How the encryption codes of Hitler's Germany were broken. What is Enigma? How to crack the Enigma cipher using modern methods

Almost at any time of the year, the English countryside looks the same: green meadows, cows, medieval-looking houses and a wide sky - sometimes gray, sometimes dazzling blue. It was just transitioning from the first mode to the more rare second mode when the commuter train rushed me to Bletchley station. It’s hard to imagine that surrounded by these picturesque hills, the foundations of computer science and cryptography were laid. However, the upcoming walk through the most interesting museum dispelled all possible doubts.

Such a picturesque place, of course, was not chosen by the British by chance: inconspicuous barracks with green roofs, located in a remote village, were just what was needed to hide a top-secret military facility where they were constantly working on breaking the codes of the Axis countries. Bletchley Park may not look impressive from the outside, but the work that was done here helped turn the tide of the war.

Crypto hacks

In wartime, people entered Bletchley Park through the main gate by presenting a security pass, but now they buy a ticket at the entrance. I stayed there a little longer to look at the adjacent souvenir shop and temporary exhibition dedicated to First World War intelligence technologies (by the way, also an interesting topic). But the main thing lay ahead.

Bletchley Park itself is about twenty long one-story buildings, which in English are called hut, and in Russian are usually translated as “house”. I silently called them “huts,” combining one with the other. In addition to them, there is a mansion (aka Mansion), where the command worked and distinguished guests were received, as well as several auxiliary buildings: former stables, a garage, residential buildings for staff.

Those same houses The estate in all its glory Inside the estate looks richer than the huts

Each house has its own number, and these numbers have historical significance; you will definitely find them in any story about Bletchley Park. In the sixth, for example, intercepted messages were received, in the eighth they were engaged in cryptanalysis (Alan Turing worked there), in the eleventh there were computers - “bombs”. The fourth house was later allocated for work on the version of Enigma that was used in the navy, the seventh - for the Japanese variation on the Enigma theme and other ciphers, in the fifth they analyzed transmissions intercepted in Italy, Spain and Portugal, as well as German police encryption. And so on.

You can visit the houses in any order. The furnishings in most of them are very similar: old furniture, old things, tattered notebooks, posters and maps from the Second World War. All this, of course, did not lie here for eighty years: the houses were first transferred from one state organization to another, then they were abandoned, and only in 2014 did restorers meticulously restore them, saving them from demolition and turning them into a museum.

This, as is customary in England, was approached not only carefully, but also with imagination: in many rooms, the voices of actors and sounds are heard from hidden speakers, which create the impression that work is in full swing around. You walk in and hear the clatter of a typewriter, someone's footsteps and a radio in the distance, and then you "overhear" someone's animated conversation about a recently intercepted encryption.

But the real curiosity is the projections. For example, this man, who seemed to be sitting at the table, greeted me and briefly told me about the local customs.

Many rooms are kept in twilight so that the projections can be seen better

The most interesting thing, of course, was to look at Alan Turing's desk. His office is located in house eight and looks very modest.

This is what Alan Turing's desk looked like

Well, you can look at Turing’s creation itself - the Enigma deciphering machine - in house number 11 - in the same place where the very first model of the “bomb” was assembled at one time.

Cryptological bomb

This may be news to you, but Alan Turing was not the first to decrypt Enigma using brute force. His work is preceded by research by Polish cryptographer Marian Rejewski. By the way, it was he who called the decryption machine a “bomb.”

The Polish “bomb” was much simpler. Note the rotors on top

Why "bomb"? There are several different versions. For example, according to one, this was supposedly the name of a variety of ice cream beloved by Rejewski and his colleagues, which was sold in a cafe not far from the encryption bureau of the Polish General Staff, and they borrowed this name. A much simpler explanation is that in Polish the word "bomb" can be used to make an exclamation like "eureka!" Well, a very simple option: the car was ticking like a bomb.

Shortly before the capture of Poland by Germany, Polish engineers handed over to the British all the developments related to decoding German ciphers, including drawings of the “bomb”, as well as a working copy of Enigma - not a German, but a Polish clone, which they managed to develop before the invasion. The rest of the Poles' developments were destroyed so that Hitler's intelligence would not suspect anything.

The problem was that the Polish version of the “bomb” was designed only for the Enigma I machine with three fixed rotors. Even before the start of the war, the Germans introduced improved versions of Enigma, where the rotors were replaced every day. This made the Polish version completely unusable.

If you've seen The Imitation Game, you're already quite familiar with the setting at Bletchley Park. However, the director could not resist and made several digressions from real historical events. In particular, Turing did not create the prototype of the “bomb” with his own hands and never called it “Christopher”.


Popular English actor Cryptocode Podbirac as Alan Turing

Based on the Polish machine and the theoretical work of Alan Turing, engineers at the British Tabulating Machine Company created the “bombs” that were supplied to Bletchley Park and other secret facilities. By the end of the war, there were already 210 vehicles, but with the end of hostilities, all “bombs” were destroyed by order of Winston Churchill.

Why did the British authorities need to destroy such a wonderful data center? The fact is that the “bomb” is not a universal computer - it is intended exclusively for decoding messages encrypted by Enigma. As soon as this was no longer needed, the machines also became unnecessary, and their components could be sold off.

Another reason may have been the premonition that the Soviet Union would not be Britain's best friend in the future. What if the USSR (or anywhere else) began to use technology similar to Enigma? Then it is better not to demonstrate to anyone the ability to break its ciphers quickly and automatically.

Only two "bombs" survived from wartime - they were transferred to GCHQ, the UK Government Communications Center (think of it as the modern equivalent of Bletchley Park). They say they were dismantled in the sixties. But GCHQ graciously agreed to provide the museum at Bletchley with old drawings of “bombs” - alas, not in better condition and not entirely. Nevertheless, enthusiasts managed to restore them, and then create several reconstructions. They are now in the museum.

It’s interesting that during the war, the production of the first “bomb” took about twelve months, but the reconstructors from the BCS Computer Conservation Society, starting in 1994, worked for about twelve years. Which, of course, is not surprising, given that they had no resources other than their savings and garages.

How did Enigma work?

So, “bombs” were used to decrypt messages that were obtained after encryption with Enigma. But how exactly does she do this? Of course, we will not analyze its electromechanical circuit in detail, but general principle It's interesting to find out about the work. At least, it was interesting for me to listen and write down this story from the words of a museum employee.

The design of the “bomb” is largely determined by the design of the Enigma itself. Actually, we can consider that a “bomb” is several dozen “Enigmas” put together in such a way as to sort out the possible settings of the encryption machine.

The simplest Enigma is a three-rotor one. It was widely used by the Wehrmacht, and its design meant that it could be used by the average soldier, not a mathematician or engineer. It works very simply: if the operator presses, say, P, a light will light up under one of the letters on the panel, for example under the letter Q. All that remains is to convert it into Morse code and transmit it.

An important point: if you press P again, there is a very small chance of getting Q again. Because every time you press the button, the rotor moves one position and changes the configuration of the electrical circuit. Such a cipher is called polyalphabetic.

Look at the three rotors at the top. If you, for example, enter Q on the keyboard, then Q will first be replaced by Y, then by S, by N, then reflected (it turns out K), changed again three times and the output will be U. Thus, Q will be encoded as U. But what if I type U? It turns out Q! This means the cipher is symmetric. This was very convenient for military applications: if two places had Enigmas with the same settings, messages could be freely transmitted between them.

This scheme, however, has a big drawback: when entering the letter Q, due to the reflection at the end, under no circumstances could it be obtained Q. German engineers knew about this feature, but did not attach much importance to it, but the British found an opportunity to exploit it . How did the British know about the insides of the Enigma? The fact is that it was based on a completely unsecret development. The first patent for it was filed in 1919 and described a machine for banks and financial organizations, which allowed the exchange of encrypted messages. It was sold on the open market, and British intelligence managed to purchase several copies. By their example, by the way, the British Typex encryption machine was made, in which the defect described above was corrected.

The very first Typex model. As many as five rotors!

The standard Enigma had three rotors, but in total you could choose from five options and install each of them in any slot. This is exactly what is reflected in the second column - the numbers of the rotors in the order in which they are supposed to be installed in the machine. Thus, already at this stage it was possible to obtain sixty settings options. Next to each rotor there is a ring with letters of the alphabet (in some versions of the machine - the corresponding numbers). The settings for these rings are in the third column. The widest column is an invention of German cryptographers, which was not in the original Enigma. Here are the settings that are set using the plug panel by connecting letters in pairs. This confuses the whole scheme and turns it into a difficult puzzle. If you look at the bottom line of our table (the first day of the month), the settings will be as follows: rotors III, I and IV are placed in the machine from left to right, the rings next to them are set at 18, 24 and 15, and then the letters N are connected on the panel with plugs and P, J and V and so on. Taking all these factors into account, there are about 107,458,687,327,300,000,000,000 possible combinations - more than seconds have passed since the Big Bang. It is not surprising that the Germans considered this car extremely reliable.

There were many variants of Enigma, in particular, a variant with four rotors was used on submarines.

Hacking Enigma

Breaking the code, as usual, was made possible by the unreliability of people, their mistakes and predictability.

The Enigma manual says to select three of the five rotors. Each of the three horizontal sections of the “bomb” can check one possible position, that is, one machine can simultaneously run three out of sixty possible combinations. To check everything, you need either twenty “bombs” or twenty consecutive checks.

However, the Germans made a pleasant surprise for the English cryptographers. They introduced a rule according to which the same position of the rotors should not be repeated for a month, or for two days in a row. This sounds like it was supposed to improve reliability, but in reality it had the opposite effect. It turned out that by the end of the month the number of combinations that needed to be checked was significantly reduced.

The second thing that helped in decryption was traffic analysis. The British had been listening to and recording encrypted messages from Hitler's army since the beginning of the war. There was no talk of decryption at that time, but sometimes the fact of communication itself is important, plus such characteristics as the frequency at which the message was transmitted, its length, time of day, and so on. Also, using triangulation, it was possible to determine where the message was sent from.

A good example is the transmissions that came from the North Sea every day from the same locations, at the same time, on the same frequency. What could it be? It turned out that these were meteorological ships that reported daily weather data. What words might be contained in such a transmission? Of course, “weather forecast”! Such guesses pave the way for a method that today we call a plaintext attack, but in those days we called “cribs.”

Since we know that Enigma never outputs the same letters as the original message, we need to sequentially match the "hint" with each substring of the same length and see if there are any matches. If not, then this is a candidate string. For example, if we check the hint “weather in the Bay of Biscay” (Wettervorhersage Biskaya), we first write it opposite the encrypted string.

Q F Z W R W I V T Y R E * S* X B F O G K U H Q B A I S E Z

W E T T E R V O R H E R * S* A G E B I S K A Y A

We see that the letter S is encrypted into itself. This means that the hint needs to be shifted by one character and checked again. In this case, several letters will match at once - move them again. R matches. We move twice more until we come across a potentially correct substring.

If we were dealing with a substitution cipher, then we could end there. But since this is a polyalphabetic cipher, we need settings and initial positions of the Enigma rotors. They were the ones who were picked up with the help of “bombs”. To do this, pairs of letters must first be numbered.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

R W I V T Y R E S X B F O G K U H Q B A I S E

W E T T E R V O R H E R S A G E B I S K A Y A

And then, based on this table, create a so-called “menu” - a diagram that shows which letter of the original message (that is, “hints”) is supposedly encrypted into which letter and in what position. The “bomb” is set up according to this scheme.

Each of the reels can take one of 26 positions - one for each letter of the alphabet. Behind each of the reels there are 26 contacts, which are connected in thick cables in such a way that the machine searches for settings on the plug panel that give sequential matches of the letters of the encrypted string with the hint.

Since the structure of the “bomb” does not take into account the switching device inside the Enigma, it produces several options during operation that the operator must check. Some of them will not work simply because in Enigma you can only connect one plug to one socket. If the settings are not suitable, the operator starts the machine again to get the next option. In about fifteen minutes, the “bomb” will go through all the options for the selected reel position. If it is guessed correctly, then all that remains is to select the settings of the rings - without automation (we won’t go into details). Then, on English Typex machines modified to be compatible with Enigma, the encryption was translated into clear text.

Thus, operating with a whole fleet of “bombs”, by the end of the war, the British received up-to-date settings every day even before breakfast. In total, the Germans had about fifty channels, many of which broadcast much more interesting things than the weather forecast.

Allowed to touch with hands

At the Bletchley Park Museum you can not only look around, but also touch the decipherment with your own hands. Including using touchscreen tables. Each of them gives his own task. In this, for example, it is proposed to combine Banburi sheets (Banburismus). This is an early method of deciphering Enigma, which was used before the creation of “bombs”. Alas, it was impossible to decipher something in this way during the day, and at midnight all successes turned into a pumpkin due to the next change in settings.

Dummy “data center” in Hut 11

What is in house number 11, where there used to be a “server room”, if all the “bombs” were destroyed in the last century? To be honest, I still deep down hoped to come here and find everything in the same form as before. Alas, no, but the hall is still not empty.

Here are these iron structures with plywood sheets. On some there are life-size photographs of “bombs”, on others there are quotes from the stories of those who worked here. They were mostly women, including from the WAF, the women's service of the Royal Air Force. The quote in the picture tells us that switching cables and looking after the “bombs” was not an easy task at all, but an exhausting daily work. By the way, another series of projections is hidden between the dummies. The girl tells her friend that she had no idea where she would serve, and is completely amazed by what is happening at Bletchley. Well, I was also amazed by the unusual exhibit!

I spent a total of five hours at Bletchley Park. This was barely enough to get a good look at the central part and a glimpse of everything else. It was so interesting that I didn’t even notice how time passed until my legs began to ache and ask to go back - if not to the hotel, then at least to the train.

And besides the houses, dimly lit offices, restored “bombs” and long stands with accompanying texts, there was something to see. I have already mentioned the hall dedicated to espionage during the First World War, there was also a hall about the decryption of Lorenz and the creation of the Colossus computer. By the way, in the museum I discovered the “Colossus” itself, or rather the part that the reenactors managed to build.

A small computer history museum awaits the hardiest outside Bletchley Park, where you can learn how Computer Engineering developed after Turing. I also looked there, but walked quickly. I've already seen enough of BBC Micro and Spectrum in other places - you can do this, for example, at the Chaos Constructions festival in St. Petersburg. But you won’t find a living “bomb” anywhere.

A very strange monument in front of the entrance to the Imperial Palace of Poznan - a monument to Polish cryptographers, decipherers of the Enigma code. The stela, in the form of an equilateral triangular prism, is 3.10 meters high and is covered with seemingly random sequences of numbers. On each side of the monument there are 21 lines of twelve numbers, without any visible meaning. In the center of each face are letters that form names.

The monument was unveiled here in 2007, on the 75th anniversary of the decipherment of Enigma by three Polish cryptanalysts Marian Rejewski (1905-1980), Jerzy Rozycki (1909-1942) and Henryk Zygalski (1908-1978). The opening ceremony took place on November 10, 2007 (all sources mention that the cipher was opened in December 1932, without indicating exact date) in the presence of Rejewski's daughter, Rozycki's son, Mr. Jan Janusz Rozycki and two relatives of Henryk Zygalski. Other relatives of the three scientists, their former managers and colleagues from BS (Biuro Szyfrów) in Warsaw, PC Bruno and Cadix (the French branch of the bureau after the German occupation of 1939) were present.

The fact is that after the end of the First World War, part of the premises of the palace were given to the University of Poznan. In 1929, courses in cryptography, organized by the Polish Biuro Szyfrów (BS), began to operate in the rooms of the Faculty of Mathematics. Three students - Re(zh)evsky, Rozhitsky and Zygalsky were especially successful.


The Enigma code was invented by German cryptologist Arthur Scherbius in 1918, and began to be used in the Weimar Republic in the mid-1920s. At first experimentally, and since 1930 more and more often. Germany's neighbors, especially France, Britain and Poland, were suspicious, especially when the Nazis came to power in Germany in 1933. As part of the rearmament of the Wehrmacht, Scherbius cipher machines (by that time, as a result of a serious modification, it had turned from a 50-kilogram model A into a model C the size of typewriter) began to be used for encryption in all branches of the military.

The French and British were unable to decipher the Enigma code (Greek Riddle), and called it “unbreakable.” However, 27-year-old Rejewski, in his work in the BS4 division, cracked the code already in 1932. This was helped by a serious mistake made by Scherbius himself. If anyone is interested, you can read the article on Wiki (for me personally, a lot of things there provoke a reaction in the form of a facepalm)

Another article is about the British Museum, dedicated to this problem. Only there the Poles are not mentioned AT ALL, which in general is not surprising for the British.

In general, everything in this story is “beautiful”, starting from either the invention of Enigma by Scherbius himself, or the purchase of a patent from the Dutchman Hugo Koch, and ending with the “chemical castration” and suicide of the outstanding British scientist Alan Turing, who is known throughout the world, like an Enigma decoder.

But let's return to the Poles. When in the summer of 1939 it was already clear to everyone that Poland was about to fall, all the work of Polish scientists was transferred to British and French intelligence. There is no doubt (for me personally, yay) that this information was decisive for the work of British decryptors. The English mathematician and cryptanalyst Gordon Welchman, who was one of the leading employees of Bletchley Park (see the link about the British Museum), directly mentioned Polish contribution and assistance, writing: “... if they had not done this, the British cracking of Enigma might well not to happen at all...” It is believed that the deciphering of the Enigma code by British cryptographers shortened the war by about 2 years and saved many millions of lives.

Now there is a small museum next to the monument.


Here you can take short courses in cryptology :)


Unfortunately, we didn’t get to the museum because we were there after everything was closed, but I have Wikipedia!


Dmitry Alexandrovich was eager to see the monument with every fiber of his soul (before entering Pozan, I had no idea that it was there at all 0_0) and was completely satisfied, in principle, from here we could go back to our room *it seems*


But I’m glad that I found out this whole story (I knew about the Polish part, about Turing before, I should now watch the film with our Cumberbatch)

The German cipher machine was called "Riddle" not for the sake of words. There are legends surrounding the history of its capture and decoding of radio interceptions, and cinema largely contributes to this. Myths and truth about the German encoder are in our material.

The enemy’s interception of messages, as is known, can only be countered by reliable protection or encryption. The history of encryption goes back centuries - one of the most famous ciphers is called the Caesar cipher. Then attempts were made to mechanize the process of encryption and decryption: the Alberti disk, created in the 60s of the 15th century by Leon Battista Alberti, the author of the Treatise on Ciphers, one of the first books on the art of encryption and decryption, has reached us.

The Enigma machine used by Germany during World War II was not unique. But it differed from similar devices adopted by other countries in its relative simplicity and widespread use: it could be used almost everywhere - both in the field and on a submarine. The history of Enigma dates back to 1917, when the Dutchman Hugo Koch received a patent for it. Her job was to replace some letters with others using rotating rollers.

We know the history of decoding the Enigma machine mainly from Hollywood blockbusters about submarines. However, these films, according to historians, have little in common with reality.

For example, the 2000 film U-571 tells the story of a secret mission by American sailors to capture an Enigma encryption machine aboard the German submarine U-571. The action takes place in 1942 in the North Atlantic. Despite the fact that the film is spectacular, the story told in it does not correspond to historical facts at all. The submarine U-571 was actually in service with Nazi Germany, but was sunk in 1944, and the Americans managed to capture the Enigma machine only at the very end of the war, and this did not play a serious role in the approach of Victory. By the way, at the end of the film the creators report historically correct facts about the capture of the encoder, but they appeared at the insistence of the film’s consultant, an Englishman by birth. On the other hand, the film's director, Jonathan Mostow, said that his film "is a work of art."

European films try to maintain historical accuracy, but there is also a share of artistic fiction in them. Michael Apted's 2001 film Enigma tells the story of mathematician Tom Jericho, who must solve the updated code of a German cipher machine in just four days. Of course, in real life it took much longer to decipher the codes. At first, this was done by the Polish cryptological service. And a group of mathematicians - Marian Rejewski, Henryk Zygalski and Jerzy Rozicki - studying disused German ciphers, found that the so-called day code, which was changed every day, consisted of the settings of the switchboard, the order of installation of the rotors, the positions of the rings and the initial settings of the rotor . This happened in 1939, even before the capture of Poland by Nazi Germany. Also, the Polish “Bureau of Ciphers,” created specifically to “fight” Enigma, had at its disposal several copies of a working machine, as well as an electromechanical Bomba machine, which consisted of six paired German devices, which helped in working with codes. It was she who later became the prototype for Bombe, the invention of Alan Turing.

The Polish side was able to transfer its developments to the British intelligence services, who organized further work to crack the “riddle”. By the way, the British first became interested in Enigma back in the mid-20s, however, they quickly abandoned the idea of ​​​​deciphering the code, apparently considering that it was impossible to do so. However, with the beginning of World War II, the situation changed: largely thanks to the mysterious machine, Germany controlled half of the Atlantic and sank European convoys with food and ammunition. Under these conditions, Great Britain and other countries of the anti-Hitler coalition definitely needed to penetrate the Enigma riddle.


Sir Alistair Dennison, head of the State Code and Cipher School, which was located in the huge Bletchley Park castle 50 miles from London, conceived and carried out the secret operation Ultra, turning to talented graduates of Cambridge and Oxford, among whom was the famous cryptographer and mathematician Alan Turing . Turing's work on breaking the Enigma machine codes is the subject of the 2014 film The Imitation Game. Back in 1936, Turing developed an abstract computing "Turing machine", which can be considered a model of a computer - a device capable of solving any problem presented in the form of a program - a sequence of actions. At the code and cipher school, he headed the Hut 8 group, responsible for the cryptanalysis of German Navy communications, and developed a number of methods for breaking the German encryptor. In addition to Turing's group, 12 thousand employees worked at Bletchley Park. It was thanks to their hard work that the Enigma codes could be deciphered, but it was not possible to crack all the ciphers. For example, the Triton cipher worked successfully for about a year, and even when the “guys from Bletchley” cracked it, it did not bring the desired result, since too much time passed from the moment the encryption was intercepted until the information was transmitted to the British sailors.


The thing is that, by order of Winston Churchill, all decryption materials were received only by the heads of the intelligence services and Sir Stuart Menzies, who headed MI6. Such precautions were taken so that the Germans would not realize that the codes had been broken. At the same time, these measures did not always work, then the Germans changed the Enigma settings, after which the decryption work began anew.

The Imitation Game also touches on the topic of the relationship between British and Soviet cryptographers. Official London really was not confident in the competence of specialists from the Soviet Union, however, on the personal order of Winston Churchill, on July 24, 1941, materials with the Ultra stamp were transferred to Moscow. True, to exclude the possibility of disclosing not only the source of information, but also that Moscow would learn about the existence of Bletchley Park, all materials were disguised as intelligence information. However, the USSR learned about the work on deciphering Enigma back in 1939, and three years later, the Soviet spy John Cairncross entered the service of the State School of Codes and Ciphers, who regularly sent all the necessary information to Moscow.


Many people wonder why the USSR did not decipher the radio interceptions of the German “Riddle,” although Soviet troops captured two such devices back in 1941, and in the Battle of Stalingrad Moscow had three more devices at its disposal. According to historians, the lack of modern electronic equipment in the USSR at that time had an impact.

By the way, a special department of the Cheka, dealing with encryption and decryption, was convened in the USSR on May 5, 1921. The employees of the department had a lot of not very advertised victories, for obvious reasons - the department worked for intelligence and counterintelligence. For example, the disclosure of diplomatic codes of a number of countries already in the twenties. They also created their own cipher - the famous “Russian code”, which, as they say, no one was able to decipher.

Based on materials from the dissertation “Encryption machines and decryption devices during the Second World War,” defended at the University of Chemnitz (Germany) in 2004.

Introduction. For the general public, the word “Enigma” (in Greek - a riddle) is synonymous with the concepts of “cipher machine” and “code breaking”, which has been taken care of by films about submarines and similar novels that have little to do with reality. That there were others encryption machines, for the “hacking” of which special decryption machines were created, and little is known to the general public about the consequences that this had in the Second World War.

And not surprisingly: there is too little information about this in popular publications. And the information available there is usually either insufficient or unreliable. This is all the more regrettable because the breaking of encryption codes was of extremely important historical significance for the course of the war, since the allies (in the anti-Hitler coalition), thanks to the information obtained in this way, had significant advantages, they were able to compensate for some omissions of the first half of the war and were able to optimally use their resources in the second half of the war. According to Anglo-American historians, if it had not been for the breaking of German encryption codes, the war would have lasted two years longer, additional casualties would have been required, and it is also possible that an atomic bomb would have been dropped on Germany.

But we will not deal with this issue, but will limit ourselves to the scientific, technical and organizational circumstances that contributed to the disclosure of German encryption codes. And what is especially important is how and why it was possible to develop machine methods of “hacking” and use them successfully.
Hacking the Enigma codes and the codes of other encryption machines provided the allies with access not only to military-tactical information, but also to information from the Foreign Ministry, police, SS and railway. This also includes reports from the Axis countries, especially Japanese diplomacy, and the Italian army. The Allies also received information about the internal situation in Germany and its allies.

In England alone, a secret service team of thousands worked to decipher the codes. This work was personally supervised by the Prime Minister of England Winston Churchill, who knew about the importance of this work from the experience of the First World War, when he was the Secretary of the Navy of the British government. Already in November 1914, he ordered the deciphering of all intercepted enemy telegrams. He also ordered that previously intercepted telegrams be deciphered in order to understand the thinking of the German command. This is evidence of his foresight. The most famous result of this activity was forcing the US entry into the First World War.
Equally far-sighted was the creation of English listening stations - then a completely new idea - especially listening to the radio traffic of enemy ships.

Even then and in the period between the two world wars, Churchill equated such activities with a new type of weapon. Finally, it was clear that it was necessary to classify our own radio communications. And all this had to be kept secret from the enemy. There are great doubts that the leaders of the Third Reich realized all this. In the leadership of the Wehrmacht (OKW) there was a department with a small number of cryptologists and with the task of “developing methods for revealing enemy radio messages,” and we were talking about front-line radio reconnaissance officers, who were charged with providing front-line commanders with tactical information on their sector of the front. In the German army, the encryption machines used were assessed not by cryptologists (in terms of encryption quality and cracking capabilities), but by technical specialists.

The Allies followed the gradual improvement of German encryption technology and also improved methods of breaking encryption codes. The Germans attributed facts indicating the awareness of the Allies to betrayal and espionage. In addition, in the Third Reich there was often no clear subordination, and the encryption services of different branches of the military not only did not interact with each other, but also hid their skills from the cryptographers of other branches of the military, since “competition” was the order of the day. The Germans did not try to unravel the Allied encryption codes, since they had few cryptologists for this, and those that they had worked in isolation from each other. The experience of English cryptologists has shown that the joint work of a large team of cryptologists made it possible to solve almost all the tasks assigned. Towards the end of the war, a gradual transition in the field of encryption began from machine-based work to computer-based work.

Encryption machines in military affairs were first used in Germany in 1926. This prompted Germany's potential adversaries to develop their own encryption and decryption methods. For example, Poland took up this issue, and first it had to develop the theoretical foundations of machine cryptology, since “manual” methods were not suitable for this. A future war would require thousands of radio messages to be deciphered every day. It was Polish specialists who were the first to begin work on machine cryptological analysis in 1930. After the outbreak of war and the occupation of Poland and France, this work was continued by English specialists. The theoretical work of the mathematician A. Turing was especially important here. Beginning in 1942, breaking encryption codes became extremely important, as the German command increasingly used radio communications to transmit its orders. It was necessary to develop completely new methods of cryptological analysis for decryption machines.

Historical reference.
Julius Caesar was the first to use text encryption. In the 9th century, the Arab scholar Al-Kindi first considered the problem of text decipherment. The work of Italian mathematicians of the 15th and 16th centuries was devoted to the development of encryption methods. The first mechanical device was invented in 1786 by a Swedish diplomat; such a device was also at the disposal of the American President Jefferson in 1795. Only in 1922 this device was improved by the American army cryptologist Mauborn. It was used to encrypt tactical messages until the outbreak of World War II. Patents for improving usability (but not for encryption security) were issued by the US Patent Office starting in 1915. All this was supposed to be used to encrypt business correspondence. Despite numerous improvements in devices, it was clear that only short text encryption was reliable.

At the end of the First World War and in the first years after it, several inventions appeared, created by amateurs for whom this was a kind of hobby. Let's name two of them: Hebern and Vernam, both Americans, neither of them, most likely, had ever heard of the science of cryptology. The latter of the two even implemented some operations of Boolean logic, which at that time few people knew about except professional mathematicians. Professional cryptologists began further improving these encryption machines, which made it possible to increase their security against hacking.

Since 1919 German designers also began to patent their developments; one of the first was the future inventor of the Enigma, Arthur Scherbius (1878 - 1929). Four variants of similar machines were developed, but there was no commercial interest in them, probably because the machines were expensive and difficult to maintain. Neither the Navy nor the Ministry of Foreign Affairs accepted the inventor's proposals, so he tried to offer his encryption machine to the civilian sectors of the economy. The army and the Foreign Ministry continued to use encryption using books.

Arthur Scherbius went to work for the company that bought his patent for an encryption machine. This company continued to improve Enigma even after the death of its author. In the second version (Enigma B), the machine was a modified electric typewriter, on one side it was equipped with an encryption device in the form of 4 replaceable rotors. The company widely displayed the machine and advertised it as unhackable. Reichswehr officers became interested in her. The fact is that in 1923, Churchill’s memoirs were published, in which he talked about his cryptological successes. This caused shock among the leadership of the German army. German officers learned that most of their military and diplomatic communications had been deciphered by British and French experts! And that this success was largely determined by the weakness of amateurish encryption, invented by amateur cryptologists, since German military cryptology simply did not exist. Naturally, they began to look for strong encryption methods for military communications. Therefore, they became interested in Enigma.

Enigma had several modifications: A, B, C, etc. Modification C could perform both encryption and decryption of messages; it did not require complex maintenance. But its products were not yet resistant to hacking, because the creators were not advised by professional cryptologists. It was used by the German Navy from 1926 to 1934. The next modification, Enigma D, was also a commercial success. Subsequently, since 1940, it was used in railway transport in the occupied areas of Eastern Europe.
In 1934 The German navy began to use another modification of Enigma I.

It is curious that Polish cryptologists tried to decrypt German radio messages classified by this machine, and the results of this work somehow became known to German intelligence. At first, the Poles were successful, but the German intelligence “watching” them reported this to their cryptologists, and they changed the codes. When it turned out that Polish cryptologists were unable to crack messages encrypted with Enigma -1, the ground forces, the Wehrmacht, also began to use this machine. After some improvement, it was this encryption machine that became the main one in the Second World War. Since 1942, the German submarine fleet adopted the Enigma-4 modification.

Gradually, by July 1944, control over the encryption business passed from the hands of the Wehrmacht to the roof of the SS, the main role here was played by competition between these branches of the armed forces. From the very first days of WWII, the armies of the USA, Sweden, Finland, Norway, Italy and other countries were saturated with encryption machines. In Germany, machine designs are constantly being improved. The main difficulty in this case was caused by the inability to find out whether the enemy was able to decipher texts encrypted by a given machine. Enigma of various modifications was introduced at levels above the division, it continued to be produced after the war (model “Schlüsselkasten 43”) in Chemnitz: in October 1945. 1,000 pieces were produced in January 1946. - already 10,000 pieces!

Telegraph, historical information.
The advent of electric current caused the rapid development of telegraphy, which, not coincidentally, occurred in the 19th century in parallel with industrialization. Driving force were railways, who used the telegraph for the needs of railway traffic, for which all kinds of devices such as pointers were developed. Steinhel's device appeared in 1836, and in 1840 it was developed by Samuel MORSE. Further improvements came in the form of the Siemens and Halske printing telegraph (Siemens & Halske, 1850), which converted received electrical impulses into readable type. And invented in 1855. The printing wheel, after a number of improvements, was still used by Hughes in the 20th century.

The next important invention for accelerating the transfer of information was created in 1867 by Wheatstone: punched tape with Morse code, which the device felt mechanically. Further development of telegraphy was hampered by insufficient use bandwidth wires The first attempt was made by B. Meyer in 1871, but it failed because the different lengths and number of pulses in Morse letters prevented it. But in 1874, the French engineer Emile Baudot managed to solve this problem. This solution became the standard for the next 100 years. Baudot's method had two important features. Firstly, it was the first step towards the use of binary calculus. And secondly, it was the first reliable multi-channel data transmission system.

The further development of telegraphy rested on the need to deliver telegrams using postmen. A different organizational system was required, which would include: a device in every house, its maintenance by special personnel, receiving telegrams without the help of staff, constant connection to the line, issuing texts page by page. Such a device would have prospects of success only in the USA. In Europe, until 1929, the postal monopoly prevented the appearance of any private device for transmitting messages; they had to be installed only at the post office.

The first step in this direction was taken in 1901 by the Australian Donald Murray. In particular, he modified Baudot's code. This modification was the standard until 1931. He did not have commercial success, since he did not dare to patent his invention in the USA. In the USA, two American inventors competed with each other: Howard Krum and E.E. Kleinschmidt. Subsequently, they merged into one company in Chicago, which began producing equipment in 1024, which enjoyed commercial success. The German company Lorenz imported several of their machines, installed them in post offices and obtained a license for their production in Germany. Since 1929, the postal monopoly in Germany was abolished, and private individuals gained access to telegraph channels. The introduction of international standards for telegraph channels in 1931 made it possible to organize telegraph communication with the whole world. The same devices began to be produced in 1927 by Siemens and Halske.

The first person to combine a telegraph with an encryption machine was 27-year-old American Gilbert Vernam, an employee of the ATT company. In 1918 he applied for a patent in which he empirically used Boolean algebra (which, by the way, he had no idea about and which was then being studied by several mathematicians around the world).
The American officer William Friedman made a great contribution to cryptology; he made American encryption machines virtually unbreakable.

When telegraph devices from Siemens and Halske appeared in Germany, the German Navy became interested in them. But its leadership was still under the impression that the British had cracked the German codes and read their messages during the First World War. Therefore, they demanded to connect the telegraph apparatus with a encryption machine. This was a completely new idea at that time, because encryption in Germany was done manually and only then the encrypted texts were transmitted.

In the USA, this requirement was met by Vernam devices. In Germany, the company Siemens and Halske took on this work. They filed the first open patent on this topic in July 1930. By 1932 a workable device was created, which at first was freely sold, but since 1934. was classified. Since 1936 These devices began to be used in aviation, and since 1941. - and ground forces. Since 1942 Machine encryption of radio messages began.

The Germans continued to improve various models of encryption machines, but they put the improvement of the mechanical part in the first place, treating cryptology in an amateurish manner; manufacturing companies did not involve professional cryptologists for consultations. Great importance For all these problems, there were works by the American mathematician Claude Shannon, who was well-read since 1942. worked at Bell Laboratories and conducted secret mathematical research there. Even before the war, he was famous for proving the analogy between Boolean algebra and relay connections in telephony. It was he who discovered the “bit” as a unit of information. After the war, in 1948. Shannon wrote his main work, The Mathematical Theory of Communications. After this he became a professor of mathematics at the university.

Shannon was the first to consider the mathematical model of cryptology and developed the analysis of encrypted texts using information theoretical methods. The fundamental question of his theory is: “How much information does ciphertext contain compared to plaintext?” In 1949 he published the work “The Theory of Communications” secret systems”, in which he answered this question. The analysis carried out there was the first and only to quantify the strength of an encryption method. Post-war analysis showed that neither German nor Japanese encryption machines were unbreakable. In addition, there are other sources of information (for example, intelligence) that greatly simplify the decryption task.

England's position forced it to exchange long cipher texts with the United States; it was the great length that made deciphering them possible. In a special department of the British secret service M 16, a method was developed that increased the degree of secrecy of the message - ROCKEX. The American encryption method for the Foreign Office was broken by German experts and the corresponding messages were decrypted. Having learned about this, the United States in 1944. replaced an imperfect system with a more reliable one. Around the same time, the German Wehrmacht, Navy and Foreign Ministry also exchanged encryption technology for newly developed ones. Soviet encryption methods were also insufficiently reliable, which is why they were hacked by American services and many Soviet intelligence officers involved in espionage for the American atomic bomb were identified (Operation Venona - breaking).

Breaking into.
Now let's talk about the British HACKING German encryption machines, that is, the machine unraveling of the method of encrypting texts in them. . This work received the English name ULTRA. Non-machine decryption methods were too labor-intensive and unacceptable in war conditions. How were the English deciphering machines constructed, without which the Allies could not have achieved an advantage over the German code breakers? What information and textual material did they need? And was there a German mistake here, and if so, why did it happen?

First, the scientific and technical basics.
First, a preliminary scientific work, since it was necessary, first of all, to cryptologically and mathematically analyze the algorithms. This was possible because encryption was widely used by the German Wehrmacht. Such analysis required not only ciphertexts obtained through eavesdropping, but also plaintexts obtained through espionage or theft. In addition, different texts were needed, encrypted in the same way. At the same time, a linguistic analysis of the language of the military and diplomats was carried out. Given long texts, it became possible to mathematically establish an algorithm even for an unfamiliar cipher machine. Then they managed to reconstruct the car.

For this work, the British brought together approximately 10,000 people, including mathematicians, engineers, linguists, translators, military experts, and other employees to sort the data, check it, archive it, and maintain the machines. This association was called BP (Bletchley Park) and was under the personal control of Churchill. The information received turned out to be a powerful weapon in the hands of the Allies.

How did the British master the Wehrmacht Enigma? Poland was the first to decipher German codes. After the First World War, it was in constant military danger from both of its neighbors - Germany and the USSR, who dreamed of regaining the lands lost and transferred to Poland. To avoid surprises, the Poles recorded radio messages and deciphered them. They were greatly alarmed that after the introduction in February 1926. in the German Navy Enigma C, as well as after its introduction into ground forces in July 1928 they were unable to decipher messages encrypted by this machine.

Then the BS4 department of the Polish General Staff assumed that the Germans had acquired machine encryption, especially since they knew the early commercial versions of Enigma. Polish intelligence confirmed that in the Wehrmacht from June 1, 1930. Enigma 1 is used. Polish military experts were unable to decipher German messages. Even having received Enigma documents through their agents, they could not achieve success. They concluded that there was a lack of scientific knowledge. Then they commissioned three mathematicians, one of whom studied in Göttingen, to create a system of analysis. All three received additional training at the University of Poznan and spoke fluent German. They managed to reproduce the Enigma device and create a copy of it in Warsaw. Let us note the outstanding achievements of one of them, the Polish mathematician M. Rejewski (1905 - 1980). Although the Wehrmacht constantly improved the encryption of its messages, Polish specialists succeeded until January 1, 1939. decipher them. After this, the Poles began to cooperate with the allies, to whom they had not previously communicated anything. Such cooperation was already advisable in view of the obvious military danger. July 25, 1939 they conveyed to the English and French representatives all the information they knew. On August 16 of the same year, the Polish “gift” reached England, and English experts from the newly created BP Decoding Center began working with it.

British cryptologists were reduced after the First World War, remaining only under the roof of the Foreign Office. During the war in Spain, the Germans used Enigma D, and the remaining English cryptologists, under the leadership of the outstanding philologist Alfred Dillwyn (1885-1943), continued to work on deciphering German messages. But purely mathematical methods were not enough. By this time, at the end of 1938. Cambridge mathematician Alan Turing was among the visitors to the English cryptographer training courses. He took part in the attacks on Enigma 1. He created an analysis model known as the “Turing machine”, which made it possible to assert that a decryption algorithm definitely exists, all that remained was to discover it!

Thüring was included in the BP as a person liable for military service. By May 1, 1940 he achieved serious success: he took advantage of the fact that every day at 6 o'clock in the morning the German weather service transmitted an encrypted weather forecast. It is clear that it necessarily contained the word "wetter" (Wetter), and that the strict rules of German grammar determined its exact position in the sentence. This allowed him to ultimately come to a solution to the problem of breaking the Enigma, and he created an electromechanical device for this. The idea came to him in early 1940, and in May of the same year, with the help of a group of engineers, such a device was created. The task of decoding was made easier by the fact that the language of German radio messages was simple, expressions and individual words were often repeated. German officers did not know the basics of cryptology, considering it unimportant.

The British military, and especially Churchill personally, demanded constant attention to deciphering messages. Since the summer of 1940 The British deciphered all messages encrypted using Enigma. Nevertheless, English specialists were constantly improving decryption technology. By the end of the war, British codebreakers had 211 decryption devices working around the clock. They were served by 265 mechanics, and 1,675 women were brought on duty. The work of the creators of these machines was appreciated many years later, when they tried to recreate one of them: due to the lack of necessary personnel at that time, the work to recreate famous car lasted several years and remained unfinished!

The instructions for creating decryption devices created by Dühring at that time were banned until 1996... Among the means of decryption was the method of “forced” information: for example, British planes destroyed the pier in the port of Calle, knowing in advance that the German services would report this with a set of information known in advance to the British words! In addition, German services transmitted this message many times, each time encoding it with different codes, but word for word...

Finally, the most important front for England was the submarine war, where the Germans used new modification Enigma M3. The British fleet was able to remove such a vehicle from a captured German submarine. On February 1, 1942, the German Navy switched to using the M4 model. But some German messages, encrypted in the old way, mistakenly contained information about the design features of this new car. This made the task much easier for Thuring's team. Already in December 1942. Enigma M4 was cracked. On December 13, 1942, the British Admiralty received precise data on the location of 12 German submarines in the Atlantic...

According to Turing, to speed up decryption it was necessary to switch to the use of electronics, since electromechanical relay devices did not perform this procedure quickly enough. On November 7, 1942, Turing went to the United States, where, together with a team from Bell Laboratories, he created an apparatus for top-secret negotiations between Churchill and Roosevelt. At the same time, under his leadership, American decryption machines were improved, so that Enigma M4 was finally cracked and until the end of the war it provided the British and Americans with comprehensive intelligence information. Only in November 1944 did the German command have doubts about the reliability of their encryption technology, but this did not lead to any measures...

(Translator's note: Since, starting from 1943, the head of the British counterintelligence was the Soviet intelligence officer Kim Philby, all information immediately came to the USSR! Some of this information was transmitted Soviet Union and officially through the English bureau in Moscow, and also semi-officially through the Soviet resident in Switzerland, Alexander Rado.)

Chiffriermaschinen und Entzifferungsgeräte
im Zweiten Weltkrieg:
Technikgeschichte und informatikhistorische Aspekte
Von der Philosophischen Fakultät der Technischen Universität Chemnitz genehmigte
Dissertation
zur Erlangung des akademischen Grades doctor philosophiae (Dr.phil.)
von Dipl.-Ing.Michael Pröse

The Bombe machine, developed by British mathematician Alan Turing, was of great importance during the Second World War. Turing's invention helped crack coded legendary car Enigma German messages.

The Turing machine significantly increased the speed of decoding intercepted German messages. This allowed Allied forces to respond to classified intelligence within hours rather than weeks.

Much has been said about Turing's genius, his troubled personal life and his tragically early death. Hollywood even made a film about him. But how much do you know about the machine he built, the principle of hacking the machine, and the impact it had on the course of the war?

We share unknown facts about Turing's invention.

1. Turing didn’t come up with his machine himself.

In fact, Turing's ingenious invention, the Bombe machine, is a continuation of the work of Polish mathematicians Marian Rejewski, Henryk Zygalski and Jerzy Rozycki.

Poland's Bombe succeeded thanks to a flaw in German encryption that double-encrypted the first three letters at the beginning of each message, allowing codebreakers to look for patterns.

Exactly how the Bombe machine worked remains a mystery, but by using six of these machines in parallel, the all-important Enigma Ringstellung (the order of the coding ring) could be discovered in just in a couple of hours.

2. The Germans perfected Enigma

At some point, German cryptographers discovered and fixed the weakness of double encryption. Then the British needed a more advanced solution, and Turing and his team got involved in the work.

Using information provided by the Poles, Turing began hacking Enigma messages using his own "computer".

His methods were based on the assumption that every message contained a cheat sheet - a known piece of German plaintext at a familiar location in the message.

In one example it was weather forecast in Atlantic, which was recorded in the same format every day. Location detection equipment at listening stations allowed codebreakers to determine where a message was coming from, and if it matched the location of a weather station, it was likely that the word "wettervorhersage" (weather forecast) would be present in every message.

Another curious clue for Turing was Enigma's inability to encode a letter as itself. That is, S could never have been S.

3. Enigma has become almost perfect

Even taking into account all the disadvantages of Enigma, it was difficult to crack the code almost impossible. There was not enough time or manpower to work out all possible combinations. This is due to the fact that each letter, when entered into the Enigma machine, was encrypted differently each time.

So, even if you guessed one keyword that offered hints, it took reduce the odds 158,962,555,217,826,360,000 to 1– the exact number of ways to configure Enigma machines.

Moreover, every day a new code had to be cracked to account for the Germans changing the settings at midnight.

4. Turing's team did the opposite

Instead of guessing the key, Bombe used logic to reject certain possibilities. As Arthur Conan Doyle said, “When you have eliminated the impossible, whatever remains, no matter how incredible, must be true.”

This method, although successful, still provided a range of possible correct answers for the German ring settings. So more work needed to be done to narrow it down to the right one.

Using testing machine process repeated until the correct answer was found.

This gave the crackers part of the key, but not all of it. Then you had to use what you learned and figure out the rest of the key.

Once the code was cracked, Turing's team would set up an Enigma machine with the correct key of the day and decipher every message intercepted that day.

5. Turing machine today costs 320 million rubles

The "bombs" were 7 feet wide, 6 feet 6 inches tall and literally weighed a ton. They had 12 miles of wires(!) and 97,000 different parts.

The decoder prototype was built at a cost of £100,000, equivalent to around £4 million today. Almost 320 million rubles at the current exchange rate!

In essence, the Turing bomb was an electromechanical machine consisting of 36 different Enigma machines, each containing the exact internal wiring of the German equivalent.

When "Bomb" is turned on, each of the riddles is allocated a pair of letters from the resulting cheat sheet text (for example, when D becomes T in a guessed word).

Each of the three rotors moves at a speed simulating the Enigma itself, testing approximately 17,500 possible positions until a match is found.

6. Turing's genius influenced the outcome of the war

After the Enigma machine was broken, 211 Bombe machines were built and operated around the clock. They were placed in various locations throughout Britain in case of possible explosions that could destroy these very complex and expensive samples.

Due to a shortage of captured Enigma machines, British Typex cipher machines were converted to working Enigma machines.

The fully decrypted messages were translated from German into English and then passed on to British intelligence.

At its peak, the Bombe machine could hack up to 3000 German messages per day. By the end of the war, she had handled 2.5 million messages, many of which gave the Allies vital information about German positions and strategy.
It is assumed that this knowledge played an important role in key battles.

According to many experts, Turing's invention shortened the war by two years.

Bank of England to issue £50 note in Turing's honor

If you find an error, please select a piece of text and press Ctrl+Enter.