POLISH CIPHER BUREAU
Tadeusz Lisicki commanded the signal battalion of the Polish General Staff, stationed in Stanmore. Attached to it was a small cryptanalysis unit at Boxmoor near London. This unit was in constant touch with the Intelligence Department of the Polish staff and also BP, which allocated work to it. This mainly consisted of reading the German SS Doppelwürfelverfahren cipher. Rejewski and Zygalski joined this unit when they eventually reached Britain after the fall of France.
Lisicki had read The Hut Six Story and wrote to Welchman on 22 August 1982 with a few comments. Following an exchange of letters in 1982, Welchman wrote to Lisicki again on 13 August 1984. He was now working on a paper about the early days at BP and asked Lisicki a number of questions about the Polish contribution.
3 September 1984
Dear Mr Welchman,
I am so glad that you decided to write another part of the Enigma history. I hope it will be a book not an article. I was away and could not answer sooner your letter. I will try to help as much as I can and perhaps even add some new puzzles. I have access to some unpublished documents and papers (in Polish) written during the war by Langer (1) and Rejewski and over sixty letters from Rejewski written after the war.
First to keep the record straight I was NOT attached to Bruno but had a different assignment which gave me access to the work of three Polish teams working on German, Clandestine and Russian ciphers (Russian are still tabu so please do not mention), we had some success in Bruno and later in England and perhaps Russian decodes were helpful; for cribs, Hinsley (2) is a bit evasive and does not tell the whole story what Bletchley did, I think it was more than Metro (3) (after 1941). When I came back from USA to England I was OC of a Polish Unit which worked with Cadix (4) and Bletchley.
I can only add a little on Denniston contacts with Poles. The first one was in Paris 0n 8-9 January 1939, Denniston led the British delegation, with him went D. Knox and H. Foss. This conference was organised by Bertrand who with Braquenie represented the French and Langer with Ciezki the Poles. The meeting is described in Bertrand book (p.58).
The second meeting took place in Warsaw on 25-36 July 1939, again Denniston and Knox and Menzies incognito (5) represented the British. Mayer told me when he met later in London they talked about their meeting in Warsaw, he told me also that the two Enigma machines were send by diplomatic bag by sea from Gdynia to France and one was later delivered by Bertrand to London. This meeting is well described and I have nothing to add.
Perhaps it will be useful if I first tell you how Bruno was organized.
Chef F.C. Bruno – Commandant G.BERTRAND
Adjoint – LtCol. – G.Langer (and liaison to Polish Army)
Liaison to the British Army – Cpt K. MACFARLANE
1) Allemagne
Chef de S/Section – Commandant RENARD (Fr.)
Adjoint – Commandant M. CIEZKI. (Pol.)
Directionde l’inteception – Lt ARNEUIN. (Fr.)
Direction de repartition – Mr. M.Fokczynski. (Pol.)
1.1. Decriptement Procedes Manuels – Cpt Larcher + 2 cryptol. (Pol.) + staff
1.2. Decriptement Procedes Mechanique – Cpt H.Braquenie + 7 crypt. (all Pol. Rejewski, Zyglaski, Rozycki …. ) + Staff
1.3. Dedriptement Clandestin – Cpt Marliere + …… (Fr.)
2) U.R.S.S.
Chef de S/Section – Cpt Couey (Fr.)
Adjoint – Cpt Chasles (Fr.)
2.1Decriptement – Cpt. Gralinski (Pol) + 2 cryptol. (Pol_ + staff
3) Italian
The total:70 cryptanalysts and staff working on breaking codes and cyphers. (12 May 1940). As far as I know this was never published.
How Bruno collaborated with Bletchley was agreed on a conference held in London on December 1939. From France came Braquenie & Langer. I do not know who was from the British side. The main points agreed were: interception, exchange of solved keys and decrypted messages. All this collaboration was super secret, MacFarlan acted in Bruno as a cut-off and had a direct line (teleprinter) to BP, all messages past through him and were enciphered on an ENIGMA. Apart of the meeting of top people Bletchley cryptologists visited Bruno but never Polish from Bruno went to England. Whoever broke the daily key first send it to the others. I have a full list of keys broken between 17 January and 21 June 1940 but it gives only the dates of the key and the day on which it was broken (I tried to make a Xerox copy but the original is too faded, perhaps later I will type it). It do not indicate what net and who broke it, however there are days when two keys are broken, perhaps one green and one red. In Feb. 16 keys were solved, some with delay of several days and some even very old i.e. key 29 Sept. solved 23 Feb. and 30 Sept. on 23 Feb. In March 20 keys solved with shorter delays and for the first time two keys, one on 15 March, second on 28 Feb. both for 21 Feb. In April it was a great improvement 36 keys broken and 8 from two nets, from the middle of April usually next day. Perhaps here is the answer why Winterbotham told in April the Air Ministry. But here comes a new puzzle: Hinsley write (p493) “By the middle of December, two copies of these sheets were completed and one was sent across to Paris. At the end of the year, GC and CS records, our emissary (6) returned with great news that a key had been broken (October 28, Green) on the sheets he had taken with him. Immediately we got to work on a key (October 25, Green) … this was the first wartime Enigma key to come out in this country, and was broken at the beginning of January 1940. The GC and CS account continues…” In my list the 25 October key was received 27 Jan. I only suppose it was received, it is the only key from this day. In our papers the sheets arrived with Turing, NOT December, but Hinsley writes GC and CS account ….., so an account exists but the dates do not tally. I was told that the History of GC and CS was written (in 5 volumes), I think by Ronald L. but he could not get access to it.
In May nearly all keys were broken the next day or same day but there is a gap from 15 to 21 May with a notes change in the system, but everybody gives the date of change 10 May this is the next puzzle. My explanation is that the order was given for the 10 but not all nets changed on that day. Why you were not told about the exchange is really strange. I am sending this letter today and continue in a few days.
Yours sincerely
Tadeusz
9 September 1984
Dear Mr Welchman,
To continue. Col Mayer in his “Supplement to the paper of 1974” writes (p.3):
“One more anegdotic detail of this July 1939 conference. Commander Denniston learned in conversation with me that my daughter /then 14/ is taking English at the school as foreign language. He promptly and amiably suggested that we could arrange exchange of daughters during long school holidays for language studies. I would send my daughter to England to his family and his daughter would come to Poland to live with my family. Of course the outbreak of the war thwarted any possibility of such an arrangement.”
Col Langer writes “I suggested to the French to send to the 5-e Bureau de L’E,M,A. not the original decrypt but a digested form of a translation from German. The French did not agree and told me that it would be a waste of time”. How lucky that none of the decrypt fell into German hands during the total collapse of the French Army.
Langer in his account gives examples of telegrams decrypted in May and June but again they are hardly readable. He also complains that Bruno possessed only two Enigmas for decrypting and the French did not managed to manufacture in time some additional machines. The two came from Poland.
I fully agree with you that the Bletchley set up was so secret that now is very difficult to unravel some of the mystery.
To answer your technical puzzles.
Both the Cyklometer and Bomba had single ended scrambler units. The polish bomba was used when only six stecker connections were being used, that why the perforated sheets were invented when the Germans started to use firstly all 13 and later 12 connections (7). None of the ideas which you listed were Rejewski’s, he was happy with the sheets and discarded the idea of improving his bomba. The loops, the double ended Enigma, the diagonal board and the filling up of the test register (8) were all British ideas and Rejewski in his letters to me several times mentioned that he had no idea how to mechanize the search for the keys and thought that the British mechanized the sheets, but that would be useless after the 10th May 40 (9). In Bruno he had absolutely no time for creative work. The running of a number of Enigma scramblers was the only idea which the Poles first used and perhaps was born from the Cyklometer. The “Knox Method” had two parts, to find the stecker connection from the Metro code and use a kind of pre-Herival tip. The Metro code first appeared either in the end of December or beginning of January 1940. It was a simple three letter code, for example sunny – DPR, then if for the day the stecker connections were D-S, A-P, F-R, it was enciphered by these letters and DPR became SAF. It was not difficult to decode this code and to have every day the stecker connection given by the German. This code was used between airfield and a flying aircraft. The second part was to find after midnight a message transmitted from a certain net where one of the operators (lazy one) always used the three letters which appeared in the window when he reset the new daily key which was an equivalent to Herival tip (?).
Lisicki then provided some more explanation to the Cyklometer description from Cryptologia along with a hand written technical explanation, which is not included here.. He ended as follows:
Sorry for the untidy “presentation” but I am going away for few weeks and wanted to send this letter before departing.
Yours sincerely
Tadeusz
P.S. Do NOT hesitate to ask any more questions or for clarification.
31 October 1984
Dear Mr Welchman,
It took me a long time to trace Langer’s documents and I was abroad for a few weeks. But here they are with my translation and explanation. (see Appendix 5)
I have read Kozaczuk’s book and I do not think it is a good book except Rejewski’s appendix. There are some mistakes and contradiction and the translator was not a cryptologist. As to appendix F (10) I have some reservations, but have to study it once more. The book was written by Kozaczuk a Polish historian living in Poland and one can see that the censor was looking over his shoulder. I would be grateful if you could write your opinion on this book.
I am now in correspondence with Col. Paillole (11) who got some Bertrand papers and is thinking of writing a book (not on Enigma) but will mention some of Bletchley-Bruno collaboration.
With kind regards
Yours sincerely
Tadeusz
23 December 1984
Dear Mr Welchman,
Thank you for your letter dated Dec. 4, I am glad that a small book will be published and most gratefully receive a typescript. There is an important point concerning the French contribution, on the page 9 of the enclosed Rejewski’s original paper written in 1941/42 there are at the bottom the six equations, to find the connections they must be solved. (The alpha etc. are in Rejewski’s handwriting and this page could be reproduced in your book). Rejewski wrote that the difficulties in solving these equations were: firstly that both sides of the equation are unknown and only the product A1, A4,A2,A5,A3,A6 are known, the cycle theory tells that there at most up to 30-40 designations for A1 which also designates A4, the same is true for A2 A5 and A3 A6, so we could write on the left hand side of the equation all possible designation from A1 to A6, very time consuming but could be done; secondly even after the inner connections were found ( for several years the Polish team, just for theoretical reason tried to solve these equations assuming that E is known but they were always defeated by the S, at the end they found a way which needed a lot of material (intercepts) but was never finalised, I have asked Rejewski if he could write it down but he died before he could do it, perhaps you could persuade some mathematician to do it, Rejewski thought it could be a doctoral thesis, or perhaps a computer could help (?). The S were given by the French in the set of keys so the equations were solved, now the QUESTION IS IF THEY COULD NOT BE SOLVED WITHOUT KNOWING S THE FRENCH CONTRIBUTION WAS VITAL AND THE ENIGMA WOULD NOT BE BROKEN WITHOUT IT; I would be grateful for your opinion. (See Appendix 2)
I do agree that Kozaczuk book is full of errors some I pointed out when his book first appeared in polish and later answered many questions from Woytak, but never thought that he and Kasparek will make such a mess of it and it is shocking that they have not asked for your permission. All three have not even a faintest idea of cryptology, Kozaczuk was at the beginning helped by Rejewski but wrote his book with the censor looking over his shoulder and having in mind all the nonsense which Winterbotham, Cave Brown etc. wrote about the Polish contribution; “we gave them money and they did the job, so please be a bit lenient”. Woytak is a historian (California University) born in America and so not speak polish and Kasparek is a MD also borm in USA but speak polish.
Rejewski’s papers are based on his original work which gives the essence of all major methods, description of bomba, etc. This work written in France in 1941 is unsigned but as I mentioned before corrected by his own hand, it was sufficient to prime his memory. Langer’s report with the list of keys was never published. I was wrong writing that the first break was on the 17th January 1940 of a key of 25 October 39 it was 29 Oct (see Langer’s list). Kozaczuk refers to Langer’s documents in Rejewski’s file and also quotes some passages but I send Rejewski only few extracts. All documents and what I wrote is NOT classified, so it Langer’s report available to everybody. Zygalski who later served with Rejewski in my unit (Signals Battalion) helped in the seventies to reproduce their work in Poland and France and I have some tape recordings of conversations with him. (Stanmore was my HQ but we thought the WRNS had something to do with RAF Command nearby). Zygalski exchanged many letters with Rejewski.
The green key for 29 October was broken in Bruno when Bletchley people (Knox and Turing) were there, perhaps it was a joint effort as the perforated sheets just arrived. If I wrote it was in Bletchley I was wrong, I have in my notes “in the presence of Bletchley cryptologists”.
I was in touch with R. Erskine (12), his letter was recently published in Lit. Times, I enclose a copy, he is quite right that Poles who fell into German hands revealed nothing. I also think that the other great polish contribution was a moral one that the Enigma can be broken.
The Frenchman who was in contact with Erskine came to London and we had a long talk., he is Gilbert Bloch and helps Paillole who is working on a second book. Paillole wrote to me and as I understand his book will be on the counter-intelligence work but he also would like to add a fuller contribution of the French toward the Enigma solution. Block is writing a personal study “Enigma avant Ultra” which will not be published in the near future and promised to send me the typescript. He was in touch with Hinsley who admitted some mistakes in his appendix (Gilbert Bloch is not a cryptologist).
Paillole was very helpful and had an organisation in Vichy France which as Bertrand operated underground and they collaborated helping each other.
I enclose 3 pages of Langer’s report with translation of the polish text and german messages which you might find interesting.
Yours sincerely
Tadeusz
(1) Guido Langer, Head of the Polish Cipher Bureau
(2) Harry Hinsley who wrote the Official History of SIGINT in WW2
(3) A code used by the net of German meteorological stations which broadcast information on air force operating conditions three times a day
(4) Codename for Bruno
(5) Lisicki was wrong. It was Sandwith not Menzies
(6) Alan Turing
(7) 10 were ultmiately used not 12
(8) The key ideas behind the British Bombe
(9) When the Germans changed the indicator system. The Polish methods exploited weaknesses in the original system
(10) Polish and British Methods of Solving Enigma by Christopher Kasparek and Richard Woytak, essentially a critique of Welchman's book
(11) The illegal operations of Bruno and other clandestine centres in Southern France were shielded by French counterintelligence under cover of the Rural Works Enterprise. Paillole headed the unofficial counterespionage organisation in the free zone
(12) The historian Ralph Erskine
Image: Marian Rewjeski, key member of the Polish Cipher Bureau, 1932.
