A sword showcased in a Havana, Cuba museum bears an encrypted inscription. Can a reader decipher it?

Here’s really thrilling crypto mystery I found (like so many others) on Reddit.


The sword

Four weeks ago, a Reddit user calling himself VinceClorthoEC posted a number of photos of a Napoleonic sword that bears a cryptic inscription. For copyright reasons, I can’t repost these pictures here, but I can provide an excerpt that shows the message. Here it is (click on the picture to enlarge):

Here’s a transcript VinceClorthoEC provided:


In his post, VinceClorthoEC writes:

The sword currently resides in the Museo Napoleonico in Havana, Cuba. It was acquired by Julio Lobo, a Cuban industrialist in the 1940s/50s. His collection of Napoleonica at the time was the largest and considered to be the best outside of France. After the Cuban revolution, his collection formed the vast majority of the new Museo Napoleonico. There’s not much more I can tell you about this specific sabre except that there are 2 or 3 examples of sabres sold at auction with similar styles of writing … but not the same.



The inscription consists of 37 letters and is based on a 24-character alphabet. Considering that this sword is about 200 years old and that sword makers usually weren’t knowledgeable in cryptology, it seems likely that the encryption system used is a simple one. As monoalphabetic substitution (MASC) is an obvious candidate.

Anyway, it might be difficult to solve this cryptogram with standard methods such as frequency analysis and word guessing, as it is short and doesn’t include spaces. In addition, the plaintext language is not known, though French and Latin appear to be options. I could imagine that a hill climbing attack might work if it takes different languages into account.

The fact that there are 24 different letters in such a short cryptogram is unusual. One possible explanation is that diacritic letters such as E, É, and È are encoded in different symbols. If this assumption is correct, Latin and English are unlikely plainext languages, as these don’t have diacritic letters. French, on the other hand, might be a good fit.

Can a reader break this encryption? In addition, I would be interested to know if other swords of this kind are known to exist. According to the Reddit post, this is actually the case, and similar swords were on auction. It goes without saying that additional ciphertext would be helpful for breaking the cipher. Perhaps, a reader can find out more about this.

Further reading:
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 (8)

  1. #1 David Oranchak
    9. Mai 2020

    The code is short so it is easy to find things that fit. For example, here’s a seemingly nonsensical French phrase that seems to fit the constraints of the purported cipher text:


  2. #2 Kerberos
    9. Mai 2020

    “”Here’s a transcript VinceClorthoEC provided:””
    VinceClorthoEC names it a “pattern”
    The sword seems to be “Damast” , or maybe plain
    steel colorized to look like Damast.
    The letters remind me of gothic (Sütterlin) letters.

  3. #3 Norbert
    9. Mai 2020

    @David: Are you sure that your example fits the pattern?

    It provides equal plaintext letters for equal cipher symbols, but shouldn’t we also make sure that different cipher symbols are assigned to different plaintext letters? That is not the case with your text.

    Some other points:
    – The language could be French, but also Latin imho
    – If it’s a MASC, double letters might have been enciphered only once
    – I see the tiny difference between the symbols transcribed as E and B, but they might also be the same

  4. #4 David Oranchak
    9. Mai 2020

    Good point! Looks like I ran it as a homophonic.

  5. #5 Klaus Schmeh
    10. Mai 2020

    Tony Patti via e-mail:

    Kerberos state above that “The sword seems to be ‘Damast’, or maybe plain steel colorized to look like Damast.”

    For those not familiar with “Damascus Steel”, please see https://en.wikipedia.org/wiki/Damascus_steel

    If you look at a close-up of the sword, for example at


    you will see (especially on the bottom edge) that the pattern does *NOT* extend to the edge.

    The pattern continuing to the edge (however faintly) which would be the hallmark of true Damascus Steel.

    As such, the pattern is painted/fired on, including the writing.

  6. #6 hias
    10. Mai 2020

    It looks to me like every symbol consists of two letters

  7. #7 Klaus Schmeh
    10. Mai 2020

    Puzzled Catepillar via Facebook:
    My first thought would be to look at the style of sword itself for clues about who might have owned it. Was it an upper Officers in shape, intended to be used vs. purely ceremonial but meant to be referred to as a Cryptologic pangram key (one of a few?) and *not* a message in itself? My first impressions — and depending on the date it was forged, what did/didn’t it have to do with the “Cipher of Paris” the Cryptographers under Napoleon cam up with …

  8. #8 Norbert
    Berlin Homeoffice
    11. Mai 2020

    @Puzzled Caterpillar:

    The “Great Paris Cipher” of 1811/1812 (I think you refer to that) was a nomenclator based on the “Great Cipher” which had been invented by the Rossignols in the late 17th century. It used about 1400 different codegroups in form of numbers.

    I see no relation to the inscription on the sword. Such an inscription doesn’t need a sophisticated cipher system. (It’s more the look of the symbols themselves that adds an archaic, magic air to the weapon.)