D-Day and the role of Colossus

back in May of this year we pulled together some the veterans the surviving Colossus veterans to come along and to learns a bit more about the work they had been doing during the war because of course they had to keep this extraordinarily secret not until the 1970s was there any information about Colossus not and and even then wasn't quite sure what that machine was doing then in the early 2000s the general report on Tunney started to reveal a lot more of the story and these Ren's these veterans who have just seen in the video were quite startled did to learn more and more about what they'd actually been doing because they had been kept very quiet all those decades and and suddenly they were learning what they were doing and suddenly they were able to talk about it but it's always been a mystery exactly what the information was and it was so difficult to get information about what Colossus was doing and what enigma was doing and so on and must say it's thanks to John at gely Hawk a volunteer at the Museum who's been ill recently but is recovering well I believe I said to John John is there any information we can get abide the Colossus decrypt what is there anything been overlooked and he a regular visitor to the National Archives at Kew started looking in greater depth and led us on the trail that has led to these talks today and so we're very grateful indeed for Mark Priestley who will speak first and Gavin Clarke who speaks second and they will tell us a lot more than we ever knew about Colossus and this activity are indeed a quite sure there's a lot more to commit when if it'll ever come out who knows but we're getting more information now than we've ever had before and Gavin in fact put together with the information from mark himself some a display board in the in the closest gallery about d-day so you haven't seen that very definitely go and see that very nicely presented information indeed so I'd like to thank them very much indeed for coming they to give a repeat of their talks with some developments I believe and I'd like to hand over straight to mark mark Priestley okay thank you Steve with that introduction and thanks to the invitation to talk this afternoon I'm sort of I'm sort of a historian of computing and for the last few years I've been involved with a colleague in a sort of project based around Colossus and so I've made you know several visits to the National Archives to look in the archive material there around the whole this issue of Colossus and what it was used for in the war so when this d-day project came up in John Gallin walk told us that he'd come across this new information the information that's in the National Archives but there's a lot of information in the National Archives and this was a particular file that I hadn't come across before that I was kind of fascinated and the opportunity to sort of dig much more in much more depth into in particular what Colossus was used for up until that point the focus had been very much on Colossus the machine how it was built how it worked have a replica have the replica over the over the way that was built this this this this project leading up to d-day he gave me a chance to look at the actual material the output the effects of Colossus is decodes and the and how that was used in the wider context so the title is using the fish decodes now the word fish wasn't used in in the internet in the videos or in the introduction talk fish was the kind of Bletchley Bletchley Park named for the German codes that used this Lorentz encryption device so they used an old-fashioned telly types and one remembers those and it was an attachment on these devices which encrypted the messages that the user would type it would automatically encrypt be sent usually by radio so it could be intercepted by the people in the UK and then would be automatically decrypted by an equivalent machine on the receiving teletype so the Germans used it to connect their kind of army commands it was their kind of high alert it was a higher-level network than the Enigma network it's less well known but a similar level an even higher level of security and some even more sort of significant strategic messages going across it the fish codes were first broken by Bletchley Park in 1942 and they carried on reading these messages till the end of the war and the story is really one of the Germans changing things and the Brits trying to keep up so in 1942 when they first broke it all the decription was done manually there were no mushy Colossus wasn't built until the beginning Colossus didn't come into service until the beginning in 1944 so in 1942 they were reading quite a lot of these fish messages but decrypting them manually and towards the end of 1942 the Germans made a change which meant there was it was impossible to do it by hand it was just too complicated too many things to consider so the so the people of Bletchley Park began a program of mechanization the planning and building machines it would help in a particular attack on this code the first machine was the one called Heath Robinson and there's a replica again over in the other building that we can now um see and I believe it's now in more or less in working order and then by the end of 1943 beginning 1944 the first Colossus machine came into operation and then by the beginning of 1944 there was new Colossus machines of a sort of improved mark to came into service about once a month so by the end of the war though I think ten colossi in operation it's important to realize I think that these the Colossus these machines and particularly the Colossus they weren't actually decrypting machines the the the project of decoding and decrypting these messages involved a lot of kind of statistical work a lot of linguistic expertise you got to know German you had to know a lot of the properties of the you know how frequently the letters occurred in the German texts that the Germans were using for example so there's a lot a lot of statistics there was still a lot of manual work done that people linguistic experts reading the German understanding the German you know making breaks in various ways but the the machines the Colossus machines were absolutely central in that they've just provided the huge resource of computational power that allowed the codebreakers to make the initial break into these messages okay so without the machines I mean this from you know from the end of 1942 onwards the black sheep art would only be not to break a really insignificant a tiny tiny number of these were fish messages if it hadn't been for the machines but it's important to remember that there was a lot more to the fish section the people who did this work than just the machines the machines were crucial magnificent creations an absolutely crucial but not the whole story there was a fish section they called themselves a fish section the people based in in these buildings can around here who worked on decoding the fish codes the Lawrence encrypted messages roughly speaking there were two parts to this you may've heard in the videos people refer to the tester II that was a section named after its head major test that major Ralph's tester about whom very little is known eating - I think he worked in some in business as an accountant or something before the war but he turned out to have some sort of genius for managing a cryptological organization and then after the war he and he disappeared back into obscurity I mean I've looked but you know very the very little biographical information about about him but he was in charge of the linguistic section so he was in charge of like german-speaking experts mathematicians chess players who had that kind of crossword type of mind who could work with these like almost like puzzle solving who could work with these codes the other half of the fish section was called the neumann Airy named after its leader max Newman who's much better known he was a Cambridge mathematician it was a kind of mentor about ensuring before the war and it was involved in one of the early British commuter projects after the war at the University of Manchester and he was seems to pin his idea to use machines at the end of 1942 to build these machines to help in the initial break into these messages so and he's much better known than Ralph testa but I mean they were the two people in charge of the fish section and they were they were calm they were very complimentary and the work these two sections did was very complimentary but between them by the by by setting by the middle of nineteenth by late middle at the end of 1944 between them they were breaking most of breaking into these fish messages on a very regular basis was like a kind of factory a kind of decryption factory what so what I want to do in this talk really is say a little bit about the output from that process so I'm going to treat the fish section itself as a kind of black box and then what what output did they produce what happened to it and how did their work how they work guided and how does it fit into the larger the larger Bletchley Park context so in 1942 at the beginning when the they first broke into the fish message this is the kind of organization chart that was going on on the very left and the green box we've got the people doing the interception of the radio signals so most of these messages were sent by radio and by putting up big enough aerials on the South towns you could intercept some of these messages and in 1941 they began to realize it was a new type of message was appearing and it turned out to be very important and then no cult of anyone knows knock halt on the South Downs they over the next couple of years they built a large interception substation there which basically was a lot of a lot of rather grumpy people who spent their war doing very boring very boring difficult work stuck in offices listening just to it just listening to just a noise coming over the earphones and then trying to write down you know some interpretation of what this meant incredible a soul destroying job it must have been and there's fascinating stuff in the archives about the difficulty of you know motivating these people and getting the staff and motivating for this really intense difficult boring work to make it worse as the war went on the Germans made various changes in order to decrypt the messages the Bletchley Park people needed longer and longer messages more and more accurate so these people at the intercept people were not allowed even a single character wrong would would sort of make MIDI there was no longer possible to decode that message so the pressures under which these people were working were incredible anyway they produce these intercepts as they were called which is basically just the a string of characters a list of a meaningless list of letters and numbers the the encrypted version of the the German plaintext message that they were trying to recapture these messages were delivered to the fish section here at Bletchley Park and with a combination of machines human intelligence and all the rest of it they decoded a certain number of these not or not all of them by any means they decoded a certain number of messages and then these the the decodes were then put into the Bletchley Park system to be circulated to other parts of the war effort now so that's from the the right-hand half of this slide which says Hut three right is kind of distribution of the messages produced by the fish section and by and large it was quite fortunate that all that mechanism kind of existed already because what happened to the fish messages because they were a very very high level German code as was the Enigma code the Enigma in terms of handling the information handling the intelligence the Enigma and fish messages were kind of grouped together and by that stage in 1942 Bletchley Park had already been handling enigma intelligence for a couple of years so the fish stuff racing g-scan slotted into the same system of distribution I mean there were there are some teething problems and some differences but by and large there was there was a there was an organization already there when they started breaking fish that would allow the messages to be decoded so within Bletchley Park the messages went to three different places so in the test theory the linguistic part of the fish section there were some German speaking experts who read the decodes and worked out and decided to send them to one of three places so going from the bottom up ice sauce is rather curious Bletchley Park acronym stands for something like intelligence services Oliver Strait she and all of us straight she was brother I think of the the more famous Lytton Strachey and he ended up in Bletchley Park as a code breaker and he headed that section originally so OS there stands for a person's name by 1942 he'd long gone and son else was in charge of the system but in the the eccentricity of Bletchley Park they carried on using the same acronym all the way through so and what what that section did was deal with intelligence that was to do with the app there which was the German Army Secret Service I think I mean probably some of you know more about the ins and outs of the German military than I do but it was basically German secret service messages which passed across the fish network and they were handled within actually Park all messages to do with the app there were handled by this ISOs section so some of the decodes went went there and I can show you an example of one of those so there's an example from the from the National Archives there's an example of one of these messages with Secret Service intelligence being sent from the fish section to ISOs who then decided you know they play them as you can see from the distribution list there they would then pass it on to mi5 mi6 the Department of Naval Intelligence anybody else so they thought might be interested I'm not sure if you can read that terribly carry but it's it's quite a nice one it's the Germans planning to send the spy to America in a submarine and they're very worried that this poor spy by time he gets to America will be so kind of shattered by the experience you'll be kind of completely useless for any kind of undercover work in America so they kind of I think they decided against it with it son it's quite it's quite a funny quite a fun message so that was the first category of decode Secret Service information that went to Isis the second section the second of the middle one naval section was messages it had a particular relevance to naval naval operations by and large the fish networks there was a diagram of the network and one of the videos and I think Gavin will show us a more detailed Network later but by and large this thieve the fish Network the lorentz network connected Army Headquarters and so there weren't too many naval messages naval had good naval commands tended to communicate my different means predominately enigma the naval enigma but you did get some naval messages going across the fish network within Bletchley Park hut-three dealt with intelligence for the army basically and the Luftwaffe and there was a separate thing I think it was hutch six which dealt with the naval messages called the naval section so the second thing that the people in the testtree were looking for when they had a new decode was if it was to do with naval matters they would send it to the naval section who would then deal with it and then more often than not pass it on to the Admiralty we haven't yet managed to track down any kind of direct evidence of this in terms of messages so I can't show you an example of one of these messages but the it did happen but the bulk of the decodes were information to do with the German army and the German air force and hut-three was the grouping within Bletchley Park that dealt with that kind of information when it was produced from enigma messages and the fish messages were just fed into this same decoding this is the same intelligence distribution mechanism so HUD 3 did a number of things with these messages they were in communication with two groups of people so the top two on the right hand side now the orange boxes are they kind of recipients of this intelligence the top two boxes are actual army commands people in the field so well active commands are intelligence operation that tells it's organisations attached to army commands in the field and the shape shape the supreme headquarters Allied expeditionary force from late 1943 onwards was discovered a different sort of thing again they got various bits of information from r3 and they also sent information to the various ministries in London the Air Ministry the walk was dealt with the Air Force obviously the War Office which dealt with the army and the Admiralty which don't give the Navy and hut-three also sent material to you know mi6 so they so they were sending material to London and also to active commands in the field and that system of distribution was worked out well in advance of the fish messages being decoded a first time in 1942 so we've got some examples of that and it was they had three ways of distribution and really what they did was make their decision based on how urgent they thought the information was so this is hut-three now they had teams of people who would be reading these decodes and then deciding what best to do with them actually the first thing to point out probably is that when it just says decodes on that slide there and then the one I showed you has got a nice nice readable text that says you know points out to the meais the mere strain of a passage to the USA under the present it's it reads like a proper piece of English that you can just read and understand what's being said that's not magic the the German the actual German coded messages tended not to be such nice readable bits of prose they were much more kind of abbreviated much more terse all sorts of military abbreviations because basically it cost money every character sent across for the radio waves cost money and time and increase the risk of encryption so the Germans kept their messages down to a kind of as short as possible by using terse language and abbreviations so the first thing that hut-three had to do was take these very hard to understand messages and translate them and into something that people at the air ministries and people in the army would be able to understand and read much more easily and get the intelligence value from it so so the real job that hut-three was doing was taking these really terse messages giving us some context and then producing output that would be meaningful to people in the ministries and people in the army they did this by means of a vast card index so all the messages they had a huge card index of information so everybody every every German soldier that was mentioned by name in a message had a card every army unit had a card practically every submarine had a card yeah they took this extraordinary fund of information so that for example when a message came in and said you know Colonel Prince a good German name here Colonel Hitler Colonel Vaisman you know is being shipped to Kiev or something like this the people in HUP 3 would go off and look at their card index for colonel zeisman and discover what he'd been doing for the last year and produce a little mint in some cases produce a little mini biography and say well okay we we know that he was associated with such and such an army group so this is evidence that they are going to Kiev as well and they just drew the most extraordinary inferences from these really terse messages filled filled out all that kind of intelligence into these the transcripts that we've now got in the in the archives which are much more readable and you know something that you can imagine a stressed intelligence officer you know Army headquarters in Sicily actually get actually appealed to make some kind of sense of there are very very few as far as I'm aware there are very very few actual raw decodes in existence I don't believe there are any in the National Archives who's been reproduced in books and people are very cagey about saying where they are we're here so secrecy secrecy still lives anyway so HUD 3 they did this thing called it was they called it amending which was translating the the the terse German message into something comprehensible and useful for the for the Armed Services and then depending how important they made then made a judgement because I mean they actually knew more about the conduct of the war than anyone they were getting all this information from enigma and fish was going through Hut 3 I mean they they they they really knew a lot about the progress of the war it could make judgments about what was input strategically important what was tactically important what you know what some what what some commander needed to know right now and what what it would be good enough just to send to the Ministry and it would get forwarded on the week's time so they made that sort of judgment about priorities and they did one of three things if something was really important they would send a signal to an active command this would be a copies of these in the National Archives and they look like this I'm not sure that's a terribly interesting one numerous items are its yes England numerous items of information about the alleged postponement of invasion or about its complete abandonment in favor of intensified air warfare and small-scale local undertakings are to be considered as planned cover for the actual intentions so this is the Germans saying you know we don't believe we're not we're not we're not being tricked by these Brits trying to pretend they're responding the invasion Bletchley Park is translating that and sending that back to the active command saying you need to know this this is what the Germans are thinking so these went out in the form of you know active live signals to the to the to the commands the next level of importance down they would send what we'll call tele prints to the ministries and these included copies of the signals so the signals that went to the commands the ministries got copies of those so they would know what the Army's the Armed Forces they were responsible for they would know what they were being told they got copies of the signals they also got additional information they've got a slightly different translation of the signal because sometimes the the people in the field needs to know a certain bit of information but the actual decode had wider significance so the ministry's got the full translation where sometimes the people in the field just got the details I needed to know the ministry has got all the extra context and the extra information and they would be Bletchley Park would tell them that we've this is the following we've sent the following signals to you know such-and-such armed commands stuff that was less important it was typed up every day in a sometimes quite a long report of less urgent material and this again was sent to the to the ministries and it's it's it's this this tends to be kind of quite boring stuff source saw a note of 4th of May signed Pohlman oberleutnant and adjutant just some details about personnel and but remember the personnel details may sound very boring but gradually they built up this extraordinary picture about in a way the German military were moving to and from that you could deduce information about you know German troop movements which are obviously very important for the army noticed this thing it says source throughout this entire process the Bletchley Park made their well in conjunction with mi5 and mi6 maintained this fiction that there was a very highly placed spy somewhere in the German High Command that they report they they had various names for him throughout the war there's this this mysterious person so often they say source and it's that keeping up this fiction that there was an actual person who was providing this and for more than one I mean there's so much information that presume nobody believed it was just one one spy but there were they kept at this fiction that the word people providing this information there were a couple reasons for this one that one of them was obviously security that if the German they didn't want people to know that we were intercepting radio transmission because then the Germans would take precautionary measures to you know change the change of frequencies more often or whatever you can whatever you can do so it's funny you know if if the Germans got information that we were breaking these messages or this information was available it will be better if the Germans believed it came from a spy not from radio interception so that's obviously a security issue there but there's also some there's also some documents which suggest that people just thought it was in bad form to listen in to the enemy like this yeah it's there is a kind of extraordinary messages I mean people might not approve of what we're doing if they realize we're still eavesdropping on their radio communications because that's private correspondence I think come on guys you know this is but yeah because you need to remember that I guess that this radio transmission like this was still quite new it was any I think to what you know I think the First World War was the first time I'm not even sure a radio radio transmission was used in active active you know encryption in the First World War but the whole idea that you could do this was maybe 20 years old max and there's protocols about manners we're still still surprisingly surprisingly strong so all these messages start off about source and Churchill Churchill that I forgotten the name so they won at the beginning of the war they actually had a name for this person that Churchill got very attached to and carried on using the name for this fictitious spiral through the war but it's just coming out of my brain I'm afraid to say mm-hmm Boniface that's right yeah Boniface they checked yeah so throughout the war Churchill's using it referring to bani faces the the mythical spies producing all this information okay so john gala hawk discovered these teleprinter the National Archives so that's these ones the signals have been known about historians have worked with the actual some of the signals for a long time because quite a long time ago the National Archives released the Admiralty files which contained a lot of these messages and historians have worked from those quite a bit later on they reduced they produce these tele prints which contains the same information but with extra information as well and the best of my knowledge these tele prints have not really been used in historical writing historical research based on this yet which is the John garlic John Gallin was great discovery and the route one of the really cool things about these tele prints is at the bottom of them they have they've they're filed what-what-what's preserved in the National Archives is a hut-three files for these things so there's like thousands of these books up getting on 4,000 of these things about this thick full of these tele prints and it's a mixture of enigma and fish messages but rather helpfully someone in the hut three at the bottom of each of these has written this as written by and because he didn't want this information going to the gang outside Bletchley Park but it was written by hand what where the messages were came from so this says time of omission 9:00 10:00 in the morning 20th of March 1944 the frequency and then jellyfish which is the name of the code the British code Bletchley Park code name for one of them one of the links in the fish Network and then a message number message number seven that they decoded on the jellyfish link and the best of my knowledge nobody's really sort of picked up on the fact that if you go through these files which is partly very boring and partly quite fun you make a count then you can make a count of exactly how many messages to work which again nobody I don't think anyone's done this done this before and so it's been possible to compile statistics on how many of these fish messages were decoded and there's a great here's a graph showing how many messages were how many fish messages were decoded in the first six months of 1944 I either six months leading up to d-day so let me start with this is a slightly complicated diagram first thing to note is that there's a big dip at the beginning you airy and this is because the Germans made some technical alterations to the way that they encrypted the messages which made it much harder for Bletchley Park to decode them and in fact if it hadn't been for a course disappearing just at that same time by chance things have been much worse I mean it the the Brits understood what the change was and what it meant but the fact that it made the the decryption of the messages much more kind of labor-intensive or the initial break-in much more labor-intensive and I mean for Colossus and it's additional computational power this would have been much a much more serious interruption to service and it was all it was a fairly serious interruption to service it would have been much worse without Colossus so the dip there at the beginning is this change and then the gradual increase is the first courses coming into service and they're getting used to using Colossus effectively and using it on these this new type of encryption that the Germans were using but the same the same that's just a slightly different way of using the encryption and then closest to and came into operation at beginning of July and then from then on the graphs go up but this this is just showing the first six months so the that line there the kind of purple line is the number of messages that were decodable the blue line is the total number of messages that the people are not Holt actually intercepted and you'll notice that there's many more of them what makes a message decodable well it's a lot of its to do with link in order the the British break into these messages was statistical and to get statistics to work you need to have quite a lot of data to begin with so long messages are good some of the messages that were interrupted with is too short that they wouldn't have been it wouldn't have been practical to work out how to decode them to work to work on decoding them other messages I guess probably that the reception was so bad that they couldn't rely they couldn't transcribe the encrypted message accurately enough I mean Bletchley Park really needed every single character had to be correct if this decryption with Colossus was going to work so a combination of short messages and poor reception meant that you know getting on for well at least but somewhat in fifteen seventy five percent of the messages that knock halt intercepted were not in principle even accessible for decoding of those messages that were available for decoding again it's something like maybe a third on average so much in a third and a half were actually successfully decoded by Bletchley Park by the fish section and in this period they said the Green Line there is the total number of decodes and there's a site there's a slight thing about differences so the Green Line is a number that I've generated by going through these files in the archives actually counting how many messages were decoded the purple line is taken from a black sea park summary at the end of the war saying only messages as well but it but you know there's this it's a very small fraction of the available messages that they're actually successfully managing to decode the blue line here in the red line are the two major links that they were decoding at that period the blue one is called bream which connected Berlin and Rome so that was like a really important link when the Italian campaign was going on the jellyfish link connected Berlin in Paris so as you're building up to the invasion of Normandy any messages passing between Paris and Berlin are important important obviously crucially important as well so in this period those are the two links that they were really concentrating on on trying to read breem was more productive than than jellyfish but the point is that the number of decodes they might reproduce it was really a small fraction of the total number available so how did they decide which ones to work on I didn't you know you've you've got this vast mass of material how do you decide which ones you're going to work on and you can't just pick a message at random you can't look at it and say oh that looks like an interesting message because it's encrypted you've no idea what the message says so how do you how do you how do you make a choice among this massive material how do you focus your effort so you get the best value the best value intelligent or maximize your chances of getting valuable intelligence out of this decryption effort and they recognized this was a problem so here's a here's a note that someone wrote in November 1943 I can't I've seen this signature before like I can't remember whose it is I'm afraid I just added this slide a couple of days ago but it's addressed at the highest level to the deputy director of Bletchley Park and it says some intelligence guide guidance is very desirable in order that interception of fish traffic may be as efficient as possible I understand that Hut 3-page who's the guy in charge of Isis now and the naval section are all concerned so you know the recipients this intelligence are expressing concern that they're not getting the best kind of value out of the the the the intelligence that's coming from the fish section and I suggest that Urso is a man in hut-three has considerable experience of intelligence grading this is working out what's the best intelligence and where to get it the others be asked to collaborate with him to produce a fish grading which will be given to tester who with Newman the two people in the fish section will adapt it to Krypto needs and communicate it to Kenworthy Kenworthy is a guy in charge of the interception so so this this this this little note here mostly names all the people involved in my my first slide all the heads of the various sections and there's a sense that they get their getting together here to sort of really focus on how to get the best value out of the decryption effort the description resources they've got available for fish how can they get the best value out of that the black sheep are people that always work closely with Kenworthy and the intercept people but this this represents I think a ramping up of the collaboration between these various groups and this worked in two ways first of all they set priorities and they called it prioritization this is something that hut-three worked on so hot three remote the people who translated and then passed on the intelligence in conjunction with the fish section they set priorities and this priorities were based on a number of things sort of like they've got the various links Rome bream Berlin bream so the messages went in two directions and they used you know there were different characteristics on the two directions of these links that may want to that could make one of them easy to decrypt and one of them harder so they don't just talk about bream they talk about Rome bream Berlin bream sometimes so this this is a list of priorities issued by hut-three in April 1944 couple months before d-day telling the fish section what they should concentrate on first of all concentrate on messages from Rome then concentrate on the jellyfish messages then from Berlin and old jellyfish messages that you haven't translated yet and then some other things less important ones it may seem slightly unusual in the build-up to d-day they're saying pay more attention to messages from Rome than from Paris but on the other hand remember first of all that the Italian campaign is still going on and also that one of the things that determine these priorities were how easy it was to break into the message it's a kind of trade-off between how easy that the decryption is and how valuable the intelligence is balancing up those two considerations give from this priority link traditionally the Rome and mean by far away the easiest link to break it was easier to read messages sent from Rome to Berlin than it was to read messages sent from Berlin back to Rome so that's where that comes first jellyfish I think was kind of harder to read then there are Finnish there are many fewer jellyfish messages being translated than bream ones I said jellyfish was harder to break but because its importance makes it up to position two in the league table at that point so the hut three people get together with testa and Newman and make these priority lists on a pretty much on a weekly basis and these these are passed back to the fish section to say these are the messages these the links you should concentrate on in addition to that there was a an activity called prediction which was done this is something now new this was this was I'll show you the overall fish so this was there's hut-3 doing the priority setting and feeding back the information to the fish section saying this is what you should concentrate on the predictions were done by a different group known as sixta and the TA at the end of six stands for traffic analysis which is a technique for getting as much information as you can out of German messages before they are decrypted so you can find things like you can work out where they're coming from for example where they wing sent from is that like to be interesting or not interesting is it long is it short those are kind of really obvious things that you can get for a message but they they also manage to I don't know the details of how they did this but they also managed to get much more detailed information out of these undie cryptid messages and you might have thought feasible and they fed back this information as predictions again to the up to the fish section people saying these again focus on these messages so in that there's this effort started up in January 44 so again it's part of this it's part of this pre d-day activity of ramping up the interest in fish before d-day and this must be a consequence of the consequence of this this note here I think injera 1944 special traffic analysis section was set up to study non Morse networks non Morse is just what they mean by by fish the runts messages and found out that traffic analysis could perform the same services here as in other fields owing to shortages staff it concentrated in helping cryptography and left to the intercept station the task of controlling his own interception so you know there's all sorts of things here about you know Bletchley Park balancing resources but they're doing traffic analysis on the on the and the encrypted messages and giving predictions to cut the fish section on what to concentrate on the bottom document here again dated from April 1944 just shows you some of the things that they could some of the things they could tell so enough to come around here to read it so it says four sets of predictions are made daily by sixth er after a study of bream logs so the logs are the information coming from the intercept station about just the messages how long they are what time they were sent where they were coming from and they're used as decoding priorities by major tests this section okay so the 6-2 people get information about transmission from knock halt and set priorities they pass on to major tester and they've so there's there's four things here limbo / find I believe finds is the German word for enemy so the for some reason they could they could pick up messages I had this particular German word in it I don't know how Panzer lings who knows messages with double serial receipts so that there's a categories message where it's got a double serial number I said that I don't quite I don't fully research the details of what these things mean yet but it's just picking up its ways of picking out particular messages from the unencrypted text to say this one looks interesting messages was observable priority marking so the Germans might have put a priority marking on the encrypted message which is visible even though you haven't um decrypted it so by 1940 format the earlier diagram showed you the the 1942 set situation when the fish section just got started this diagram shows you in the first half of 1944 when it was becoming much more integrated in with the other work of Bletchley Park it's the same diagram except we've now got hut-three feeding back priorities we've got the intercept station at NOC halt sending logs of transmissions to this 60 group who would then base predictions on that and again feedback some more information to the fish section giving them really quick by this stage in the war quite a lot of guidance on which messages they could focus on that were more likely to give valuable intelligence every worthwhile spending some of the scarce decryption resource on ok and so hopefully that's been of some interest and give me some insight into it's not just about courses and decryption but this is how how that fitted into the much bigger environment of Bletchley Park thank you very much thank you very much indeed Marc fascinating stuff just to see what was going on behind the scenes I didn't hear that back in May because I was running around doing all sorts of things so absolutely amazing I know I Gavin Clark is going to tell us about what was actually happening on the grind as a result of these messages and invasion thank you I'm gonna start with a photograph and a number a very big number which you can see here and first all before I starting to thank Mark for a really great presentation as an amazing insight and thank you all for coming today on a Saturday afternoon a gorgeous autumnal afternoon as well this number was quite interesting this picture that's actually photograph by Frank Capra photographer you might be familiar with and it's the only photograph he got the d-day landings and it was it he messed it up as well they're actually developer messed it up he got a lot of pictures the developer messed up in processing and that's the end result but he captures perfectly the chaos of that day 75 years ago the drama the dangers all summed up on one photograph that number was the number the Allies thought expected they were not taking casualties 40,000 that's not just dead they've been Clues injured and dead overall they're expecting 40,000 casualties on that one day taking the beaches of Normandy as it turned out they didn't and the reason they didn't is the reason we kind of here we're here today it was because they had a lot of information on this side now I worked on they're as stephenson I worked the the anniver the display in the closest gallery on the 74th anniversary and we went for a lot of information and through that we discovered the pivotal role that Colossus played in making sure that their number didn't actually hit 40,000 now as you pretty well where you another museum very well the story of Colossus come over shadow I think of the public world by by enigma but I think we have to concentrate on Colossus there were two main ways that Colossus was pivotal and why that number didn't hit 40,000 great extent and that was because it provided the the work you just heard about provided a lot of the pre-invasion intelligence that enabled them the allies in general to build up a very comprehensive very rich picture of what the forces were arrayed against them so they knew kind of what to expect and the second thing was it gave them the confidence I suppose as a result of that they knew to proceed with their plan they knew they were kind of pushing in the right direction so before we kind of go any further we have to kind of look at what was that plan and that's what what I'm kind of gonna get into first of all so at a high level what was d-day you know Operation Overlord was d-day was to plan to send 150,000 troops into Normandy that's beach landings and that's airdrops as well the plan was to link up these bridge heads and to progressively you know liberate France and liberate Europe drive to Berlin etc there will be over it's all very simple obviously it's very simple and paper apart from this guy he wanted to obviously stop that Adolf Hitler had other ideas he knew at some point in the invade he'd been concentrating he hit the war in the East at this point he knew invasion was bound to come at some point in the West question was when so around 1943 he began he's out the order this is a Hitler Fuhrer directive number 51 he often put our Fuehrer directives which gave great statement orders he let other people fill out the details and the result of Fuehrer directive number 51 was the massive reinforcement of the defense's along the Atlantic which near the Atlantic wall which we kind of see as the North Sea in the English Channel so they expanded they reinforced the Atlantic wall with 10,000 more positions they expanded the amount of troop joints as well 30 divisions Williams were 33,000 kilometer fronts they got very dense those troops were trained they were made to go on the alert look out for this forthcoming evasion which they didn't quite know was gonna come now obviously we we kind of alluded to this picture earlier on obviously North asleep but the German military at the time had a very vast and complex communications network which Mark alluded to there this is the fish Network well that's what they called it here to think the Germans didn't call it the fish network the Crypt offers here and the intelligence they called it the fish Network connected the whole of Europe as you can see and through this when you know orders situation reports observations instructions on tank movements troop movements daily food rations bullet allocation everything went through this network so again if you could crack that you've got a great insight to what was going on in German military machine this was highly encrypted by the Lorenz device which you I think we're all familiar with and the particular X section we were interesting but again I think he might have middle is that a bit there between Paris and Berlin in the context of d-day that's because this connects the whole of Europe where we interested in just in in the Paris Berlin section here and now Colossus was sent to attack that section with jellyfish it was the people who are working fish section were looking at that section and they were tired attack the codes that Lorenz was generating these fiendishly complicated codes now who would have received that and again mark alluded to some of some of that's put some faces on who was receiving this information Eddie the whole of the Allied political and military command tree were receiving this information from the highest level to the deepest depths so you've got your general eisenhower commander a shave fill Marshal Montgomery come on the bridge British 21st Army also in charge of ground troops on d-day and weeson Churchill who sure we're all familiar with he received the entire a select number of the intelligence reports but there are so many more people got it as well the spies mi6 would receive that information tactical air units on d-day the commanders in the field would receive us information as well so from the highest to the lowest were receiving this information and they were receiving the information to into kind of ways they were receiving the as mark alluded to the raw decrease that were coming out from the work associate applause also this complex intelligence analysis about what these situation as men now before we go any further we've got to talk about we've must discuss inflatable tanks and players now and this is where Colossus really comes into its own the Allies decided they need to deceive the Germans you can't build up a vast army in Britain without going unnoticed so they can put the energy to get this massive deception campaign cut operation of fortitude and it involved being create the creation of a fake army which was the u.s. first Army Group which was based in the South East of England and was under the command of General Patton it had its own fake radio communications traffic that's own call signals its own staff and signior and everything and it went on its own fake army maneuvers as well I mean army units were actually based in the west of Britain but having maneuvers in Dover and that all that information was going back to the Germans Germans picking up all this stuff and it had it had a point all this stuff it was to deceive the Germans who'd be listening in to and watching into not knowing when the invasion was can take place where it's going to take place and the size of the force they'd be facing that was the whole point of this these kind of inflatable tanks and planes and that's where Colossus that's where the D Crips came in because listening to that they understood that the Germans were taking this information and believing it to a great extent so what did they get so reading this this information these decrypts that we that we were lucky enough to read but you can you can see the the material in the gallery which we worked on what what what kind of size force they're gonna face well the Germans were taking low flying aircraft over the country they were taking and sort of visual counts on what they saw building up they expected an allied force of about 20 divisions and what they eventually faced On June 6 1944 was it was eight divisions vastly smaller now you might think that's okay the balance is in the Germans favor there but actually it wasn't because when Normandy happened they thought it was diversion they were expecting the bulk of the troops are still going to land in the north somewhere so they kind of it was vital that they kind of they didn't know how big this forces gonna be they were going to face they didn't know when it was gonna take place either if you read these communications the the German command is really in the dark they're making the the information for their own spies and from of their reading is essentially there's there's nine tangible and there's no there's no obvious dates popping out of them their own spies are saying kind of could be the state could be that date and the best they got it down to it would be the first half of May which was we all know wasn't the case and they were making its informed conclusions about what time of day would be based on tide table of consulting tide tables but when it might be so they were thinking I've been in coming tide before the dawn sometime all the things you might think I mean it went to the such extent that Field Marshal Rommel who is the one of their famous more famous commanders he was in charge of the German defenses in the Normandy area on June the 5th he left his post he went back home to Germany to visit his wife so he wasn't in position when the Allies landed on June the 6th that's how little they knew about when it would actually take place nowhere with the invasion take place again reading the communications you see the German military had no idea where it was going to get them as gonna go names keep coming and going and was kind of guessing I think it could be here it could be there they thought it's me across a broad front they really thought it was gonna be the the north of France around that sort política pas de calais kind of area so contributing to that was of course this resistance of this ghost army which is which isn't based at all intents and purposes in the southeast they assumed it would be the narrowest possible sea crossing which obviously surged over Calais so they were thinking the invasion would take place there but other things kind of came to mind if you're assembling 20 divisions you they thought well the odds are going to need deepwater port to land that stuff so they drop a list of French harbors and ports they thought would be suitable to invasion say they thought love Dover sorry Calais and Blouin would be the kind of the ones that go for Rommel kind of did float the idea he thought he looked at he was the one who seriously looked at beach landings in Normandy but it wasn't taken seriously at any higher level so again they were kind of off the mark there so that's kind of three buckets that's that's where the deception really played off and listening to this information that was the chatter on on the jellyfish net we're really enabled the Allies to the confidence and the reassurance to know that they could they could proceed with the plan which was to target in Normandy they could target they could land in June and despite the massive German preparations that he knew going on they knew they could still catch the Germans out but that was only a snapshot they got vastly more information the Allies taking this Colossus information looked vastly more insight into what was going on and it kind of breaks down in three three kind of areas buckets of in for a liberal information first of all was was this as we saw its Fuehrer directive number 51 the the German military was engaged in a breakneck campaign to beef up the defenses along the Atlantic Wall so along the coast of France and Netherlands and Belgium on a huge huge impressive scale I mean they were directing forward defenses so like these concrete blocks in the sea with like mines on top of them which the British called Rommels teeth designed to blow ships off a sailor came in on landing bay and and sort of massive construction campaign you know new gun positions new bunkers built a built in using slave labor and also using local labor and they lay the million crazy oh my it's more than six million mines by late November 1943 they were great they were vastly preparing so all this come all this level of information comes through and these broken communications I mean facts and figures you won't believe the impact assessment now the Allies were conducting a massive bombing campaign to distract attention away from normally they're bombing the northern France area around Calais north of Paris to make it seem like this is going to be the area that they're going to attack and what happened was course they were listening to the chatter that's coming back from that and what they learned was that several things kind of came out one is by listening to chat I learned the amount of physical damage that's being done to the German military units are being destroyed dislocation of bridges and tunnels all this data that was coming through them very nicely they also got an insight into the amount of dissipation that was going on in terms of supplies that were moving around they came to really become very of it a very obvious that supplies we're not making through making it through to the front as it would be so ammunition was not making it through fuel was not making it through tanks and troops not be able to move because railway lines were being attacked during the day bridges would be blown up so they've got a good insight to the amount of dislocation what's going on and secondly most importantly they got this information that demonstrated to them that the chatter proved that the Germans were convinced it's gonna be the past de Calais area they thought this is it they're bombing it we we know we know what's going on it's like the first world war you bomb it then you send your troops in over the top so they kind of seen this path they knew they had no idea what's gonna be Normandy so this comes through in the chatter as well and the really important thing I said I won't maybe I find fascinating I suppose being a slight student of nerd history of Second World War was the appreciation of just how much the German army was not the force it was in a by 1944 the force that invaded Poland and France in 1939 1940 was not what it was I mean you look at this information this decrypt you're getting a detailed breakdown of really my new like movement of generals and commanders because it's construction and composition of movements and units and where they're being allocated to and whose can what the battle plan is going to be but also became clear was how they were weak and what happened slightly before d-day was that a lot of Divisional Commanders a battalion level and local level and regional level were assigned to Italy so they were taken away from France they were put in Italy where the campaign was active because the troops down there was suffering from a lack of training and battle readiness and they were losing the battle in Italy so these more experienced troops that were thought commander level would move down there suddenly they've weakened the army the German armed is now much more weaker command level in France so that was very interesting so they know but it's very rich in science the state of battle readiness of the German army now that's all really good but for me that I suppose the I suppose nothing symbolizes more of me the the success clausus had the outing of Hitler's special tank units the Panzer so none of you are familiar with parents units now it was kind of felt that tanks could could swing the success of Normandy for or against Rommel knew this he had a lot of experience of Tanks he wanted as many tanks as possible on the beaches really behind the beaches to repel an invasion so there ten Panzer units in allocate de France but no but the Brits didn't bring nowhere they world the Allies didn't Reno where they were so below us think they want to do is they want to send they do not want to send a and evasion force in against a consolidated coordinated efficient tank defense cuz it'll slow you down it will lose a lot of troops so what is a what I don't know how much you guys are familiar with with the minutiae of German military but quite like tanks what is a Panzer Panzer units were a combination of troops and vehicles in particular tanks like this this is a mark for tank it's from the Panzer Division 21 which I've come back to later on later on roller and but this was this is a this was a baby tank by comparison they had some really hardcore tanks the tiger tank was probably the most infamous tanks the Second World War wasn't why the best to a certain extent but has failings it had a 88 millimeter gun on the front which is an artillery piece that could shoot down planes it was thickly armoured as well I mean this you didn't want to rub up against a tiger tank it which it would cook Sherman tanks for an even for breakfast basically so you didn't wanna run into a targeting so the pansy units also there was a combination of Panzer units you know some are very new there are contantly being reformed some are very old the common denominator really was they were all very very experienced battle-hardened troops mean these guys are fought in North Africa they fought on the Russian front they were they could be really nasty I mean some of them engaging if they hadn't been involved in atrocities by Normandy some did so after Normandy they thought nothing about killing innocent men women and children reprisal appearances would do that monks other units but they they were tougher units so if anybody could live up to Hitler's fury director 51 it was thought the Panzers can do it well fortunately we had this man to thank for revealing the locations of the pounds of units through the work of Colossus this is Carl Heine's godean he was the Inspector General of mechanized units and he decided to Undertaker's I suppose any good commander would a tour of all the Panzer units in France in April May 1944 and being very efficient the Germans transmitted his itinerary through the jellyfish Network and it got picked up and it got picked up and decoded and they worked out the locations of the units now you can see from that a couple of things one is really important you can seal from the success of operation fortitude the Germans don't quite know where to position all their tanks so they've split them which is very bad thing to do so you can see that they're split across across North and South of France but what's I think alluding to what Mark said earlier about the level of work that was undergo was that was facing the the team working on classes was this is a real catch I mean by this time they were receiving a thousand messages a month they were only able to attack half of that number here so you got to bear in mind they really look at getting his itinerary it's a real needle in a haystack job and yet they got it the his they were suffering a chronic shortage of people and computers so it was really dedication and diligence that hard work and read in that terse German that revealed this man's itinerary imagine if they hadn't got the job that would be a different cup of fish I think for the Allies and much more a much more kind of like nervous bunch of troops going in there and commanders this is the revelation was so important it was one of the message that found its way up on to Churchill's desk and we know very relieved too and Churchill read this communication as well so what did it tell us what was I guess we want to find out the the 65 million dollar question or whatever it is where were those units well what they discovered was that there was out of all those ten there was only one unit that actually posed any realistic immediate damage a danger to the landing forces and it was the 21st Panzer Division which was based in Rennes which I think's around there somewhere it's 185 kilometers away so again it was suitably out of position on that first day I think if you want to understand the significance of the parents units how they could really wreak some havoc the 21st Panzer did make it to to the frontline very quickly actually when they drove hot that horrible day to get there along with other German units and they made to Conn and the ally plan was to take Conn on the first day of d-day they failed it took a month to take common because units such as the 21st Panzer did make it there so if you think about that on a larger scale if those pounds units had been in that area had been masked it would create am I able to slow down d-day to such an extent we don't know what the outcome it's well those what if history is what might have happened we know what had happened but what if all those pans has had been there and consolidated its once I could discuss a thing after finally I suppose another interesting thing what Colossus revealed was I suppose the structural problems at these pans are these elite Panzer units were facing you know what you see here our next couple slides I'll show you the told you through these are actual words from the commanders in the field they're expressing the trouble that all this Allied bombing campaign is causing for them so you're looking at you know there's a shortage of fuel and that lack of fuel is mean that the the Panzers can't go on maneuvers as much they'd like to so it's restricting their kind of fighting capability trading supply of troops is breaking down they can't get the troops they want the troops are insufficient quality the officers aren't available so the quality the Panzer units is now breaking down as well and finally I think this is really interesting and this kind of links back to why the points earlier on which was about the the bombing campaign in the north it will kind of commish's quite nicely Mellie camp was a massive or it may was a very large panzer training ground to the east of paris and there was an the Allies conducted a huge bombing campaign things he made one night in May there was about a thousand or so Allied aircraft involved they obliterated the place and you can see here the Allies got discover what exactly what material damage they had done as you can see there quite a lot of damage so once again the Panzers have been hitting there one of their key training facilities they've lost men they've lost material they've lost tanks so they're they're on the back for him and the Allies get to know this stuff by reading the decrypt so I suppose something that all up mean you can see I mean this for me those three areas based on the intelligence information that we read the D Crips there's three areas in the Colossus the Machine and all the men and women who worked on it they contributed to d-day without really mean they all had a snapshot of it they didn't read at see the bigger picture this is the bigger picture you can see these are the ways that Kloss has helped indeed aim and was so pivotal to the operation as alluding to the start they were figuring on 40,000 casualties if they were lucky eventually was 10,000 actually was just over 4,000 dead I mean that number could be not larger and I think it's safe to say that without Colossus without the work they've done here they probably would have been as well and that's all we've got to say happy to take any questions thank you very much indeed goin great complement to the to the reconstructions that are across the way and the actual original Lorentz equipment that you can see today great piece of research thank you very much indeed mark and Gavin and I think clearly an inflatable tank needs to join the display across the way let's open up the questions just wait until we get the mic in front of you and both mark and Gavin are willing to take questions thanks what you've described there the enormous amounts of information I deduced that most of that is from before the 1st of June which is when the mark 2 comes online with these extra speed now conventionally we talk about the extra speed of the mark 2 allowing messages to be broken within 24 26 hours or better if lucky can you tease out how much effect that mark 2 has on the intelligence at the beginning of June as opposed to all this build-up that you've described I think mark you had some data on that if you look at if you refer back to Mark's chart earlier on you start seeing the slow scalation how much information they were able to process I think as you as you correctly said all that information is based on the shoulders of the one Colossus unit and so and they were also processing information that was they were in arrears on information they were posting half the mount messages they were actually receiving and there was this constant churn of they would take a pile get back to if we can so there's a vast amount information they never process we don't know how important it would have been that's the mark you want to take the question about how much Colossus - yeah of course there is there are sort of apocryphal stories there are stories accounts of people saying that closest to made sort of an enormous breakdown and breakthrough that was um critical to the success of d-day closest to specifically I never seen hard evidence of this actually I'm slightly skeptical for the following reasons Colossus too did come into service on the 1st of June but it was sort of like you know it was the rocket there are documents saying that was essentially it was like 3 o'clock in the morning when the people from Dallas he'll finally got it more or less working to meet a contractual deadline which is completely different from saying it was available you know to be slipped into the operational procedure on the 1st of June so you know it's not it's not impossible but you know it took clauses one for example about a month after it was officially started working it was still partially complete and people were talking about teething problems people writing memos saying now that Colossus is finally coming online there's good there's quite a long you know with the best will in the world you have to accept that it's gonna be a lot of tea there's quite a long teething Pro Set there teething process you know I mean yeah before before before it's gonna really start delivering so it's not impossible of courses to was used to decode messages before d-day itself but you know I'd be I'd like to see hard evidence about that the other thing is that the codes for the codes that the Germans used for encrypting the messages changed at the end of the month so that at the end of May the Germans would have changed the codes for the for the the way that Lorentz machines were set up and the jerk the Brits would not be able to read any June messages until they'd worked out the new wheel settings and technical term the new wheel settings so it's a two-stage process you've got you've got to work out the new configuration of the Lorentz machine which is a really hard job and then you've got to decode individual messages and they took them it could be easy it could be hard but it never it always took them a week ten days sometimes the best part of a month before they could start reading messages in that month so the idea that any messages at all and that I can that I can pretty much guarantee that no no June messages would have been read before d-day simply because they hadn't mentioned wheel breaking before d-day itself so I know there are stories about courses too but I mean the best one in the world you know I'd like to see some some hard evidence which probably doesn't exist so I'm gonna be slightly skeptical about me I'm gonna be slightly skeptical about the about the importance of Colossus to it for those reasons yeah that's just my take so dude so just in June in June for example I think I remember correctly June was actually quite a difficult month I think I think the graph dip down like maybe July the grant the graph dip down and what they the total number of decodes was quite high because they've got this new machines but what they were actually decoding was messages from some other link from like a month or two ago which if you read through the messages you get quite a different feel to the distribution of the distribution of how and the messages they were successfully decoding and it's not quite it's not always the you know the onward knotwood story of untroubled progress that you might have expected but yeah it's no criticism at all it's dis reflection how bloody difficult it was and the changes the Germans because the German I mean it round about in the middle of the roundabout June July the Germans did another thing which was the whole the whole thing is like it's like the journey so let me wind back so there was through a three bit connect I'll just spend a couple minutes on this if no one else has got any other questions so in the course of thought there were like three moments where the Germans did things that really mess things up so in October 42 they stopped having little headers on the messages it said this is how we've set our machine up and the Brits had to work that out from scratch and that took a like a year and they had to build these machines to do that the in January 1944 the Germans started using something called the p5 limitation which just made it much much harder to apply the statistics to these messages and just coincidentally enough closest was there to sort wasn't what Colossus was designed to do although it could cope with it but the computational power of Beck's and it was available when it was necessary in june/july the Germans started changing the wheel settings of these machines on a daily basis rather than a monthly basis which meant that the really hard bit of the code breaking was suddenly 30 times as difficult had been before and just at that period that cost us two and the three four and five come into service but again it's you know they had no idea the Germans were going to do that but there was this kind of you know there's a couple of occasions where you think there's suddenly they suddenly got this new additional power just when the Germans do something and it's like it's it's kind of weird it's the whole thing looks a bit can Heath Robinson like you know it's like almost accidentally the way the way that the the colossi were there when they were needed for things that nobody could anticipate we're gonna happen if that makes sense accidental success I strongly recommend that you have a look at the display that Gavin put together in the Colossus gallery it's a new addition to the courses gallery and of course the the Tunney gallery is is there also I think it's fantastic that 75 years on this story still develops and quite possibly more to come we know there is some documentation that is there but we don't even know what it is and whether it will ever be released probably not so that's traffic analysis [Laughter] so thank you very much indeed for coming thank you very much indeed Gavin Clark and Mark Priestley who have done this research for the museum and it's tremendously to the story thank you very much [Applause]

Channel: tnmoc

Views: 5,319

Rating: 4.8800001 out of 5

Keywords: D-Day, Colossus, Bletchley Park, World War 2, Hut 3, TNMOC, Lorentz, codebreakers

Id: vsMH7OKgrzE

Length: 70min 2sec (4202 seconds)

Published: Wed Oct 09 2019