Within a week, it had surrendered. Churchill, who was in Washington conferring with Roosevelt, was bitterly disappointed. ‘I did not attempt to hide from the President the shock I received,’ he would later recall. ‘It was a bitter moment. Defeat is one thing; disgrace is another.’
But the tide was about to turn, and as a direct result of a dramatic improvement in the work of the codebreakers at Bletchley Park which included much greater coordination between the codebreaking process and traffic analysis.
During 1940, three teams of officers and mostly female civilians began working on trying to analyse the radio networks and in particular break the German system of daily changing callsigns. They were based at MI8 offices in Caxton St in London, at Harpenden and in Hut 3. This process, subsequently known as ‘log-reading’ and then later ‘traffic analysis’, had produced a wealth of usable intelligence during the First World War and had already proved its worth in Hut 4, the Naval Section, where Harry Hinsley was obtaining so much important intelligence. It was not initially regarded by many of the new young codebreakers and intelligence officers working on German Army or Luftwaffe traffic in Hut 6 and Hut 3 as likely to produce much intelligence. The reverse soon proved to be the case. Using a variety of information noted down by the ‘Y Service’ operators on their message logs, including callsigns, locations from direction-finding or identifications from RFP or Tina, and the simple German operator plain language chatter, which when units came under pressure during the battle often contained violations of basic communications security, they were able to build up complete pictures of the units involved and provide fresh intelligence and cribs which would help in the breaking of Enigma. A new unit, VI Intelligence School, with an establishment of 233 staff, a mix of male members of the Army Intelligence Corps and female ATS soldiers, was set up in March 1941 and based initially at Beaumanor. Even at this stage it was suggested by some in Hut 6, now converted entirely to the idea that ‘traffic analysis’ was a producer of vital intelligence, that it should be based at Bletchley. It moved there in May of 1942 and was attached to Hut 6, initially as the ‘Central Party’ but later as Sixta, a title derived from Six Traffic Analysis.
‘Our job was to analyse the operator chat,’ recalled Jimmy Thirsk, a 27-year-old librarian recruited into the Intelligence Corps to work as a log-reader.
If you had a section of the Luftwaffe, say they were in France with their headquarters in Dijon, and perhaps they had ten outstations round in that area. Every morning they would start up. Each station had a three-letter callsign and they would change that at midnight according to a pre-set pattern and then they would call up the outstations just to make contact. Just to make sure they were all awake and working and they would chat to each other in clear German. This operator chat was going on all day long and the intercept operators logged it all down.
Most of the messages were teleprinted from the intercept stations but the logs used to be brought mainly by motor cycle dispatch rider. You would be allocated a number of nets. If they were small ones you might have two or three but you might have just one. Each day you would plot the radio net. We had coloured pencils, and you made a circle with a dot in the middle as the HQ and then the outlying stations were round the circumference of the circle with a line drawn from each of the outstations to the centre and you would note the number of messages passing on each link. Then at the end of the week you had the dreaded weekly report and you had to go through your stuff for the week and compile a report. It was pretty dreary stuff at times.
Their work was assisted by the capture of the Luftwaffe callsign book, the Rufzeichentafel Ausgabe B, from a German Army unit in the North African desert in December 1941. The Luftwaffe was clearly unaware of its capture and continued using the same system for the next two-and-a-half years.
Longer-term traffic analysis was carried out in the Hut 3 Fusion Room which used all the information produced by various parts of Bletchley Park to build up a complete picture of the enemy radio networks. Joyce Robinson had a degree in German and after a brief spell in the Civil Service joined the ATS and was posted to Bletchley where she was allocated to ‘fusion’.
It was really a sort of consolidation of information from a lot of quarters after material had been dealt with operationally. It was departmentalised according to networks, or keys, where a group of one, two or more people considered the behaviour of certain things so that you knew your network. You were sometimes able to help Hut 6 when they had difficulties with decoding, with a change in the wheels.
One of her ATS colleagues was Jean Faraday Davies, who was pulled off a German course at the University of London and sent to Bletchley. The Fusion Room coordinated information produced by the log-readers with the intelligence from the decyphered messages, Faraday Davies said. ‘Our function was to take these two sources and feed it out in two directions to enable interception to go on or to help decoding.’
The move of the traffic analysts to work alongside Hut 6 had increased the availability of cribs, together with an increased understanding of how the Army Enigmas worked. Up until now the delay in breaking Chaffinch, the main Afrika Korps Enigma, had been up to a week. From the end of May, they were able to break it daily, albeit with some delay. Another Panzer Army Enigma, designated Phoenix and broken briefly at the end of 1941 following the capture of three machines and a number of keys, was read continuously from 1 June. A third Army Enigma, Thrush, giving details of air supplies, was also broken.
Hut 6 was now able to read all the Luftwaffe keys, including the Red which had been continuously broken since May 1940; two relatively minor keys called Locust and Gadfly; and two much more important keys which, in a security blunder by the Germans, were closely linked. Primrose, the cypher of the Luftgau Afrika, the air formation responsible for administration and supply of the Luftwaffe forces in North Africa, was not only an important source in its own right, its keys were also used later by Scorpion, the cypher used for communications between the Flivos and the ground forces.
During the North Africa campaign, Primrose became as high a priority for Hut 6 as the Red cypher. ‘It was very important,’ said Susan Wenham, one of the Hut 6 codebreakers.
Every day the key was changed and about 3 o’clock in the morning Primrose used to send a tuning message through, a very short little message. It was always on the same wavelength and it was always recognisable. So people on the nightshift would watch out for this message and, when they got it, they would tinker about with it and we could quite often break on that message. Red had an enormous quantity but it didn’t have a nice convenient tuning message that you could find.
Hut 6 was given much more time to get into the Army keys as a result of an extraordinary error by the Luftwaffe, Stuart Milner-Barry recalled, describing 1942 as Hut 6’s ‘annus mirabilis’. Paradoxically, it was due to an increase in the number of different Luftwaffe Enigma keys introduced with a view to making the systems more difficult to crack. It actually achieved the exact opposite, Milner-Barry said.
The Germans suddenly realised that there was no objection, and obviously great advantage in security, in using a large number of different keys for the different major units of the Luftwaffe. However, with characteristic blindness the enemy undid much of the good that this step might have done him, for instead of making up entirely separate keys he rehashed old ones on a delightfully simple plan. The effect of this was that, every other month, the majority of Air Force keys were in our hands for the decoding and a tremendous boom ensued which taxed our resources to the utmost.
Since Scorpion’s keys were now predictable, Hut 6 decided that it could be decyphered in Heliopolis, making it available to the commanders far quicker than any other material. Because the Flivos needed to keep in close contact with the battle in order to coordinate air attacks with the movement on the ground, Scorpion provided more details of the fighting, the troop positions and air activity than any of the other cyphers had before.
‘It contained much �
�hot” operational news; it was easy to break, for the daily settings could be predicted in advance,’ said Bennett. ‘It was decided to radio them in advance from Hut 6 to Cairo and to send an experienced officer out to compose signals on the spot.’
Meanwhile, during July and August, the number of mobile Y units in North Africa was doubled and they became much better integrated into the command structure. Traffic analysis and direction-finding were improved and the Intelligence Corps and RAF codebreakers attached to the Special Wireless Sections expanded their exploitation of enemy tactical codes and cyphers.
Ultra was now totally in the ascendant. In the first nine months of the Special Signals Link to the Middle East, between March and November 1941, Hut 3 had sent just over 2,000 signals to Cairo. Between November 1941 and July 1942, it had sent five times that figure.
Desperate for a victory and fully aware of the information from Ultra that was now available, Churchill decided it was time for change. Auchinleck was a fine general but did not have the necessary killer instinct. He brought in General Harold Alexander as Commander-in-Chief Middle East and appointed General Bernard Montgomery to take command of the Eighth Army.
Within days of his arrival, Montgomery was the beneficiary of a major piece of Ultra intelligence that was to change the military’s view of the codebreakers. On 15 August, Rommel, newly promoted to Field Marshal, explained to Hitler what he planned to do next. The details of the plans had to go first through his direct commander, Field Marshal Albrecht Kesselring, the German Commander-in-Chief South, and they were transmitted using the Red cypher, which Hut 6 had no problems reading.
Two days earlier, Montgomery had outlined what he believed the Desert Fox would do. It matched the signal sent to Kesselring almost to the letter. Rommel intended to attack around the time of the full moon due towards the end of August, swinging south around the end of the British lines before striking north to come up behind the Eighth Army, cutting it off from Cairo. But to do so he would have to cross a major obstacle: the Alam Halfa ridge.
‘Monty arrives in the middle of August and is told not to go and take charge until the next morning,’ said Ralph Bennett.
He goes up to Alam Halfa to have a look round, one day before he is going to take over. He sums up the situation and realises that if Rommel is going to attack he will almost certainly do so on a route that will take him through the Alam Halfa ridge.
A few days later, Rommel tells Hitler what he is going to do, which is exactly that. We get this signal and we tell Monty. So there is Monty, the new boy, who has just made a pep talk to his troops, now knowing that his hunch as to what Rommel will do is exactly right. He can’t tell anybody about it but when Rommel attacks Monty is ready.
Two other new developments at Bletchley Park also helped Montgomery. Hut 3 had just started receiving reports decyphered from the Chaffinch cypher, giving a complete breakdown of the fighting strengths of the Afrika Korps and comprehensive returns on the availability of tanks. Meanwhile, its joint operation with Hut 4 to detect the Axis supply lines had been improved, partly by the decision to place Royal Navy advisers inside Hut 3 to track the convoys and, more importantly, by Hut 8’s breaking of Porpoise, the German Navy’s Mediterranean cypher.
Throughout the second half of August, the RAF and the Royal Navy redoubled their attacks on the Axis convoys. Meanwhile, the codebreakers were able to monitor a series of high-level exchanges between the German commanders. Those between Kesselring and Rommel showed the two were barely on speaking terms. They also revealed that the Desert Fox was unwell. Then came the approval, first by Hitler and later by Mussolini, of Rommel’s plans.
More importantly perhaps, given the Desert Fox’s predilection for ignoring orders, Bletchley and Heliopolis were able to chart the regrouping of the German forces in readiness for the attempt to outflank the Eighth Army as well as the problems and delays to the operation caused by the non-arrival of two of the supply ships. Montgomery had briefed his troops on what Rommel was about to do. Then the supply problems led to a four-day postponement.
‘Believing that the confidence of his men was the prerequisite of victory, he told them with remarkable assurance how the enemy was going to be defeated,’ said Williams, the Eighth Army commander’s chief intelligence officer.
The enemy attack was delayed and the usual jokes were made about the ‘crystal-gazers’. A day or two later everything happened according to plan. The morale emerging from the promise so positively fulfilled formed the psychological background conditioning the victory which was to follow.
After finding his way through the Alam Halfa ridge blocked, Rommel was forced to retreat for lack of fuel. From then on, Ultra played a privileged part in Montgomery’s plans. He allowed Williams access to his command post day or night with any new information the codebreakers produced.
‘Imagine the situation in the desert in the late summer of 1942,’ said Ralph Bennett, who was sent out to the Middle East to report from Cairo on the Scorpion traffic.
There is Montgomery. He’s got a little truck park with his own command truck and the Army and the Air commanders forming three sides of a little square. Then the fourth one is the wireless truck to receive the Ultra signals so that Williams can make immediate contact with Montgomery and the other commanders to give them the urgent Ultra information.
Ultra played no significant part in the Battle of el Alamein itself. But Montgomery knew from Chaffinch and Scorpion the precise numbers of troops and tanks he faced, while the sinking of the supply ships, 50,000 tons in October alone, nearly half of the cargo which left Italy for North Africa, had a crucial influence on the Afrika Korps’ ability to resist. So tight were its margins of supply that the sinking of an Axis convoy during the battle itself had a direct influence on the fighting.
On the afternoon of 2 November, with Montgomery having punched two holes in the Panzer Army’s defences and about to force his way through, Bletchley decyphered a message from Rommel to Hitler asking permission to withdraw. ‘Panzerarmee ist erschopft’, he said. The Panzer Army was ‘exhausted’ and had precious little fuel left. The response from Berlin was that Rommel should stand his ground at all costs. He was to ‘show no other road to his troops than the road leading to death or victory’.
But in the face of far superior troops, he was forced to retreat along the coast road towards el Agheila. Why he wasn’t pursued at speed and destroyed either by intensive RAF bombing raids or Montgomery himself remains a puzzle. Chaffinch revealed on 10 and 11 November that one of his Panzer divisions had just eleven tanks while the other had none at all. Five days later, with all attempts to resupply the German troops being frustrated with the aid of Ultra, the Red Enigma carried a special situation report from Rommel to Hitler in which he described his fuel supplies as ‘catastrophic’.
Bennett was by now in Cairo as the experienced Hut 3 officer who was to oversee the issuing of reports from the Luftwaffe Enigma decyphered in Cairo.
My temporary absence meant that among much else I missed the fierce indignation and dismay felt throughout the Hut at Montgomery’s painfully slow advance from Alamein to Tripoli, incomprehensible in the light of the mass of Ultra intelligence showing that throughout his retreat Rommel was too weak to withstand serious pressure.
Edward Thomas, who was the Hut 4 liaison officer in Hut 3 during this period, was a first-hand witness to the fury within Hut 3 at Montgomery’s inaction even when told, absolutely correctly, that in the aftermath of Alamein, the Desert Fox had only eleven tanks and virtually no fuel.
After the war, I found my initials at the bottom of the signals giving details of three supremely important tanker movements at the time of el Alamein. Their sinking was largely responsible for Rommel’s long and halting retreat westwards. I well remember the frustration that exploded from our Hut 3 colleagues at Montgomery’s failure to overtake and destroy him.
Their anger was clearly shared by Churchill. Following the victory at el Alamein, he had said: ‘
This is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning.’ Now he bombarded Alexander with pieces of Ultra suggesting that the Eighth Army kill the Afrika Korps off for good.
‘Presume you have read the Boniface numbers QT/7789 and QT/7903 which certainly reveal a condition of weakness and counter-order among the enemy of a very remarkable character,’ one signal from the Prime Minister stated with obvious impatience, while another pointed out: ‘Boniface shows the enemy in great anxiety and disarray.’
But Montgomery feared that Rommel’s greater mobility might allow him to turn the tables on the British yet again and decided to err on the side of caution, ignoring the repeated suggestions of Churchill, Alexander and Air Marshal Arthur Tedder, the commander of the Middle East Air Force, that the Desert Fox should be pursued and annihilated.
‘Unfortunately after Alam Halfa, Monty was inclined to be a bit boastful about having got it right,’ said Ralph Bennett.
He was inclined to think he was right all the time. At el Agheila, he insisted on being cautious, which of course was Monty’s great thing most of the time, although he knew perfectly well, because we had told him over and over and over again, that Rommel had inferior defences and very few tanks.
Within days of the victory at el Alamein, Allied forces landed in Morocco, Algeria and Tunisia as part of Operation Torch, the invasion of North Africa – designed to provide a base from which to attack Italy and southern France. British and American troops under General Eisenhower pushed east with the aim of linking up with Montgomery. Meanwhile the Axis forces began pouring troops into Tunis, a reinforcement chronicled in some detail by Hut 3 from the Luftwaffe Enigma, the Italian C38m and Porpoise, the German Navy’s Mediterranean cypher.
The Secrets of Station X Page 21