The Secrets of Station X (8 page)
Read The Secrets of Station X Online
Authors: Michael Smith
For example you might expect that a message might start â
An die Gruppe
' something or other, just an address. So you make a supposition that it started like this and you might be able to get a very little confirmation that if you wrote
An die Gruppe
something or other under the message, the one thing that the encoded message couldn't for instance have is the A of
An
as A, it could be any one of the twenty-five letters other than A and the second letter couldn't possibly be N and the sixth letter couldn't possibly be G of the word
Gruppe
so if you had quite a long thing you might have far from certain evidence but quite a feeling it might very well be right.
Cribs could appear at any point in the message. Even
Keine besondere Ereignisse
was likely to be preceded or followed by some piece of routine information. But the fact that none of the letters in the crib could ever be matched up with the same letter in the encyphered message made it much easier to find out where they fitted.
âThink of it as a sort of crossword technique of filling in what it might be,' said Mavis Lever, a member of Knox's team.
I don't want to give the impression that it was all easy. You did
have inspired guesses. But then you would also have to spend a lot of time, sometimes you would have to spend the whole night, assuming every position that there could be on the three different wheels. You would have to work at it very, very hard and after you had done it for a few hours you wondered, you know, whether you would see anything when it was before your eyes because you were so snarled up in it. But then of course, the magic moment comes when it really works and there it all is, the Italian, or the German, or whatever it is. It just feels marvellous, absolutely marvellous. I don't think that there is anything one could compare to it. There is nothing like seeing a code broken. That is really the absolute tops.
But in Hut 6, the codebreakers would not sit and decypher whole messages. They broke the keys and once they had done that left it to other less qualified staff to decypher the actual message. âWhen the codebreakers had broken the code they wouldn't sit down themselves and painstakingly decode 500 messages,' said Peter Twinn. âI've never myself personally decoded a message from start to finish. By the time you've done the first twenty letters and it was obviously speaking perfectly sensible German for people like me that was the end of our interest.'
Diana Russell Clarke was one of a group of young women in the Hut 6 Machine Room, decyphering the messages. âThe cryptographers would work out the actual settings for the machines for the day,' she said.
We had these Type-X machines, like typewriters but much bigger. They had three wheels, I think on the left-hand side, all of which had different positions on them. When they got the setting, we were to set them up on our machines. We would have a piece of paper in front of us with what had come over the wireless. We would type it into the machine and hopefully what we typed would come out in German.
The decyphered message then had to be distributed in some way that made it clear the information was important and was authoritative while at the same time preserving the security of the source. It could not be revealed that the British were
breaking
Enigma. A new section was formed in Hut 3, next door to Hut 6, in what had previously been the Army section, in order to report the material down the line. The section was made up of just three men. It was headed by Commander Malcolm Saunders RN, even though Hut 6 only dealt with German Army and
Luftwaffe
messages. He was assisted by
Squadron-Leader
Courtley Nasmith Shaw, an MI6 officer with
experience
of running operations inside Germany, and by F. L. âPeter' Lucas, a fellow of King's College, Cambridge, and a famous writer and poet who is best remembered now for his scathing attacks on the modernist poetry of T. S. Eliot.
The material was to be disguised as an MI6 CX report, the standard format for reporting information collected by MI6 agents abroad, de Grey recalled. MI6 was supposedly in contact with a German left-wing organisation which was feeding it the new material. âNow the essence of the security of an agent is that he should never be recognised as such,' de Grey said.
He must always masquerade in sheep's clothing with a solid âcover-story' in case suspicion is aroused. Some sort of
cover-story
had already been produced for Bletchley Park â the air defences of London or the like â cover-stories were in vogue. Since the orders were that the material was all to be cast in the form of a report from an agent, conventions had been adopted, such as âsource saw an order to â¦', or in the case of corrupt messages âfrom part of a torn document source was able to report â¦' Only the pith of the messages was extracted. Ingenuity would be exercised but none the less the reports had none of the cut and dried smack of a service telegram such as would carry conviction to a service mind.
Therein, initially at least, lay the problem. The three services were already suspicious of anything that MI6 reported so while the format adopted by Hut 3 was secure, it took a while before the material began to be taken with the respect it deserved. At this stage, only thirty officers outside of Bletchley Park were aware that Enigma was broken â the six Royal Navy officers âin the know' included Ian Fleming, the creator of James Bond, then the main naval intelligence officer liaising with MI6 and Bletchley Park. Hut 3 was only to pass the material out via MI6 itself so the reports had to be sent by teleprinter to the MI6 Air Section, Section II, if they were about the
Luftwaffe
, and to the MI6 Military Section, Section IV, if they were about the German Army, and at night they must be passed only to the MI6 duty officer. Although, initially, the standard of the intelligence Hut 6 was reporting did not appear to justify the stringent security measures surrounding it.
âOn a snowy January morning in 1940, in a small bleak wooden room with nothing but a table and three chairs, the first bundle of Enigma decodes appeared,' said Lucas.
The four of us who then constituted Hut 3 had no idea what they were about to disclose. Something fairly straightforward like German Police, or something more like diplomatic â neat and explicit documents straight from the office-tables of the
Führer
and the
Wehrmacht
that would simply need translating and forwarding to ministries?
They were neither. In after-years, even the
Führer's
orders were duly to appear. But meanwhile here lay a pile of dull, disjointed, and enigmatic scraps, all about the weather, or the petty affairs of a
Luftwaffe
headquarters no one had heard of, or trifles of
Wehrkreis
business; the whole sprinkled with terms no dictionary knew, and abbreviations of which our only guide, a small War Office list, proved often completely innocent. Very small beer in fact, and full of foreign bodies.
I
n the early hours of 9 April 1940, the Phoney War came to an abrupt end with the start of Operation
Weserübung
, the first German military operation in the West. The
Wehrmacht
invaded first Denmark and then Norway, occupying Copenhagen and landing in the Norwegian ports of Narvik, Trondheim, Bergen and Stavanger that night. Almost immediately a new Enigma key made an appearance. Six days later, the
Yellow
key as it had been dubbed in Hut 6 was broken, producing a mass of
intelligence
on the German operations in Norway. The intercepted messages told the British virtually every detail of what the advancing Germans were doing, said Peter Lucas, one of just three intelligence reporters then working in Hut 3.
Very shortly Hut 6 began breaking it, not occasionally, but daily, and often within a few hours of the introduction of each fresh day’s setting. The volume of traffic was considerable, the information important and often urgent: frequently it was read in Hut 3 within a few hours or even less than an hour after it had been sent out by the German station. Hut 3, and through them the ministries, were thus able to follow closely the Norwegian campaign from the point of view mainly (but not exclusively) of the
Luftwaffe
. They learned, not merely what had been done, but what was intended. They received information both strategic and tactical, information which covered not only air operations, but also military and naval operations, and not operations alone, but matters of supply, organisation and politics.
Both Hut 6 and Hut 3 were unprepared for such a huge amount of material and began working around the clock to break the keys and get the intelligence they were producing to London. Up until now, shifts had worked from 10am to 6pm with all the reports being bagged up at the end of the day and sent down by van to MI6 headquarters in Broadway. They were then passed on to the War Office, Air Ministry and the Admiralty, but with British troops deploying to Norway, and Royal Navy ships also taking part in the operation, the Bletchley intelligence was needed urgently. Initial attempts to pass Hut 3’s reports by telephone were abandoned as insecure and a teleprinter was installed in Hut 3 so that the intelligence reports,
carefully
couched in the language of an MI6 report from a spy, a so-called CX Report, could be sent direct to either Section II of MI6, the Air Section, if they related to the
Luftwaffe
or Section IV (military) ground forces operations. Nigel de Grey said they were then distributed by hand to the War Office, Air Ministry, Admiralty and Foreign Office.
The Enigma traffic, from having been interesting and
informative
, became suddenly highly operational. The business of Hut 3 and Hut 6 was to ensure that the facts as reported by the enemy reached the three Service Ministries in the shortest possible time. This entailed working round the clock seven days in the week and the Hut 3 party were formed into three eight-hour watches.
The authorities were so concerned over the propriety of having young women working alongside young men overnight in Hut 6 that they insisted that where women were working night shifts there must be at least six women, recalled Stuart Milner-Barry, the deputy head of Hut 6.
The innovation was thought to be not only a strange fad but dangerous to the morals of a mixed community. Indeed, a
total of three girls, which was all that we required, was thought to be insufficient to ensure the observance of the proprieties and – presumably on the principle that the men would be overworked by such large numbers – a minimum of six was insisted upon.
As a result, three girls from another department had to be put on the night shift, not to work but simply to act as ‘dummies’, Milner-Barry said. Fortunately, this was ‘a precaution that was dropped by tacit consent after a short interval’.
The number of Hut 3 intelligence officers, all of whom were at this stage male, only doubled to produce two-man watches which rewrote the information completely to remove any suggestion that it had come from decyphered Enigma messages and to give the impression that it was the product of
Source Boniface
, a supposedly left-wing German recruited by the British and running a network of spies inside the German armed forces. At this stage it was known only as ‘Source CX/FJ’, the FJ being a designation for the supposed MI6 spy.
R. V. Jones, who as MI6 scientific adviser was a frequent recipient of reports from Bletchley Park, said they were
disguised by some introduction such as ‘A reliable source recovered a flimsy of a message in the wastepaper of the Chief Signals Officer of
Fleigerkorps
IV which read…’, or in the case of an incomplete decrypt, ‘Source found a partly charred document in the fireplace of….’ I can remember handing a disguised decrypt to Air Commodore [Charles] Nutting, the Director of Signals, who exclaimed: ‘By Jove, you’ve got some brave chaps working for you.’ Inevitably, there was speculation about the identity of the supposed secret agent or agents who were sending back such valuable reports. Gilbert Frankau, the novelist, who had a wartime post in intelligence, told me that he had deduced that the agent who could so effectively get into German headquarters must be Sir Paul Dukes, the legendary
agent who had penetrated the Red Army so successfully after the Russian Revolution.
It wasn’t easy by any means to give the Enigma messages the appearance of credible agent reports. They were in a very formatted, specialised military German that was very different from the language as studied by the German scholars who were working on it and were more used to the writings of Schiller and Goethe. It was ‘a difficult and fundamental problem’, recalled Peter Lucas.
It was not a matter of receiving straightforward messages and translating them: it was always a matter of receiving
material
which was nearly always more or less imperfect, often incomplete, rarely intelligible with ease and at its worst totally meaningless even to the best German scholar.’
Mistakes by the German officer who drafted the original signal, or by the German operator who sent it, or by the British
operator
who intercepted it, or simply gaps in reception caused by bad atmospheric conditions all provided potential to make the decyphered message difficult to understand and were compounded by the military jargon or technical terms in use, or simply by colloquial usage of words and phrases that meant something quite different in ‘
Hochdeutsch
’, the German
equivalent
of ‘Oxford English’, that the codebreakers had studied at university.
‘The first task of the Hut 3 watch, from the very beginning, was to produce out of these corruptions, gaps, abbreviations and technical terms, a reliable and intelligible text, and to put that text into English,’ said Lucas.
To do this needed all the technique of the academic editor, with the great disadvantage, as compared with an academic editor, that the work generally had to be done in a hurry. Corruptions
had to be emended and reasonable conjectures had to be made as to the contents of missing passages, if this could all be done safely. Something more than ‘academic’ knowledge of the German language was necessary for this sort of work. Undoubtedly it was generally true that the best linguists did the best work. Nevertheless, there were one or two striking instances of men whose academic training was incomplete, or whose knowledge of German had apparently been more or less casually picked up, achieving remarkable success with Enigma texts. They were, however, exceptional.
Our experiences in Hut 3 demonstrated clearly enough, as a general proposition, that what we really needed – but did not get, at any stage, in sufficient numbers – were men with a
first-class
training in German, plus a sort of flair for handling these particular linguistic problems, which some of the less likely workers displayed, but some of the more likely did not. This flair is difficult to define and was impossible to predict in any untried watchkeeper. A combination of academic knowledge and flair carried a watch worker further than anything else, but not many watch workers possessed both in sufficient degree. The technical terms, many of them quite unknown at the beginning to anyone save a few specialists in Germany, and the abbreviations, and the plain bad German usages sometimes yielded to straightforward academic or technical knowledge, sometimes to more or less inspired guesswork, sometimes not at all.
The Norwegian campaign was an exhilarating time for both the Hut 6 codebreakers and the intelligence officers in Hut 3. There was a feeling of triumphant excitement that the system put in place by Welchman and Travis was actually doing its job, that the relatively sedate organisation that GC&CS had been between the wars was being turned into a codebreaking
production
line that was reading the Nazis’ most secret signals. The Hut 6 codebreakers were also relieved to discover how much
easier it was to crack the Enigma keys when there was a mass of operational signals on which to work.
While Bletchley was now producing valuable operational intelligence, it was not yet appreciated for what it was among the customers, not least because the tight security surrounding the break into Enigma still meant that a total of only thirty Army, Navy and RAF officers knew the truth about where the
Boniface
intelligence was coming from. Everyone else believed that it was the product of an MI6 agent network. Unfortunately, there was little confidence among the service intelligence departments in MI6 agent networks and their reports and among the vast bulk of recipients who knew nothing of the Bletchley secret, Source Boniface was regarded with outright suspicion. Nor had anybody given any thought as to how this intelligence might be passed to the troops on the ground where it could be of some use.
The solution came as a result of the fact that the MI6 head of station in Oslo Frank Foley was retreating through Norway with the Norwegian commander-in-chief Major-General Otto Ruge. Foley was passing messages back to MI6 from Ruge asking for assistance. He was encoding them in the MI6 emergency code. This was a ‘book code’ which used an 1865 edition of John Ruskin’s book
Sesame and Lilies
. Both Foley and the MI6 decoding section based at Bletchley had exactly the same edition and the messages were made up of words taken from the text and identified by page number, line number and position of the word within the line. This link was used to send Foley details of German progress garnered from the Bletchley Park decodes and disguised as information ‘from our own forces’.
The German occupation of Norway was still
continuing
, providing large amounts of intelligence from the
Yellow
Enigma, when the Germans invaded Holland, Belgium and Luxembourg, sweeping into France. Unfortunately, at the same time, they changed the Enigma indicating system on the Army and
Luftwaffe
cyphers they were using, preventing attempts
by Hut 6 to break the Enigma cyphers and leading to concern among the codebreakers that they might not get back into it at all. No matter how hard they tried, they simply could not get back into the Enigma cyphers.
The solution came as a result of a moment of inspiration by John Herivel. The 21-year-old from Belfast was brought in at the end of January 1940 by Gordon Welchman, who had been his mathematics tutor at Sidney Sussex College, Cambridge, and recognised that he had exceptional talent as a mathematician. He was one of the new boys trying to think of ways to break into the
Red
. After being taught ‘the mysteries of the Enigma’ by Alan Turing and Tony Kendrick, Herivel was sent to Hut 6.
‘I had been recruited by Welchman and I was going to work in his show,’ Herivel recalled.
I do remember that when I came to Hut 6, we were doing very badly in breaking into the
Red
code. Every evening, when I went back to my digs and when I’d had my supper, I would sit down in front of the fire and put my feet up and think of some method of breaking into the
Red
. I had this very strong feeling: ‘We’ve got to find a way into the
Red
again.’ I kept thinking about this every evening and I was very young and very confident and I said I’m going to find some way to break into it. But after about two weeks, I hadn’t made any progress at all.
As he thought about how the system worked, and might be unravelled, Herivel tried to get into the mind of the operators who were setting up the Enigma machines. How did they go about it; what were they thinking when they did it? The operators using Enigma began each day by putting the correct wheels and ring settings into their machines. They then selected an
opening
position for the wheels themselves – there was no laid-down starting position – and then they sent that starting position as a three-letter indicator at the beginning of their first messages.
‘Up until the middle of February, I had simply been
thinking
in terms of the encyphered messages which were received daily and which came to Hut 6,’ Herivel recalled.
Then one evening, I remember vividly suddenly finding myself thinking about the other end of the story, the German
operators
, what they were doing and inevitably then I thought of them starting off the day. I thought of this imaginary German fellow with his wheels and his book of keys. He would open the book and find what wheels and settings he was supposed to use that day. He would set the rings on the wheels, put them into the machine and the next thing he would have to do would be to choose a three-letter indicator for his first message of the day. So I began to think, ‘How would he choose that indicator?’ He might just take it out of a book, or he might pluck it out of the air like ABC or whatever. Then I had the thought, suppose he was a lazy fellow, or in a tearing hurry, or had the wind up, or something or other and he were to leave the wheels untouched in the machine and bang the top down and look at the windows, see what letters were showing and just use them.
Then another thought struck me. ‘What about the rings? Would he set them for each of the three given wheels before he put them into the machine or would he set them afterwards?’ Then I had a flash of illumination. If he set them afterwards and, at the same time, simply chose the letters in the windows as the indicator for his first message, then the indicator would tend to be close to the ring setting of the day. He would as it were be sending it almost in clear. If the intercept sites could send us the indicators of all the
Red
messages they judged to be the first messages of the day for the individual German operators there was a sporting chance that they would cluster around the ring settings for the day and we might be able to narrow down the 17,576 possible ring settings to a manageable number, say twenty or thirty, and simply test these one after
the other in the hope of hitting on the right answer. The next day I went back to Hut 6 in a very excited state and told my colleagues of this idea. ‘Oh, brilliant,’ they all said. Welchman immediately arranged, very discreetly, for first message
indicators
on the
Red
to be sent early each day to Hut. It was a simple matter to look for clusters. The idea, as my colleagues said, was a good one, and it was faithfully tested every day. Unfortunately, it never worked.
