[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

VMs: King Tut Word Game, or the EKT Hypothesis

	Very early in the history of the list, Andras Kornai
suggested that a word game like Pig Latin in 
English could explain the low entropies of Voynich
text.  I became interested in the King Tut word game:

A - a                   I - i           R - rur
B - bub                 J - jug         S - sus
C - cut                 K - kam         T - tut
D - dud                 L - lul         U - u
E - e                   M - mum         V - vuv
F - fuf                 N - num         W - wuv
G - gug                 O - o           Y - yec
H - hush                P - pup         Z - zuz

Q (as is)
X (as is)

- from *The Cat's Elbow and other Secret Languages*,
collected by
  Alvin Schwartz and pictures by Margot Zemach, 1982; a
children's book,
  although it also has good scholarly references.  King
Tut or
  Double Dutch, p. 45-47.

	Thus one substitutes a different CVC combination for
each consonant.  The subsituents could be short words. 
Also, there could be more than one substituent for each
consonant; the operator's personal choice of
substituents could explain the difference between
Voynich A and B.  

	However, if one applied such a system directly to
words, one would get long words, much longer than one
sees in the VMs.  It could be that the underlying
source is divided up into syllables.  This would allow
the underlying language to be a common European one.  

	I've elaborated this idea in more detail as the EKT
Hypothesis:
http://www.geocities.com/ctesibos/voynich/ekt.txt

	In addition, if we assume that the underlying text is
divided into syllables, that would make French a good
candidate for the underlying language, since the spoken
form of French, since medieval times, does not contain
such word markers as syllabic stress, etc., and most
morphemes are bound.  

	What are the arguments against the hypothesis I've
just set out?

1)  *Word Structure and Word Internal Structure of
Voynichese*  I've attached a relevant post by Philip
Neal. I think his points are still valid even with the
King Tut cipher.  

2)  *Number of Voynichese Words*  Gabriel Landini told
me that Voynichese has 8200 words (dictionary entries).
Consider French once again.  The great 19th French
crippie Etienne Bazeries broke Louis XIV's Royal Cipher
and found that it operated on syllables.  The
ciphertext consisted of 3-digit number groups.  Since
some of the groups stood for words or homophonic
equivalents, let us say that French contains 500
syllables for simplicity.  If each consonant had 3 King
Tut substituents, for each CVC syllable we would have 9
possibilities, for each CV syllable 3 possibilities,
and for each VC (word-initial) syllable we would have 3
possibilities.  Thus we might have, say, 6 * 500 = 3000
enciphered 'words', which is at least in the right
neighborhood.

3)  *Consonant-Vowel Alternation of Voynichese Words* 
If the hypothesis is correct in its simple form, we
should see mostly CVCV , CVCVCVC , and VCVC Voynichese
words, while I do not believe we see this sort of
consonant-vowel alternation.  Of course, consonant
clusters could stand for a single consonant phoneme, as
German sch / English sh / Hungarian s   all stand for
the same phoneme.  

	Comments are invited.

Dennis
--- Begin Message --- Dennis wrote

	I must be dense.  Could you be more specific? E.g.,
French, Italian, etc. syllables don't have the
paradigmatic nature of Voynichese words, don't have the
relative small number of endings that lead one to think
of Chinese, have too many/too few open syllables, or
something I can't think of?


I think it can be shown very simply that Voynich 'words'
are not transliterated words or syllables of French, German
or Latin. Consider what letters can occur initially and
finally in words of those languages and which ones can be
doubled.

French

initial		abcdefghijlmnopqrstuvxz
final		acdefgilmnoprstuxz
double		elmnprst

German

initial		abdefghijlmnopqrstuvwz
final		bdefghlmnprstuz
double		abeglmnoprst

Latin

initial		abcdefghijlmnopqrstuvx
final		abcdeilmnorstux
double		bcdfgilmnprstu


In all three of these languages, more or less any letter which
can occur finally in a word can also occur initially and many
letters commonly occur as double letters. Furthermore, the
letters which occur as doubles are also common initially.

In Voynichese, only c and (if it is an independent letter) i
occur as double letters at all frequently. i is never initial
and c seldom initial. Furthermore, Voynichese (a)in and (a)iin
are almost exclusively found in final position and never occur
initially.

There are other patterns in Voynich 'words' which have no
parallel in these three languages. For instance only a minority
of Voynich characters ever occur twice in the same word, and
most characters have a strong preference for a fixed position in
a word relative to other characters (e.g. where a word contains
both c and k, the c almost always comes between the k and the
end of the word).

French, German and Latin are my strongest foreign languages, but
I have never noticed the Voynich patterns in languages like
Italian which I know less well.

Philip Neal



_________________________________________________________________
Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

---
[This E-mail has been scanned for viruses and found to be virus free]



--- End Message ---