A telegram sent by a British colonel in 1916 still waits to be solved. The encryption method used might be a letter-pair substitution.

Blog reader Tony Gaffney, who recently broke the encrypted diary of a pedophile priest, has made me aware of an interesting cryptogram that was published in 2014 by the Bury Times, a local newspaper from Bury, England.


An encrypted telegram

The Bury Times article reports on an encrypted message sent in 1916 by a Colonel Frederick Arthur Woodcock. When the article was published, this message was unbroken and it probably still is. Here is a scan of it (sorry for the bad quality):

Source: Bury Times

As can be seen, this is a telegram. It was sent on October 16th, 1916, to a recipient named Caledonia. The Caledonia probably was a ship (Wikipedia lists several ships of this name). The telegram is noted on a form of the Marconi International Marine Communication Co. (M.I.M.C.Co.), a subsidiary of the Marconi Company. Here’s a M.I.M.C.Co. avertisement poster published in 1917 (it has no relationship to the Woodcock telegram):

Source: Grace’s Guide to British Industrial History

Considering that Woodcock was a colonel, that the recipient was a ship crew and that the telegram was sent during the First World War, it seems plausible that the message was military in nature. The telegram form doesn’t look military to me, but perhaps a commercial radio station was used to transmit military message.

Here’s my transription of the message:



The key (?)

The most popular method to encrypt telegrams in WW1 was using a codebook.

In the military, many codes were super-encrypted, which means that an additional encryption step was applied on the codewords used (e.g., adding the current date). The message on the Woodcock telegram is consistent with a codebook encryption. Codebooks with five-letter codewords were quite common.

However, according to the Bury times article, the Woodcock telegram was not encrypted with a code but with a cipher based the following key, which is contained in a notebook Colonel Woodcock owned:

Source: Bury Times

However, in my view, this is not a key but the result of an encryption, i.e. a ciphertext. It looks similar as the telegram message. I don’t think that these lines were produced by looking up words in a codebook or by a codebook super-encryption.

As the text contains vertical lines that separate letter pairs, the cipher used might be a letter-pair substitution, e.g., a variant of the Playfair cipher. The Playfair was a quite common cipher in the first half of the 20th century. If Woodcock really used a cipher of this kind, the ciphertext might be too short to break the message.

Tony Gaffney has emailed the librarian in Bury asking if this cryptogram was ever solved and if it is possible to get more scans – but he didn’t receive a reply. Can a reader solve this mystery anyway?

Further reading: The Top 50 unsolved encrypted messages: 46. Unsolved ADFGVX cryptograms from World War 1

Linkedin: https://www.linkedin.com/groups/13501820
Facebook: https://www.facebook.com/groups/763282653806483/

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Kommentare (18)

  1. #1 True Dan
    19. März 2019

    It appears that, if you account for transcription errors, the telegram is the same as lines one and four of the notebook.

  2. #2 Thomas Bosbach
    19. März 2019

    Obviously this has to do with Playfair: ER – GT – NE and RE – TG – EN (lines 1, 2, 3 belong together, the same is the case with lines 4, 5, and 6). Hence these bigrams form squares in a Playfair table so that the cipher bigram becomes reversed when the plaintext bigram is reversed. Given the bigram combinations in the notebook, reverse engeneering the Playfair table should be possible.

    How do we know that Woodcock was the sender and not the recipient who tried

  3. #3 TWO
    19. März 2019

    Friedman IC: 1.4667
    Kappa-PT: 0.0564

  4. #4 TWO
    19. März 2019

    Dear the

  5. #5 Thomas
    19. März 2019

    Lines 2 and 5 seem to contain the “plaintext” of lines 1 and 4, (1 and 4 contain the ciphertext of the telegram), since 1 and 4 are derived from 2 and 5 using a Playfair matrix with the keyword “STURDILY”. I suppose that lines 1 and 4 can be derived from 3 and 6 with a different Playfair matrix, still haven’t figured out the keyword. Since 2, 5 and 3, 6 don’t contain real plaintext, maybe the Playfair was a second step of a double encryption. Or maybe the notebook contains attempts to decipher the telegram message with two differrent Playfair keywords, but to no avail. I wonder whether Woodcock had sent or received the telegram…

  6. #6 George Lasry
    20. März 2019

    My Playfair could not find a valid solution. With such a short length (40), the text has to be with exceptionally favourable ngram statistics, or the solution will be ‘buried’ under spurious solutions with better scores. This does not mean this is not Playfair, however.

  7. #7 Thomas
    20. März 2019

    Did you take into account that Klaus’s transcript might contain a typo?: The last letter of the 3rd group must be V instead of Y, otherwise line 2 wouldn’t fit to line 1 in Woodcock’s notes (who had the original at hand). Does your algorithm generate arbitrary grids or does it perform a dictionary attack with a reduced keyspace (grids only containing keywords of length n and 25-n letters in alphabetical order – as Woodcock apparently assumed according to his decrypting attempts in the notebook)?

  8. #8 George Lasry
    21. März 2019

    I tested only arbitrary grids. I can also try a dictionary attack but I need to implement it.

  9. #9 Alex
    21. März 2019

    Habe den Blog gerne gelesen, doch wann wird dieser wieder auf Deutsch erscheinen?
    Finde der Blog ist eine der interessantesten im Netz die es gibt.

  10. #10 Tony
    21. März 2019

    I’ve just received the following reply from Bury Art Museum –

    ‘The images that were taken and put on the Bury Times website were the only ones taken.
    As I recall the pocket book did not contain any other information.
    I will get the book back out of the archives next week and see if I can get some better images.
    We had no response at all so the note remains unsolved.’

    I will pass on to Klaus any new images etc.

  11. #11 Thomas
    24. März 2019

    Did you make up your transcript based on the extremely blurry image of the telegram? More reliable should be the transcript in the notebook (lines 1 and 4), since Woodcock saw the original.

  12. #12 Norbert
    25. März 2019

    Woodcock’s deciphering attempt in line 3 and 6 seems to base on the keyword SWITZERLAND. However, some letters do not fit (interestingly, at more or less the same positions where for line 2 and 5 STURDILY does not fit).

    This is the most I have made of my dictionary attacks so far.

  13. #13 Tony
    26. März 2019

    Norbert – Woodcock made a slip when using SWITZERLAND – instead of decoding the 5th pair FX he decoded the GW beneath it from his first try with STURDILY

  14. #14 Thomas
    26. März 2019

    Woodcock didn’t know the plaintext of the telegram, for he was neither the sender nor the recipient. Maybe he had found the (outgoing) telegram and tried to decrypt it with the keywords “sturdily” and “Switzerland”.

    To the background:
    Is there a connection to Lawrence of Arabia? At the time of the telegram Woodcock served in the 125th Lancashire Fusiliers Brigade (former 1st/5th battalion of the Lancashire Fusiliers, https://www.ww1infantrycos.co.uk/lancsfus.html) as part of the Expeditionary Force in Egypt (https://www.lancs-fusiliers.co.uk/tourspostings/1-5thLFToursPostings.htm). T.E. Lawrence served as a British intelligence officer in Cairo and was sent to Saudi Arabia to support the Arab revolt against the Ottoman Empire – according to Wiki – on 16th October 1916 (date of the telegram). In his autobiography “The Seven Pillars of Wisdom” T.E. Lawrence mentions a Bedouine leader (Sheik Auda ibn) “Zaal” (see the office of origin in the telegram). The recipient of the telegram was S.S. Caledonia, a ship required by the British government for troop transportations in the Mediterranian during WWI and torpedoed in December 1916 (https://navalwarfare.blogspot.com/2013/06/ss-caledonia.html).

  15. #15 Thomas
    27. März 2019

    Moreover, Ronald Storrs, Oriental Secretary to the British residence in Cairo and mentioned by Lawrence in his autobiography, praised Lawrence’s knowledge of the Playfair cipher in a diary entry from 14th Oct. 1916 (https://archive.org/details/in.ernet.dli.2015.174934/page/n225), two days before the telegram was sent!

  16. #16 Thomas
    27. März 2019

    Wipe out “Zaal” in #14. The sender was “Zaaw”, here a much better image of the telegram that I recommend for each decrypting attempt: https://mobile.twitter.com/BuryArchives/status/495151484376731648/photo/1

  17. #17 Thomas
    27. März 2019

    “ZAAW” was the call sign of a British patrol boat, https://archive.org/details/n10wirelessage04nyne/page/758

  18. #18 Klaus Schmeh
    30. März 2019

    >Is there a connection to Lawrence of Arabia?
    If so, it would be a great story.