Despite it's actual use
It's a thing of beauty.
A World War II German Enigma cipher machine is on the block at Bonhams, the London auction house, this month. The 1941 oak model, described as an "extremely rare example", is expected to go under the hammer on 14 November for an estimated £40,000-£60,000. In 2010, a 1939 Enigma fetched £67,250 at auction - that model was …
I recall reading that the 4th rotor is what game them the big break; Some U-Boats would operate with the 4th rotor disabled for messages encoding on just the three (weather reports I think?) and this allowed them to see the changes between three / four rotor encoding messages and move on form there.
Or something like that, I'm sure someone will be along to clarify / flame as incorrect*
*delete as appropriate
From my reading all Enigmas had three rotor positions.
The configuration for the day fitted the available rotors in different positions, each with defined side to side orientation.
To improve the security of the system the possibilities were increased by adding rotors to the kit. The options are to the power of the number of rotors. The army added one, the submarine service added two.
A guide at Bletchley assured me that rotors were extremely difficult to make.
And the bombe was designed by Polish cryptanalysts who cracked the system. They managed to get their result to the UK as Poland fell. Most of the team were interned, and kept shtum.
@Badooka, @Alan Firminger
as the third wheel already rarely moved (theoretically one step per 26^2 message characters), the fourth wheel on the M4 would essentially be just a static selector. The Germans knew this, and didn't adapt the design to make the fourth wheel driven. Together with the Umkehrwalze (which was also made changeable on the M4) it did increase the number of starting positions, but not the cryptographic complexity.
The bombe is usually attributed to Turing, not Polish cryptanalysts. It automated his "probable phrase" attack.
The Polish cryptanalysts are usually credited with the different, but equally amazing feat of reverse-engineering the design of the military Enigma, using an attack based on a flawed operating procedure. This attack was aided by knowledge of the pre-war civil Enigma procedures, details of which had been leaked to the French. The German authorities knew of this leak, but failed to realise its significance for the security of the military Enigma, nor that Enigma's "no self coding" constraint itself leaks sufficient information to break the cypher. It was knowledge of the design of the military Enigma, obtained from Poland, and its "no self coding" constraint, that led to Turing's "probable phrase" attack, and hence the bombe.
It may be the case that Polish mathematicians had worked out the theory of the "probable phrase" attack, and the possibility of automating it, before Turing. It would be good to know if there is any evidence that they, rather than Turing, deserve the credit for these insights. Either way, it could be a bit misleading to say "the bombe was designed by Polish cryptanalysts".
@Chris Fox
From http://users.telenet.be/d.rijmenants/en/enigma.htm :
"When the Wehrmacht introduced the plugboard on the military Enigma, this added an astronomical number of possible key stettings. The general idea was that this military Enigma, unlike the commercial types, would be impossible to break. No one even tried to break it. However, in 1932, Poland's Biuro Szyfrow (Cipher Bureau) initiated attempts to analyse and break the Enigma messages. Although the chief of this Bureau received copies of codebooks sold by the German spy Hans-Thilo Schmidt, he did not give them to his codebreakers. He thought that keeping this information from them might stimulate their efforts.
Marian Rejewski, Henryk Zygalski and Jerzy Rozicki were convinced that mathematics could solve the problem and succeeded in breaking the Enigma messages. They also developed an electro-mechanical machine, called the Bomba, to speed up the codebreaking process. Two major security flaws in the German Enigma procedures were the global groundsetting and the twice encodes message-key, a procedure to exclude errors. These flaws opened the door to cryptanalysis. In 1939 the Bureau was no longer able to break the codes due to increased sophistication in the design, new procedures and lack of funds for the code breakers. When Germany invaded Poland, the Polish Biuro Szyfrow passed its secret knowledge and several replica Enigma machines to the baffled French and British intelligence. The work of the Biuro Szyfrow was vital, not only because their pioneering work itself, but also because it convinced other cipher bureaus that it was possible to break Enigma."
This post has been deleted by its author
"Haha "even the ONE-TIME-pad can be broken if it is over-used"
Not sure if a joke or a FAIL..."
I think you'll find the OP meant to say 'the original code' should not be repeated, in part or in whole, across a batch or set of one time pads, or indeed ever. Call it 'the full set', to use logical language. Repeating code segments across one time pads was a mistake that people made, believe it or not.
kind of like good old fashion 1337
If message length is kept short, then other encryption methods work just fine - it is too much repetition that is the killer, and even the one-time-pad can be broken if it is over-used. ->
1Ph /\/\3$$493 L3|\|97|-| 1$ |<3P7 $|-|0r7, 7|-|3|\| 07|-|3r 3|\|(r'/P710|\| /\/\37|-|0D$ \/\/0r|< jU$7 Ph1|\|3 - 17 1$ 700 /\/\U(|-| r3P371710|\| 7|-|@ 1$ 7|-|3 |<1LL3r, 4|\|D 3\/3|\| 7|-|3 0|\|3-71/\/\3-P4D (4|\| b3 br0|<3|\| 1Ph 17 1$ 0\/3r-U$3D.
"If message length is kept short, then other encryption methods work just fine - it is too much repetition that is the killer"
Indeed and, IIRC, in the early days a stage in coding was cracked because a signaller repeated an exact message with one error included. Additionally ISTR that a phrase was repeated in another message elsewhere, which helped enormously, and added to the overall picture of success in attacking the German code.
"even the one-time-pad can be broken if it is over-used."
Not quite; reusing the code in a one time pad which, by definition, is used only once. This is where we can discuss what truly random means, whether this is meaningful and the like. I recently read an article in which someone suggested that a shortened password is less vulnerable to being cracked than a long one with digits, symbols, upper and lower case symbols. That has not changed my policies one bit.
There's a strong case that the Germans always thought the Enigma was unbreakable. See the book "Delusions of Intelligence: Enigma, Ultra, and the End of Secure Ciphers" by RA Ratcliff. To quote from the blurb on Amazon: "German intelligence experts conducted numerous internal investigations that all certified their ciphers' security".
The Germans knew that one of their guys had handed over papers and settings to the French in the early 30s, which enabled the Poles to crack the pre-War Enigma. They didn't realise that wartime Enigma was to some extent compromised by that leak.
On the other hand, Donitz suspected on numerous occasions that Enigma had been broken - hence the Navy's 4-rotor machines and more complex crypto procedures. But even he was persuaded that any compromises were only temporary.
Whether it was unbreakable or not, for most of the messages transmitted on the Enigma it didn't matter. They were used quite extensively by the German land forces for messages that were acted on almost immediately, meaning that if they were intercepted, then by the time they were decoded at Bletchley Park the intelligence gleaned from them would have been useless. The decryption was far more useful in the war against the U-Boats, since the messages were quite often full of long term operational information. German high command also had the more sophisticated Lorenz cipher machines for communicating with regional commands, and it was decrypting this cipher that the Colossus computer was used for.
@Chris Wareham
on the other hand, the Wehrmacht had quite a lot more machines in use, and their operators weren't as highly trained (and rigorous) as those in the Kriegsmarine. Forgetting to change the rotor settings from the previous day (and then resending the same message once that was rectified) was a regular enough occurrence to help Bletchley Park quite a bit. And then there was the German insistence on fully adressing the recipient, name, rank and unit. Although those would obviously differ per message, it significantly limited the possible words at the start of the message, making rejecting incorrect decoding settings much easier.
@Chris Wareham
even if most of the messages are time-limited regarding the usefulness of their information, knowing that message #1 indeed meant "Unit X moves to take up positions east of Landmark A" when you have information from another source that Unit X has now moved and is east of Landmark A, means that you can validate other messages from that day, some of which may contain more strategical, long-term info.
An interesting moral conundrum was when British ships and allies were deliberately put at risk by not acting on information discovered in the messages, if the British were too obvious with their information gathering, then Enigma would have been known as insecure by the Germans (rather than just suspected insecure) and they would have blanket changed to 4 rotor, or more (was there a 5 rotor transvertex?), perhaps introduced plugboards eairlier, a complete new set of differently encoded 5 rotors or changed the daycode more regularly.
In reality, there were sacrifices to make it look like the British had not cracked everything, but balancing that with the need to have a tactical advantage must have been a horrible situation to be in, being in the war room, knowing where the enemy was but not acting on the information so obviously that they know you know.
I believe this technique was mainly dropping mines in very specific areas close to ports, and was referred to as 'Gardening". It gave the excuse to return to the same place again and again, which was far less likely to happen on land (if the RAF could find exactly the same place at night anyway).
What no-one ever mentions is that, of course, the Germans were breaking our codes too, and doing so, I think, quite successfully. Indeed, there is at least one case, I think, where a convoy was rerouted based on reading German codes, and the Germans then read the British rerouting instructions and told the U-boats where the convoy would now be.
It's kind of sad that the Wikipedia entry on "B-Dienst" (the German naval codebreaking organisation) is one short paragraph. I mean, they clearly were the bad guys of course, but it would be interesting to get an even slightly unbiased opinion as to what actually went on. Apart from anything else it might help understand how the battle of the Atlantic was actually won, which probably really wasn't the heroic people at Bletchley, but rather a combination between US support and (mostly) the RN and RAF finally pulling their fingers out and talking to each other properly and sorting out air cover that worked (in particular the RAF were obsessed by bombing Germany and just willfully ignoring the "if we don't fix the U-boat problem *right now* we will not be in the war in 6 months" problem, and the RN really didn't get the "you have to defend the convoys, that's all that counts" thing for a long time).
One author, I think it was Cave Brown, reported that the Germans had a team trying to crack the British TypeX machine, which was in principle the same as the Enigma but had 8 rotor positions. Year by year as they made no progress the team was reduced in size until there was one person on the task.
This is interesting logically. If they believed that the Enigma was secure what hope did the have for six years of breaking something clearly far better ? Why did they not apply the team to breaking their own machine first ?
Cracking the Enigma depended on having a long crib to match against the cyphertext from which was derived all the settings. As the Enigma never encoded a character to itself the match was exact. Odd. The reason for this give away is in the circuit. The signal passes once through the rotors, is reflected at the end and returns through the same cats cradle on different wires to the input side of the first rotor. If the reflected signal were picked up after passing through the middle rotor then although the complexity of the system would have been reduced, possible combinations are unchanged and the method of decryption would have been frustrated.
What no-one ever mentions is that, of course, the Germans were breaking our codes too
And compromising resistance networks, most notably the Dutch one. The Abwehr managed to capture a Dutch radio operator and his equipment, then proceeded to hoover up the rest of the operators. Meanwhile they used the original operators' equipment to carry on receiving transmissions, since the Allies hadn't twigged that anything was wrong. This is why the paratroop drops and ground offensive to take the Rhine bridges (Operation Market Garden) was such a disaster - the Germans knew exactly what was going to happen, and despite being so stretched at this point of the war they had an armoured division waiting. As a result, one of my grandfathers - possibly the oldest enlisted volunteer - spent the rest of the war in captivity. The truth was considered too embarrassing in the immediate aftermath of the war, and was only declassified a few years ago.
@Chris Wareham
not only the operator's equipment, but, more importantly, they turned the operator, so that messages were still sent with his hand.
But about Market Garden: I don't quite buy it. If the Germans knew exactly what was going to happen, why did they only have one, severely cut down and recuperating, Panzer division waiting (if that) at the end of the string of bridges? Things were already going badly for them, and trading off the fact that the Allies would know (or at least strongly suspect) their crypto was broken against the possible loss of those bridges, and subsequently everything west of them, including the Antwerp and Rotterdam ports would have weighed strongly in favour of putting a few more units up elsewhere too. In the whole operation, even around Arnhem, there are enough situations of Germans being caught with their pants down to make it clear that large chunks of the German forces around didn't quite know what was up, taking a day or more to spring into action.
(my dad spent the rest of the war as an evacuee; my grandfather's shop was about a kilometer away from the Arnhem bridge)
You can almost imagine the scene, with bureaucrats from both services pointing that they role was fighting the German Navy and bombing Germany, not running around after merchant shipping or pointless flying around the North atlantic looking for submarines which were probably all underwater
There is a story regarding the massacre of the Jews in all this. It was known, via deciphered communications, that this was going on, not always in time to do anything but that it was known what was happening.
The story goes that a town in northern Italy was due to have it's Jewish population sent to the concentration camp, some 60,000 people. However, if the British high command told the resistance cell in that town and got them to act on it, it would be known that we'd broken the code and would have caused problems for the D-day plans.
Churchill made the call to keep it quiet.
It's mentioned in some of the many biographies of Churchill. A decision on a scale the vast majority will never, ever have to take.
"Yes you are right the German Navy did switch to 4 rotors then."
Wasn't that when the US navy hunted down a German naval unit and captured an undamaged working copy of the latest enigma machine, thereby saving the human race from certain destruction? ;-)
I saw it on Tee Vee. Honist.
"The website Dirk Rijmenants' Cipher Machines and Cryptology has a very good account of Enigma, complete with technical details,"
http://users.telenet.be/d.rijmenants/en/enigma.htm
I'd just like to endorse that statement. Great website, great Enigma simulator, and - from the email exchanges I've had with Dirk - a great guy too. Highly recommended!
To a large extent, this machine was the key to many of Rommel's successes.
Hitler would send Enigma-encrypted orders to Rommel, the orders would be intercepted and de-crypted at Bletchley Park and forwarded to Auchinleck (or Montgomery), Rommel would blithely disobey Hitler's orders.
If only I had £50,000. I'd spend them on drugs and teenage prostitutes of course, but still...
"Lordy. Not so keen on Heydrich, one could only hope? ;)"
No shame in admiring Rommel. He was, after all, a gifted military strategist and supreme tactician. He was a humane officer who ignored orders to kill civilians, captured Jewish soldiers or commandos and treated prisoners of war well.
And he was complicit in a plot to kill Hitler which ultimately led to his death. When his involvement was discovered he commited suicide by cyanide capsule in order that his family be spared execution.
All in all, a noble man respected by his own men and his enemies alike and who died with dignity.
Heydrich, however was a right shitbag.