined the messages toward the end of 1937 which we expected to be continuations and wrote the expected beginnings under them. The proportion of ‘crashes’, i.e. of letters apparently left unaltered by encipherment, then shows how nearly correct our guesses were. Assuming that the change mentioned by the prisoner had already taken place we found that about 70 per cent of these cribs must have been right.
Gordon Welchman explained how the Germans used the machine and safeguarded its security with regard to their compatriots as well as their enemies. His analysis provides a detailed picture of what Allied codebreakers were up against:
The Germans had adopted the principle that the security of their communications must rest not on the machine itself, but rather on a ‘key’ that would determine how the machine was to be set up for a particular purpose. Moreover, they were concerned with both external and internal security. They wanted to prevent their enemies from reading their messages and also to prevent their own units from reading messages that were not intended for them. For example, three of the many different types of Enigma traffic were messages between operational units of the regular army and air force; messages between units of Hitler’s private army, the SS or Schutzstaffel; and messages involved in the training exercises of new signals battalions. All three kinds of messages were enciphered on identical Enigma machines, but the regular army and air force units were not to be allowed to read the text of SS messages. Nor were the trainees to be permitted to read the texts of the other two types of traffic. Consequently, different keys were issued for different types of traffic. This, however, did not quite solve the problem, because messages of different types were often transmitted on the same radio net. It was therefore necessary to provide means by which a receiving unit’s operator would know what type of clear text was hidden behind the enciphered text, and whether he had the necessary key to read it.
The Germans chose to solve this problem by using a three-letter ‘discriminant’ transmitted in the unenciphered message preamble. This discriminant was not part of a key. Its purpose was simply to indicate which of many keys was being used. A cipher clerk would examine the discriminant of each incoming message to determine whether he had been issued the key used for its text encipherment. If he did have the key, he could set up his Enigma and decode the message. If not, he couldn’t.
The particular needs of the Wehrmacht (the Nazi armed forces) led them to develop a sophisticated form of communications network:
The Germans, in their concept of a blitzkrieg, reckoned that many groups of fast-moving fighting, command, support, and staff elements would need effective communication among themselves wherever they might be, and furthermore that the activities of these groups would have to be tied in the higher command system. The elements of each cooperating group were to be served by signals detachments operating a ‘radio net’ on an assigned radio frequency. Under ideal conditions any message transmitted by any radio station operating in the net on the assigned frequency would be heard by all the other stations. One station of the net would act as control, to ensure that no two stations would cause interference by transmitting at the same time. There were to be many such nets, and a station could operate in two or more nets, so that messages originating at any point could be relayed to any other point.
The call signs were simply the means of identifying the individual elements that were participating in this overall radio communication system. When passages were passing between elements within a single radio net, the preamble would contain the call signs of the originator and intended recipient(s) of each message. When a message was to be forwarded to other elements, their identifying call signs would also be included in the message preamble. Thus, by studying call signs, we had an opportunity to learn something about the structure of the enemy’s forces. As the call signs were changed every day, however, the detective work had to begin anew every twenty-four hours.
Our intercept operators listened to the Enigma messages and their preambles, writing them out by hand on standard message forms. The main part of the form was used for the succession of five-letter groups, or ‘words,’ which constituted the indicator and text of a message enciphered on an Enigma machine. At the top of the form was a space in which the intercept operator entered the preamble that the German radio operator had transmitted ahead of the message. Indeed the form used by our intercept operators must have been very similar to the form used by the German cipher clerks. The intercept operator, however, also entered the radio frequency on which he was listening and the time of intercept.
The beginning of the codebreakers’ war is seen as the dawn of the computer age. The mathematical endeavour needed to keep pace with the increasing complexity of Axis cryptographic communications gave an enormous spur to the development of what would later be called computer science, though its story was by that time over a century old.
The computer was an outgrowth of the calculating machine, and that boasted a long and illustrious pedigree. The word ‘calculate’ stems from the Greek for pebble, and for most of history the task of adding figures was assisted by such basic objects. Leonardo da Vinci conceived the notion of a machine that would carry out calculations, though he simply described the scheme in his notebooks and it was never built. Worked by a handle, it was to consist of 13 wheels, each cogged and able to turn individually. In 1642 a similar machine was, quite independently, devised in France by the 19-year-old Blaise Pascal, to assist his father, who was a tax-collector. This would have swiftly performed addition and subtraction and saved a good deal of labour, had it been possible to build it. Like da Vinci’s concept, however, it was beyond the reach of contemporary technology. Just over 50 years later a machine designed by the German Gottfried Leibniz established basic principles that would remain unchanged until the 20th century.
The development of the computer itself began in 1821, when Charles Babbage, a mathematics professor at Cambridge, wondered if he could avoid repeating tedious calculations by creating a machine to perform the task. He set out to take Leibniz’s ideas further by building a bigger and better cog-wheeled structure. The result, which he called a ‘Difference Engine’, was in fact a series of calculating machines working in unison. These were able to perform a number of mathematical tasks, and produced calculations that were often more accurate than those resulting from unassisted brainpower. Having a degree of engineering skill, Babbage was able to work on the parts himself, and he received funding from the Royal Society. He moved on to a more advanced concept: the Analytical Engine. This had a much wider range of functions, including basic versions of all the elements of a present-day computer: a memory, a control unit, and input and output mechanisms. He had the notion of a machine’s operating instructions as something that could be manipulated, and thus conceived the ‘computer programme’. He made use of punched cards to input data and give instructions to the machine, which would then complete the work without further human involvement – a concept borrowed from the French textile manufacturer Joseph Jacquard, who used such a system (he called them ‘operation cards’) to control the patterns in his weaving looms.
Contemporary with Babbage was another brilliant theorist: Ada, Countess Lovelace. She studied his work, adding to and improving the concepts of programming with cards and of solving mathematical problems by having the machine deconstruct them. Though all this proved, once again, beyond the capabilities of technology, it established the concept that in future would define what a ‘computer’, as opposed to a calculator, actually was. The latter could perform tasks only if constantly attended by an operator. The former could make its own decisions without supervision.
In spite of these discoveries, the development of the computer went no further in the lifetimes of these pioneers. Neither Babbage nor Ada Lovelace was taken seriously by contemporaries, because the equipment they designed or described could not be built. Only near the end of the century, in 1890, did an American statistician pick up the torch. Herman Hollerith was employed by the US Census Bureau, and he invented a tabulating machine that used Jacquard’s card system to automate the analysis of census returns. It was not the beginning of the computer age, but neither was it, this time, a false dawn; Hollerith founded a company that, in the next century, would become one of the giants of the communications industry: International Business Machines, now universally known as IBM.
With a new age came the power of electricity and the revolution in communications represented by wireless, and these were ultimately to open new possibilities, as technology at last caught up with the dreams and schemes of inventors. As RB Davison, a writer on computer science, put it:
Up to this point all calculations had been done mechanically, by means of cogged wheels turning numbers in registers. When electricity was used it was simply as a motive power, to drive the wheels more quickly and for longer periods. But with the development of radio and the new science of electronics, a new idea began to emerge – instead of counting by moving cogged wheels, it should be done by electronic impulses.
In 1936 an electronic computer was patented, but not produced. The first to come into use, for performing mathematical calculations, was therefore a model built at the Bell Telephone Laboratories in New York by Dr GR Sterbity in 1939. Parallel with this, and working in complete isolation from other scientists, a young German called Konrad Zuse built another electronic computer at his parents’ home in Berlin.
It would be the young Alan Turing, spurred on by the demands of the codebreakers at Bletchley Park, who made the next significant contribution. His wartime colleague, IJ Good, described the moment in which Turing’s eyes were opened and his sense of vocation began:
Turing had gone to King’s [College, Cambridge] as a mathematical scholar at the age of 19, and was elected to a Fellowship at 23. M.H.A. Newman had been a university lecturer at Cambridge, and it is believed Turing’s work was sparked off by one of Newman’s lectures. This was a lecture in which Newman discussed Hilbert’s [David Hilbert, the German geometrician] view that any mathematical problem can be solved by a fixed and definite process. Turing seized on Newman’s phraseology, ‘a purely mechanical process,’ and interpreted it as something that could be done by an automatic machine. He introduced a simple abstract machine in order to prove that Hilbert was wrong, and in fact showed that there is a ‘universal automaton’ that can perform any calculation that any automaton can, if first provided with the appropriate instructions to input.
Turing was thus the first to arrive at an understanding of the universal nature of a (conceptual) digital computer that matches and indeed surpasses the philosophic understanding that I believe Babbage had attained, a century earlier, of the universality of the planned (mechanical) Analytical Engine. Central to the Universal Turing Machine is the idea of having data, and input data in particular, represent a programme (called a ‘table’ in Turing’s paper).
Turing carried out further research at the Institute of Advanced Studies at Princeton. Here he was a colleague not only of Albert Einstein but of Alonzo Church – who was doing similar work – and of John von Neumann, who was also his supervisor and whose ideas, running parallel with his, were to create EDVAC (Electronic Discrete Variable Analytic Computer), in June 1945, the first modern computer. Turing was, in other words, in precisely the right place for working with the other major pioneers at a time when their own most important work was being done.
It was while at Princeton that Turing proved his concept was mathematically valid. As computer historian Neil Barrett wrote:
In a universal Turing machine, we see a mathematical expression of an idealized device that can be programmed with instructions and data to solve any problem that can be solved. The universal Turing machine is the theoretical expression of a ‘stored programme, general-purpose computer’, the computational model missing from Babbage and Lady Lovelace’s earlier work.
Even more important, in view of coming events, was the fact that Turing’s machine could be used for analysing ciphers. By this time it was 1938. War was averted in that year, but was a real possibility in the near future. Turing felt he should return to England, and he was immediately attached to the Foreign Office’s Department of Communications as a temporary civil servant. He was, in fact, a member of the Government Code and Cipher School.
The technology necessary for unravelling the mysteries of the formidable Enigma ciphers had not yet been perfected, but the development of the computer had reached a stage at which it could make a unique and effective contribution to the most important event of the century. It was also to initiate, unnoticed by all but the few who had knowledge of it, a new era. One of the Bletchley machines, nicknamed Colossus, was the first-ever programmable electronic computer.
War is often the midwife of scientific progress, and a new era in communications was brought about by the urgent need to know the enemy’s thoughts in what was the most scientific conflict yet fought. With foreknowledge of its intentions, it was possible to deploy limited resources to best advantage, to take its forces by surprise or to cut off and destroy its sources of supply. In several instances this information directly affected the outcome of major campaigns. It therefore saved many thousands of lives and hastened the defeat of the Axis. It has been estimated that the work of the codebreakers shortened World War II by between two and three years.