Devils-Letter-bar

A 17th-century Italian nun claimed that the Devil had dictated an encrypted letter to her. Scientists have now allegedly broken this cryptogram. Details are not known yet.

One of the best sources for unsolved historical cryptograms is Google. Using search terms like “encrypted letter” or “postcard in code”, I often find interesting topics to write about. Besides English and German, I use Russian, Italian, Portuguese and other languages for my searches – thanks to Google Translate this works quite well, though I have no command of these tongues.

 

A letter received from the devil

However, in spite of many Google searches and in spite of consulting the crypto history literature, it seems that I have missed a very interesting and well documented cryptogram – the Devil’s Letter.

The Devil’s Letter seems to have some popularity in Italy. There is even an Italian Wikipedia article about it. However, this encrypted note is not mentioned in the standard crypto history literature and it doesn’t show up on one of the numerous top ten unsolved ciphers lists that are available on YouTube.

The reason why I have now learned about the Devil’s Letter is that it has recently been broken (allegedly). This has been reported by several English and Italian news portals. Thank you very much to blog reader Eberhard Bauer for the hint.

According to legend, the Devil’s Letter was written in 1676 by an Italian nun named Sister Maria Crocifissa della Concezione from the convent of Palma di Montechiaro on Sicily. She claimed that the devil had dictated it to her in order to convince her to turn away from God.

Devils-Letter

To somebody interested in crypto history this letter looks like a text written in a secret alphabet. If so, the encryption can be regarded as a mono-alphabetic substitution cipher (MASC). Usually, cryptograms of this kind can be solved.

According to several press articles that have now been published, many codebreakers and historians tried to decipher the Devil’s Letter, but to no avail. If this is correct, I am a little surprised that this cryptogram is not mentioned in the crypto history literature I know. Maybe this is because so far mainly Italians have occupied themselves with it.

 

A solution – is it correct?

Earlier this week, researchers from the Ludum Science Center in Sicily announced that they had deciphered the Devil’s Letter. Of course, one needs to be careful with claims like these. After all, dozens of nonsensical solutions to all kinds of unsolved cryptograms have been published in recent years – just think of the alleged Voynich Manuscript solution I wrote about in my last post. However, this time I am a little more optimistic than usual, as the Devil’s Letter looks solvable.

To my regret, the Sicilian researchers have not released the complete solution at this time. Instead, they shared only a cleartext passage stating “God thinks he can free mortals. This system works for no one.” I hope, we will learn more soon.

The scientists that claim to have deciphered the Devil’s Letter are said to have used a special computer program they obtained on the Dark Web. This software is allegedly used by secret services. I have no idea, what it is about. One of the researchers is quoted by The Times as follows: “We primed the software with Ancient Greek, Arabic, the Runic alphabet and Latin to de-scramble some of the letter and show that it really is devilish.” This is not necessarily a statement of high scientific value.

I hope that the Sicilian researchers will publish a detailed account of their work soon. If a reader has more information about this cryptogram or can break it, please let me know.


Further reading: A great event: the European Historical Ciphers Colloquium 2017

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

Subscribe to Blog via Email

Gib Deine E-Mail-Adresse an, um diesen Blog zu abonnieren und Benachrichtigungen über neue Beiträge via E-Mail zu erhalten.

Kommentare (12)

  1. #1 norbert
    10. September 2017

    This is not necessarily a statement of high scientific value.
    Nice said :-)

  2. #2 Eberhard Bauer
    11. September 2017

    Vielen Dank für die Recherche!

  3. #3 Thomas
    11. September 2017

    There are (at least) two different versions of this letter, one kept by the archive of the cathedral chapter in Agrigent, the other one by the convent of Palma di Montechiaro, https://sites.google.com/site/beatacorbera/lettera-del-diavolo.
    Did the nun write two letters heated by a Sicilian summer night?

  4. #4 Thomas
    11. September 2017

    @Klaus:
    Maybe you can ask Daniele Abate for further information? http://www.museodellascienza.com/?page_id=28

  5. #5 Davidsch
    11. September 2017

    I like referring to it as “Lettera del Diavolo” and not the English translation.

    @Thomas: it clearly says below: (Questa …originale.) Nice find b2way!.
    transl:
    The above is the facsimile of the existing document in the Cathedral of Agrigento, which is a copy. The original 1842 is published by Vitello, document kept in Palma di Montechiaro.

  6. #6 Paolo Bonavoglia
    Venezia
    11. September 2017

    I tried a statistics of the first 4 lines.
    I see some Italian clear text: il dì (the day), io (I), fu (was), e ho (?? and have)
    For the rest it’s hard to find two equal signs; I count about 35 different signs out of 39.
    A quick look to the rest shows even more signs, mostly different.
    This seems to exclude a mono-alphabetic cipher.
    It is like Suor Maria invented new signs at every step (or … the devil suggested them); the resulting cryptogram is like a Vernam.
    And from a Vernam you can obtain any clear text you like; just invent the “ad hoc” key!!

    Maybe I’m badly wrong, but I feel something of the like happened here.

  7. #7 Davidsch
    11. September 2017

    …that being said, it means that the image above this article displays the wrong item…

  8. #8 Thomas
    11. September 2017

    @Davidsch
    Do you think one is a copy of the other? There seem to be obvious differences, e.g. certain chars.

    @Paolo Bonavoglia
    As you are a native speaker – if you look at the three sentences in the plaintext cited by Davide Abate in the La Stampa article, can you identify the corresponding ciphertext passages in the letter? Abate mentions only the signs which stem from various scripts, but not wether the plaintext language is Italian.

  9. #9 Paolo Bonavoglia
    Venezia
    11. September 2017

    @Thomas
    I did it already. I could not see a clear correspondence between the “Poiché Dio Cristo Zoroastro seguono le vie antiche e sarte cucite dagli uomini,” before “Ohimé” sentence and the cipher-text. The first has 66 chars (in Latin maybe a bit less), the second has, as far as I can discern, about 31 signs, looking all different. Digraphic signs? Homophones? Nulls? Any way you could find [almost] infinite solutions of such a cryptogram, if it is a real cryptogram, as I wrote above.
    Surely Abate should give more details …

  10. #10 Charlotte Auer
    12. September 2017

    @Paolo Bonavoglia

    I don’t see “Ohime” really as such, I’d rather read it as
    “Chi me” even if there’s not a clear space between “Chi” and “me”. To me this would make sense, but my Italian…
    What do you think as a native speaker?

  11. #11 Norbert
    13. September 2017

    @Charlotte Auer:
    I am not a native speaker but an opera professional which is why the term “ohimè” (alas!) is not unknown to me 😉 See, for example, this dictionary from 1700:

    https://books.google.de/books?id=qdAtp13yp_AC&vq=ohim%C3%A8&hl=de&pg=PA538#v=onepage&q=ohim%C3%A8&f=false

  12. #12 Paolo Bonavoglia
    Venezia
    14. September 2017

    @Charlotte Auer: I read clearly “Ohime” without space; in Italian “Ohimè” or “Ahimè” is a common lament meaning “poor me” “dear me” “oh dear” …