Discussion:
cipher of historical interest
(too old to reply)
Michael Friendly
2009-07-14 15:15:00 UTC
Permalink
Raw Message
Hello

I have little detailed knowledge of cryptography, but in my research I have
encountered a ciphered text that is of great historical interest and would
like some assistance or suggestions about how to go about decoding it. I'd be
very grateful for any help or pointers to other mailing lists, sources of
expertise, software, etc. One source of difficulty is that the plaintext, if
there is any, would be in old Spanish.

The text appears in a 1644 publication by Michael Florent van Langren titled
"La verdadana de longitude par mer y terra". In this, he claims to have
discovered a new, improved method to determine longitude more accurately than
before, a topic that was of major interest in all the courts of Europe. But,
so that no one would steal his method, he says he is describing it in code,
that he will reveal to his patron, King Phillip of Spain. Almost certainly,
the method concerns some variation on what is called the 'lunar method',
whereby time difference between two places can be determined by measuring
angles of lunar events and translating time difference into degrees longitude.

Decoding this text is the final puzzle piece in some historical research on
which I've been working for the last year. It is very important for me to
understand whether this was in fact the 'secret' he was guarding, or perhaps
just a ruse to make it look like he had discovered something new.

I've placed as fair a transcribed copy as I can at:
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/cipher.pdf
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/cipher.doc

For context, and an image of the original manuscript portion, see:
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/Verdadera-transcription-p6-8.pdf

Again, thanks in advance for any help with this.

-Michael
Ilmari Karonen
2009-07-18 04:17:54 UTC
Permalink
Raw Message
Post by Michael Friendly
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/cipher.pdf
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/cipher.doc
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/Verdadera-tran
scription-p6-8.pdf

The letter frequencies in the enciphered text on page 8 seem to match
those of the plain text on page 6 quite closely, suggesting that this
is very probably some kind of rearrangement cipher.

The distribution on the numbers mixed in the text is quite curious:
the numbers 5 and 9 occur very frequently, while 0, 1 and 2 never
occur at all (and 2 and 6 each only once). It's possible that this
merely reflects the occurrence of said numbers in the plaintext, but
another possibility is that the numbers might also stand for letters:
indeed, both the simple mapping 1=a, 2=b, etc., as well as it inverse
9=a, 8=b, etc., would map both 5 and 9 to vowels. I'm not sure what
the point of such a substitution would be, though, unless it's simply
to make the cipher look more cryptic.

--
Ilmari Karonen
To reply by e-mail, please replace ".invalid" with ".net" in address.
Michael Friendly
2009-07-27 15:24:54 UTC
Permalink
Raw Message
Thanks for your kind reply and initial analysis of the letter
frequencies. I suppose that a longer
portion of the Spanish plaintext would help confirm your idea that he
is not using a substitution cipher.

One guess is that the digits might be dates, but then, I'd expect 1
and {5,6} to be the most frequent,
and this is not borne out in the ciphertext. The frequent use of
capital letters also suggests that
these may be the initial letters of names, again, just a guess.

Any further suggestions are most welcome.

-Michael
Post by Michael Friendly
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/ciphe...
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/ciphe...
http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/Verda...
scription-p6-8.pdf
The letter frequencies in the enciphered text on page 8 seem to match
those of the plain text on page 6 quite closely, suggesting that this
is very probably some kind of rearrangement cipher.
the numbers 5 and 9 occur very frequently, while 0, 1 and 2 never
occur at all (and 2 and 6 each only once). =A0It's possible that this
merely reflects the occurrence of said numbers in the plaintext, but
indeed, both the simple mapping 1=3Da, 2=3Db, etc., as well as it inverse
9=3Da, 8=3Db, etc., would map both 5 and 9 to vowels. =A0I'm not sure what
the point of such a substitution would be, though, unless it's simply
to make the cipher look more cryptic.
--
Ilmari Karonen
To reply by e-mail, please replace ".invalid" with ".net" in address.
Loading...