Fascinating.
Thanks for this!
Did the Germans ever got clued up that their supposedly encrypted message system have been compromised?
El Reg had the honour of speaking with a war hero last Friday when the UK's National Museum of Computing fired up its replica Enigma code-breaker to decrypt messages sent from Poland. Ruth Bourne was among hundreds of Wrens who worked on the front line of code-breaking on 200 or so Bombe machines1 at sites in and around …
As I understand it, the Germans received a couple of hints that codes might have been cracked. At one point they went back and interviewed the polish codebreakers they captured who had cracked Enigma before the war but the poles were able to convince the Germans that changes in procedures meant the same cracking shortcuts they used were no longer possible.
On more than one occasion I think they reviewed the information they had and convinced themselves that it was still uncrackable , rather than erring on the side of caution as the allies often did.
Did the Germans ever got clued up that their supposedly encrypted message system have been compromised?
Apparently their techies always believed that it was impossible.
Other people had their suspicions. That's why Doenitz - head of the Navy, including U-boats - changed the Naval Enigma in February 1942, from a 3-rotor to a 4-rotor machine, with the result that BP was unable to decrypt the crucial U-boat traffic for nearly a year afterwards.
"BP was unable to decrypt the crucial U-boat traffic for nearly a year afterwards."
Even when unable to decrypt the traffic, the transmission themselves provided a shitload of useful metadata - RDF fixes on the origin coupled with each operator having a unique "fist" meant that they roughly knew where the subs were, etc.
Curiously, the Italians believed the Allies were reading Enigma traffic during the North African campaign. They found that a single scout plane spotting supply ships consistently in the Mediterranean Sea was too big a coincidence.
The German Navy was worried about decrypts after suffering from the UK successes in WWI. German naval cryptologists added a 4th wheel to Enigma and tightened up operating standards.
It might help that Wilhelm Canaris, head of the Abwehr (German Military Intelligence) is rumoured to have been a double agent.
German naval cryptologists added a 4th wheel to Enigma
That only effectively added a second 'reflector' [0] to the unit, and while that added cryptological complexity it was way less than the Germans thought it would. Every keystroke moved the rightmost wheel, then on one full rotation it moved the next one on the left, etc. So the fourth wheel hardly ever moved unless they had a very long message, and only its internal wiring and the starting position added to the coding.
[0] a disc to the left of the rotors, wired so that a current through the rotor wiring got routed back through the rotors again and to the 'display', a field of small lightbulbs displaying the (de)coded character for the pressed key. This made that the coding and decoding could be done on the same device with matching rotors and starting setting.
Neal Stephenson wrote a great book about the period (amongst another story set in modern times), in the Cryptonomicon.
Made that point precisely about having to not get 100% but also to give the Axis forces enough clues to think other means were the reason the Axis losses occurred.
Did the Germans ever got clued up that their supposedly encrypted message system have been compromised?
R.V.Jones' Most Secret War refers to this dilemma: acting on information versus keeping the fact that the code was broken under wraps. In some cases where acting was the strongly preferred option because of the anticipated consequences of not acting, a 'thank you for the info' was sent to a (non-existent) agent who could have plausibly provided the pertinent info.
British wartime intelligence went to great lengths to keep the secret. The high command even (as mentioned above, and downvoted for some reason) sometimes refused to act on Ultra intel, because they felt it could blow the gaffe.
There were some close calls, and the Germans must have had suspicions from time to time, but never to the point of acting on them, at least not concertedly and effectively.
Heck, if they'd just stopped saying "Heil Hitler" in every other message, that alone would have made the job significantly harder.
I'm presuming that formal official messages would have been composed in Standard German (or the equivalent of the time) with the stilted jargon which permeates organisations. Would there have been other messages -- banter between operators -- in dialect or vernacular German?
I think I read somewhere that elements of "personalisation" in the source messages did come into it. Each operator had their own keying style, so the interceptors knew who it came from, and certain operators would have little quirks in their message content.
Real needle-in-a-haystack stuff though, nevermind the complexity of the encyphering of the message.
I've read a few books and watched a few documentaries about this subject over the years, so in theory I kind of understand the principles....but of you gave me an Enigma-encoded message I honestly wouldn't know where to start. Very much hats off to everyone involved at Bletchley Park.
"I think I read somewhere that elements of "personalisation" in the source messages did come into it. Each operator had their own keying style, so the interceptors knew who it came from, and certain operators would have little quirks in their message content."
Yes, this was a known "feature" of telegraph operators almost from the invention of the telegraph. There many stories around of telegraph operators knowing who was "calling" from way the Morse was tapped out. No doubt any Ham operators still using Morse have the same experience today with their regular contacts.
Would there have been other messages -- banter between operators -- in dialect or vernacular German?
I know that as well as looking for "Heil Hitler", details of the weather, and similar stuff - one look-out post sent "Nothing to report" day after day, using different keys, a godsend for the codebreakers - some of the operators used to talk about their girlfriends. So the codebreakers would look for the girlfriend's name - and I'm sure that some of these messages would certainly have used the vernacular!
I know that as well as looking for "Heil Hitler", details of the weather, and similar stuff - one look-out post sent "Nothing to report" day after day, using different keys, a godsend for the codebreakers
And having established which look-out post sent this same very useful "nothing to report" message every day it was decided that the disturbance of the people in it might lead to a different report being sent, so the people in this look-out post had a very, very quiet war.
I found it!
The German radioman who always transmitted 'nothing to report' was stationed in the Qattara Depression in North Africa, in case the Allies tried bringing a whole army through impassable terrain.
Cite: wikipedia.org: Qattara_Depression#World_War_II
I found this thanks to Lindybeige, who mentioned it in an otherwise-only-tangential video relating to Great British Wartime Deceptions: youtube: watch?v=6ZYadpxoUbc
one look-out post sent "Nothing to report" day after day, using different keys,
The Wehrmacht (army) had some 40.000 Enigmas in use, and more than once a sloppy operator accidentally sent today's first message with yesterday's setting, then resent it with today's. If yesterday's code was already broken, then so was that day's. And if not, it certainly helped. Repeating a particular message, with some words abbreviated the second time, that the intended receiver hadn't been able to copy down correctly also offered cracking advantages.
The Kriegsmarine had way less devices and operators, and much tighter code discipline as well.
I remember seeing one documentary, and it mentioned that they noticed that one encrypted message had no "L"s in it. As the machine always changed the letters type in, they came to the conclusion that the message sent was something like -
LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL
- or similar
This post has been deleted by its author
but I don't think the Poles who got things rolling are given their due.
They most certainly are appreciated and given their due respects within those folk that follow the Bletchley Park story, and by Bletchley Park itself.
It's Hollywood history that neglects them, and quite a lot of other stuff too.
"Of course the heros of Bletchley Park are rightly lauded, but I don't think the Poles who got things rolling are given their due."
FWIW, ever since I first heard of all this, pretty much every story has included mention of the Poles and how the British built on their knowledge.
I remember seeing one documentary, and it mentioned that they noticed that one encrypted message had no "L"s in it. As the machine always changed the letters type in, they came to the conclusion that the message sent was something like -
LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL
I didn't know we'd been spying on the Welsh during WWII
First, thanks to El Reg, and to Ms Bourne, for this very interesting piece. These stories from people who were there are now, sadly, almost impossible to get.
Second, my late mum did this in Washington as a WAVE. She never really got into detail about what she did and whose codes she worked on, and we didn't think to grill her, but she could recite the alphabet backwards and always did the crossword in the morning with her coffee.
But my (virtual) hat is off to those like Ms Bourne, who did this work while in England, where everything was scarce, there was a constant danger of bombs, and the threat of invasion. Well done, ma'am, and thank you for your service!
Greatest Generation, indeed...they did more with less, and didn't complain about it.
To pretty much all of the people that did those things, they were just doing their job.
At the end of the day, they didn't have much choice in the matter, if they didn't do it then there was a good chance the country would have been invaded, and besides, if you're called up then your only choice is doing the work or going to jail (and having to do worse work).
>She never really got into detail about what she did and whose codes she worked on, <
The IEEE published memories from a Washington WAVE ('we joined the navy to see the world, but all we saw was DC'). She reported that after a strict security introduction, another man got up. They expected the good cop after the bad cop. Instead they got the worse cop: "DON'T EXPECT THAT YOU WILL BE TREATED ANY DIFFERENTLY BECAUSE YOU ARE WOMEN. IF YOU TALK ABOUT YOUR WORK, YOU WILL BE SHOT."
So true, I found out just last year after he had died that my father-in-law was one of the intercept operators stationed just outside of Loughborough in the midlands.
Real shame as I would have liked to know more.
"Second, my late mum did this in Washington as a WAVE. She never really got into detail about what she did and whose codes she worked on, and we didn't think to grill her, but she could recite the alphabet backwards and always did the crossword in the morning with her coffee."
If you've not yet seen it, you might enjoy both series (seasons) of The Bletchly Circle, a sort of detective story where the heroines are Bletchly almumin using their skills to solve the case, the second series being set when two of them follow a lead to the US and meet one of their opposite numbers who they only knew through a code name and telegraph messages.
It's not really directly related to this story of course, but shows a little of how women who were important and doing vital war work were often "pushed back into the kitchen" after it was over.
"I think I read somewhere that elements of "personalisation" in the source messages did come into it. Each operator had their own keying style, so the interceptors knew who it came from, and certain operators would have little quirks in their message content."
Yes this was used to generate individual cribs:-
- Operators sending messages to be passed on to girlfriends etc
- Standard form test messages that they were too lazy to alter as they were meant to.
- Returned test messages to check for clarity of transmission and retransmission
- Requisitions of stores and stocks were another good one. If you got a message from a fuel dump they would report stock levels periodically at a known time each week etc.
- And the well known weather and HH (I don't even want to type those two words) cribs
And the list get pretty big but one you know the 'touch' of the operator you could look for a likely crib in the decrypts.
Obviously you focus on the operators who spew out cribs with standard form messages at known times.
There are actually two types of "personalization." The first has been mentioned a couple of times, the habits of certain individuals to compose messages the same way. I especially enjoyed the "Nothing to report" standard message as being a good crib.
The other type of personalization was the Morse code sending itself. (Being an old ham radio operator, I speak from experience.) Every Morse code operator has his, or her, own rhythm. The weight (length) of the dots and dashes vary from person to person. With practice the interceptor learns to recognize the rhythm, sometimes called "the fist," of certain operators. Joe's code sending definitely sounds different than Mike's, etc. Then, couple Joe as the guy who always sent Jim's messages, and you have valuable meta data on beginning to decode messages.
This was part of Gordon Welchman's work that revolutionised military signals intelligence: Previously, everyone concentrated on the message content, but Welchman is credited with being the first to realise that what is said in a message is only part of the information that can be extracted from it. The identity of the sender and recipient impart information, as does the schedule of transmissions, the transmission power, and the attitude of the sender: a good Morse listener can't just recognise a "fist", they can also hear how relaxed or stressed that sender is, just a you can tell how relaxed someone is by their voice.
So, if you recognise an operator in Hamburg's keying style and know he sends lazily-keyed confirmations to messages about fuel supplies every morning, usually to other operators in the Baltic, but suddenly he's responding to stations in Dover and Rotterdam as well and with more urgency than his usual lazy fist, then it suggests that there's something happening along the Channel that might be worth sending a spy-plane to look at. And you discovered this without having to decrypt a single message.
>The other type of personalization was the Morse code sending itself. (Being an old ham radio operator, I speak from experience.) Every Morse code operator has his, or her, own rhythm.
I can confirm this. Being an old military telegraphist I too have experience in this. Where a dot lasts 50 - 80 ms the variation is of the order of a few milliseconds, it is surprising how quickly you learn to recognise everyone on the net. I learned to recognise all my colleagues and no two operators ever sounded the same.
Adding to the above poster it should be noted that you had some counter measures when enemy intelligence was clued onto you: use the other hand. The "fist" for left and right hands are very different and would be confused for a different person.
There are tons of stories in the world of Morse code.
The other type of personalization was the Morse code sending itself.
Yes, but that was for the people listening to pick up. Most of those were ordinary citizens that had (or got) a suitable receiver, with motorcycle messengers collecting the messages that were copied down. They did get told, if not trained, to spot particular operators by their keying. And of course particular Wehrmacht and Kriegsmarine stations were listened to by army staff radio operators, with a quicker way to get interesting messages to the code breakers.
The transmitters used have their own personalisation.
The equipment would have their own attack and decay characteristics (rise time and fall time) which could be seen on an oscilloscope. Also overshoot and 'ringing' and ripple on the pulses.
I don't know if oscilloscopes were around during the war years but were used in the '60s.
I don't know if oscilloscopes were around during the war years but were used in the '60s.
They were essentially the basis for radar, to the point that initially radar was just displaying the scope trace for the echo from a (semi)fixed[0] antenna. Only later the rotating sweep came along.
They certainly weren't used widely for identifying enemy transmitters, if at all.
[0] The antennas could usually be rotated, but only slowly because of their size. More like getting them pointed in a particular direction and sort of tracking a target.
>They certainly weren't used widely for identifying enemy transmitters, if at all.
Was rather more low tech - from the operators PoV, they synchronised 2 different signals in a 'stereo' headset to get the range/bearing at the start of transcription. What sat behind that is presumably rather more complex than simple triangulation, but that's definitely what they did as my father was deaf in one ear and someone had to do this for him each time the source changed.
There was also the method of generating cribs which was known as "gardening".
BP would get the RAF to drop ("sow") mines into the sea at a specific location. Later, they would "reap" the benefits by looking for reports which contained the word for "mines" and specified the location.
I hadn't realised until recently that the Soviets tried to get cribs on an industrial scale in the 1940s. Guy Burgess was one of the Cambridge spies and at the time was working at the BBC. He lobbied to be given access to non-secret Foreign Office cables, in order for the BBC to be able to better report foreign news. The real reason the KGB wanted this info was to have the actual clear content of enciphered FCO cables, that they could then use as a crib so they could read everything in that cypher. Of course more important messages might be using a different code, but the more you can read of the other guy's messages the better.
At the Beeb he wasn't getting much info. Apart from making good contacts (he was the original producer of the Week in Westminster on Radio 4 - that's been going ever since) - so he was desperate for more. Sadly the Foreign Office gave him a job, before the Beeb got round to sacking him for being drunk too often.