Cryptoquip Typo

I think I found a typo in a Cryptoquip.

This one is a “celebrity cipher” but they’re so much alike you’d think that one would have sued the other out of existence, so I presume they’re actually done by the same people.

I think this ciphertext has a typo:

celebrity cipher with a typo

That’s right a typo, not a corrupted ciphertext. Can you spot it in the image? That’s a scan from the February 15, 2023 Fort Collins Coloradoan newspaper.

I used my cryptoquip solver to find the plaintext associated with the above ciphertext.

Solved letters

Cipher letters on top, cleartext letters below:

a b c d e g h i j k l m n o p r t v y z 
h s n k m v d w t e u i l a r f ? g o p 

There’s a ciphertext letter ’t’ for which the solver can’t find a plaintext letter. The plaintext word is “be”. The ciphertext letter ’t’ only occurs once in the puzzle. I think the solver can’t tell if the ciphertext word is “be” or “ye”. There’s no way for the solver to tell if the cleartext word is “be” or “ye”. There’s no other clear letters ‘b’ or ‘y’ in the solution. That’s a legitimate place for the solver to indicate it has no solution.

Solved Puzzle

eybj vpyic lzb rypvkj iaoj mj iob nmdk jy tk o dmh m gyikh jaoj m iylnh ckgkp
most grown ups forget what it was like to ?e a kid i vowed that i would never

rypvkj jak bmezbycb apkojyp
forget the simpsons hreator

I left off “Matt Groening” which enciphers to “eojj vpykcmcv” above. “Groening” isn’t in my dictionary, and its shape (“12345657”) is shared with 4224 words that do appear in my dictionary. It only causes the solver to indicate more ambiguous letters.

The typo is in the ciphertext word “apkojyp”. It deciphers to “hreator” which is pretty obviously “creator”. My solver isn’t finding the plaintext ‘h’ incorrectly - ciphertext ‘a’ appears in cipher words “iaoj”, “jaoj” and “jak”, which decipher to “what”, “that” and “the”, respectively. Those 3 deciphers are correct.

The ciphertext word “apkojyp” has (what I’ve called) a shape “0123451”, “creator” has the shape “0123451”, as do 290 other dictionary words in the file I use to solve these ciphers. The 291 dictionary words narrow the cipher letter ‘a’ in “apkojyp” to one of the cleartext letters: a b c d e f g h i k l m n o p r s t u v w z Since cipher letter ‘a’ corresponds to ‘h’ in 3 other cipher words, “hreator” slips through.

The ‘a’ in “apkojyp” is incorrect. It should be another cipher letter.

It seems unlikely that Luis Campos typed “hreator” for the plaintext before enciphering. Campos mis-typed or otherwise typoed the “a” when transcribing the ciphertext. The unused cipher letters in this puzzle are ‘f’, ‘q’, ’s’, ‘w’, ‘x’. ‘q’ and ’s’ are adjacent to ‘a’ on a standard QWERTY keyboard, I’m going to assume that Campos typed ‘a’ instead of ‘q’.

If I put in a ciphertext ‘q’ (otherwise unused in the ciphertext), my solver finds “creator”.

This is weird, a typo in ciphertext. It’s extremely difficult to retype the newspaper’s enciphered text. I often make mistakes doing that, which usually causes the solver to find gibberish as the cleartext, or to find only partial cleartexts. This phenomenon is related to the difficult task of deliberately reproducing typos, but it’s not the same.

As far as I’m aware, there’s no name for making typographical mistakes when typing ciphertext. I’ve used “typo” to cover this gap, but “typo” doesn’t really match the situation, since the cipher as end product is as close to gibberish as you can get.

See the cryptoquip construction post, which may explain a little more about how my solver works.