VMs: Gallows mega-theory...

[Nick Pelling <incoming@xxxxxxxxxxxxxxxxx>]

Hi everyone,

Here's my gallows mega-theory, which I've been piecing together for the 
last 6 months. Any suggestions as to how this might be turned into testable 
hypotheses will be gratefully received! :-)

Essentially, there are two frequently-appearing types of gallows - 
t-gallows and k-gallows. We also see a fair number of p-gallows, which 
often appear in stylistically well-defined pairs (typically two-thirds 
along the top line of a paragraph/page) containing anomalous text, known as 
"Neal keys". Another well-known anomaly is that the behaviour of the first 
letter of each line of text can be unpredictable - often looking as though 
it is "glued" onto the start of a "real" word. Gallows (particularly t- and 
k-) also have a consistently high appearance rate across all lines of all 
folios. This theory seeks to build a coherent explanation for gallows, 
linking all these properties together.

My suggestion is that p-gallows pairs [ie, horizontal Neal keys] (and 
occasionally f-gallows), and sequences of line-initial characters [ie, 
vertical Neal keys] may well define *two-character substitution ciphers, 
local to a single line*, while the (more frequent) t- and k-gallows *use or 
refer to* those local ciphers.

You might alternatively understand this as a *local steganographic 
transposition cipher*, whereby the two most prominent letters (perhaps 
consonants?) within a line are extracted, placed elsewhere on the page (ie 
within horizontal or vertical Neal keys), and then replaced with either 
t-gallows or k-gallows.

That is, if the two most common word-initial letters (perhaps disregarding 
common prepositions like "con-") for a particular line are "R" and "S", 
then *for that line only* t-gallows might well stand in for "R" and 
k-gallows for "S". To complete the picture, "R" and "S" would be encoded, 
one as a a line-initial character and the other sequentially within a 
horizontal Neal key (though ATM I don't know which way round these go).

Let's deconstruct a typical page with a well-defined horizontal key - the 
split gallows on f35r:-

cth [oo r choly] cthy choty char dy -	// Neal key bracketed, ornate t-gallows
qokeeaiin chokaiin qotchy daiin -
dchaiin cthey qotchey taiin cthory -
qotchy shetchy ckhol cheey daiin dainl -
otchor sho tcheey scheey daiin dain or -
schaiin char chan daiin -
shosaiin tchor choky -
qokchy kykaiin daiin =

I suspect that this top-line (horizontal) Neal key actually contains 8 
characters in an only-very-slightly verbose cipher (where "or"/"ol"/"qo" 
code for individual letters) (naturally, "ch" and "sh" are individual 
letters too), each corresponding to a separate line of the (8-line) paragraph:-
ch	o	or	ch	ol	y	ch	y

The line-initial characters (and hence the 8-element vertical key) are:-
(ch?)	qo	d	qo	o	s	sh	qo

However, if the encoder wanted to refer to both horizontal and vertical 
keys within the very first line, the system would not allow the horizontal 
Neal key to begin at the left edge of the line... I strongly suspect that 
this is why the Neal key typically starts two-thirds of the way along the 
top line, so as to prevent the two keys overlapping at the top left character.

I believe that all this is sufficient to explain Neal keys (their contents 
and position), split gallows, single-leg gallows, and line-initial 
anomalies... but I still can't read it. I therefore suspect that these two 
keys are encoded via (at least one) separate (global) replacement ciphers.

My best current guess is that these ciphers (for the horizontal and 
vertical keys) are trickily encoded in the various zodiacal volvelle 
labels... but perhaps we can work them out regardless. :-)

While this isn't actually a (computationally) complex system per se, its 
property of locality (which gives two replacements within a line) would be 
extremely cryptologist-unfriendly - especially when combined (as I believe 
it is) with other simple ciphers and steganographic practices.

What do you think? :-)

Cheers, .....Nick Pelling.....

PS: I don't believe there are any cryptographic / historical precedents for 
this - unless you know differently? :-o 


______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list