Voynich Welsh Numerology

[stellar]

JKP,

I'm sure if we were to run a frequency analysis this would look like a language yet as you know too hard to break without a crib.  I made this exercise for everyone in 3 hours time.  Yet once you get the feel you can make up words at a fast clip.

Notice that when I used the Pythagorean number system table I managed to make it look like a language in just 1 paragraph thus having a frequency set.

Using this system it quite easy to shorten long words and still maintain the look and feel of a language.  The system I use is kinda of a grill like Gordon Rugg suggests.

An interesting comparison would be is to know the average length of a voynich vord and compare with my little paragraph to see if I'm on to something.  Another fact is that the Voynich Manuscript could be a copy of another book as it was easy for me to just make up glyph's and use the Pythagorean table.

Just looking at my paragraph it would seem that I have almost an exact match for 5.5 average.
let me do some math 
Here is my calculation for the paragraph.

1(8 letter word) 1 (7 letter word) 7(6 letter words) 6 (5 letter words) 8 ( 4 letter words) 9 ( 3 letter words) 13 ( 2 letter words) 7 (1 letter words)

54 total vords

                                    1+1+7+6+8+9+13+7 
___________________________________________________= 6.5 average

                                              8          
After recalculating I'm of by a whole word of 1!  Maybe there is something to be said about math and language combined and some sort of universal thinking using the two.

 The average word length for voynich glyph's is 5.5.
You are not allowed to view links. Register or Login to view.

CRIB of words taken from JKP's site:

When I first saw Dee’s handwriting, I noticed similarities between it and the VMS columnar text, but after sampling the handwriting of other writers, it appears that this style of script was widespread geographically even if it was not entirely common (I encountered many other styles in the search for this handful of samples).

See cipher: