In 1991, an unknown person sent a number of letters to the presenter of the TV show “America’s Most Wanted”. The anonymous writer claimed to have committed 23 crimes. Some of the letters were encrypted in a similar style as the Zodiac Killer messages.

Click here for the complete top 50 list

Did you know that one of the USA’s most successful TV shows had its origin in Germany? I’m talking about America’s Most Wanted, which was at the time of its cancellation in 2011 the longest-running program (25 seasons) in the history of the Fox television network. This mark was since surpassed only by The Simpsons. America’s Most Wanted reported on unsolved crimes and asked the viewers for assistance in solving these. Over 1,200 people were captured because of the show.

The concept for America’s Most Wanted came from the German TV show, Aktenzeichen XY … ungelöst (German for “File Reference XY … Unsolved”), which is certainly known to many readers of this blog. One of the most mysterious cases Aktenzeichen XY … ungelöst reported on was the YOGTZE case, which involves an unsolved cryptogram.


Mysterious letters sent to the host

America’s Most Wanted was hosted by John Walsh (born in 1945). In 1991, Walsh received a number of anonymous letters. The sender, who called himself “Scorpion”, claimed to have committed 23 crimes – including robberies and murders. Five letter excerpts were published. Here’s the first one (I took all the scans from Dave Oranchaks’s Zodiac Killer site):


Two of the excerpts show encrypted messages (there were three more messages of this kind, but police kept them secret). Apparently, the sender of these ciphertexts was inspired by the Zodiac Killer. Here’s the first encrypted passage:


Here’s the second cryptogram that was published:


Here’s another cleartext excerpt:


And finally, here is the last letter passage that is publicly known:


There is no evidence that the author of these letters is identical to the Zodiac Killer. It is very unlikely that this person’s claim to have committed 23 major crimes is true.


Can it be solved?

There’s no doubt that the Scorpion cryptograms are hard to solve. What makes cryptanalysis difficult are the large alphabets. The first ciphertext contains 70 characters, 53 of which are unique. 155 of the 180 symbols contained in the second message are unique. This means that a frequency analysis is as good a useless.

If the Scorpion cryptograms are real ciphertexts, they were probably created with a homophonic cipher (i.e., a cipher that provides several cleartext equivalents for some letters). The first Zodiac cryptogram was created this way. It is possible to solve cryptograms of this type (in fact, the first Zodiac cryptogram was deciphered by Don and Betsy Harden), but it is very difficult with such short messages. Breaking a homophonic cipher is a little easier if there are spaces between the words, but this is not the case here.

All in all, I don’t think that the Scorpion cryptograms can be broken without further information. Especially, having access to the three unpublished ciphertexts would be very helpful. Or perhaps, the creator of these cryptograms used the same encryption system somewhere else.

If you have an idea how this difficult crypto mystery could be solved, please let me know.

Further reading: The Zodiac Killer code is solved – at least in this new movie


Subscribe to Blog via Email

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

Kommentare (7)

  1. #1 Gerry and Andrea
    31. Januar 2018

    The two cleartexts can be ordered the other way around, so they form a sentence in the middle: … many hundreds of hours of mindracking experimentation …

  2. #2 Klaus Schmeh
    31. Januar 2018

    Bart Wenmeckers via Facebook:
    For some reason i thought these ciphers were already solved, thanks for the post.
    The fact these have such high unique symbols to cipher text lengths makes them difficult indeed to solve. With copycat ciphers i often wonder if they are even real ciphers or just some bogus rubbish to waste peoples time. I wonder that about Z340 some days 🙂

  3. #3 Peter
    4. Februar 2018

    Hallo Klaus
    Ich weiss nicht ob Du das schon gesehen hast, aber ich konnte es mir jetzt nicht verkneifen dir den Link zu senden.

  4. #4 Marthy
    Study how the symbols are written
    29. Juni 2020

    I think, in the cipher there are not unique symbols. Just only very good hidden letters and numbers.You can normally read the cipher, than it is possible to decode writing. Like finding symbols in paintings or photography art. As children we drawed number like a main object and than drawed something else from this(animals, parts of body, geometrical pictures, ) . It was very confusing, to find the hidden original symbol, especially when it was made with 1 colour. Now I learn this my children.

  5. #5 Marthy
    Use your eyes
    31. Oktober 2020

    Use your eyes. In the first picture you can see hidden letters. In second one, it is more difficult. But if you use your eyes, you see rotated marks, doubled letters, 2 marks put in one. I thing the first step is realize how are unique marks made. This cipher is not too much about frequency analyse, mathematical and logical thinking. The more about watching. and word puzzle.

  6. #6 Septimius Severus
    A quite different view
    16. Dezember 2021

    „having access to the three unpublished ciphertexts would be very helpful”.

    Should be „four”, not „three”.

    „155 of the 180 symbols contained in the second message are unique”

    Actually 145 characters are unique, not 155.
    8 characters were used 3 times, 19 were used 2 times, 118 were used 1 time. 62 repeats.

    This mistake is on the Cipher Mysteries blog. Craig Bauer replicated the mistake in „Unsolved!: The History and Mystery of the World’s Greatest Ciphers from …” (2017, page 224).

    It’s not so easy to swiftly devise more than one hundred unique characters. But to create a shape „families” is an easy way to generate new signs. And there is something else. Lets assume that the author created the characters during the S5 writing/drawing process. A few first symbols should be deviced „from scratch”. We can suppose that subsequently the man will model the next characters after previously written ones. Not always but rather often, consciously or unconsciously. An unintended effect can emerge: especially in the upper part of the cipher (excluding a few unitial characters) similar (in some way) shapes statistically will be located closer to each other. (I assume here „normal” writing direction: from left to right and from top to bottom). When more rows of signs are created, there is more sources to imitate in the creator’s sight. So he/she can be inspired by the shapes located much farther. In the S5 some characters should be excluded from this kind of observations because of the repeat spans pattern. I’m not sure is this a true in the Scorpion 5 cipher or is this my delusion only. But if the shapes really were created during the drawing it can means:

    a) the S5 is a hoax,
    b) there is something diffent going on.

    I don’t know what the S5 is. But it is possible to deploy a way of encryption other than ordinary homophonic cipher or something inspired by Vigenere. The example is below.

    I did a „mindracking experimentation” and created a cipher using the same or similar characters as in the original Scorpion 5. 180 characters in total, 145 are unique. 8 were used 3 times, 19 were used 2 times, 118 were used one time. And „every single shape repeat spans a distance that is a multiple of 16” – just like in the S5 (but it does’t matter). Simple, logical method. There is a real message. Any ideas how it works?

  7. #7 Septimius Severus
    29. Dezember 2021

    There is a hypothesis about 16 alphabets and some kind of grid logic based of „shape families”. Or perhaps The Scorpion wrote the plaintext on 16 column grid first and then created different but similar signs for each letter according to the rule that the same signs may repeats in the same column only. If this is a case, it may be a weak point of the cipher and it seems inconsistent with the claims about „many hundreds of hours of mindracking experimentation” and „a lot of time and effort”. On the other hand there is no obvious way to group all the signs into „families”. Some of them are easy to classify, some not.

    We can find some „bigrams”, it means pairs of „possible one family shapes” but it may be also a coincidence or the effect of a copycatting the previous draw characters.