Re: VMs: RE: Word pairs

[J R Elliott <jre@xxxxxxxxxxxxxxxx>]

Greetings all,

I am, at present, in the final 'throws' of writing-up my PhD thesis on
"Detecting language-like features in 'signals'"; or to put it another way:
Language discovery and Universals.  I have been aware of the Voynich
manuscript for some time now and have consequently performed analysis on
it, as part of my work.  This includes Entropy (H1:H5), Ngrams, word pairs
etc., etc.  In fact a suite of programs from a computational linguists
point of view.

My general conclusions are that if the Voynich manuscript is a language,
then it is encoded in some format and probably 'padded' to 'mask' its true
meaning, rather than a letter substitution cipher.  Evidence from my
analysis seems to support this at all levels, as results consistently
record atypical phenomena, such as word-pairs mentioned recently: in this
case, word-pairs not 'mirroring' typical distribtion patterns.   My
'gut' feeling based on computational evidence, is that the script is
encrypted using a cipher, as I mentioned earlier, or more likely an
artificial language constructed for "mischief, such as academic fraud for
purposes of ingratiating the author, or fiscal reward".  I emphasise the
latter, as the behaviour of the script seems to have all the hallmarks of
artifical creation: symbol frequency, combinations and word statistics
etc.

I did not wish to present such a negative conclusion but the dispationate
computational evidence, across a variety of metrics, using a suite of
tools I developed for analysing represenative languages (45 in total)from
all the major language families, compels me to this: I have also included
logographic and syllabic scripts.  I would also say, that if it is indeed
encoded, then the statistics should behave much more 'noise-like', like
any efficient system.

I will post my thesis onto my web page post-viva, for your inspection and
hope, at least, that it contributes towards the ultimate aim: of
understanding what the manuscript  contains.  Nevertheless, I hope
subsequent research proves that the manuscript is a bona fide language,
which has been cleverly encoded to present us with a useful conundrum.

Very best regards to all,

John
*********************************************************
John Elliott
Centre for Computer Analysis of Language and Speech
University of Leeds.  http://www.comp.leeds.ac.uk/jre/
and Computational Intelligence Group, School of Computing
Leeds Metropolitan University
email:  jre@xxxxxxxxxxxxxxxx  or J.Elliott@xxxxxxxxx
Home: 0113 286 6517 john.elliott@xxxxxxxxxxxxxxxxxx
*********************************************************


On Tue, 8 Jul 2003, Rene Zandbergen wrote:

>
> --- GC <glenclaston@xxxxxxxxxxx> wrote:
>
> > I can't give you a count of the entire manuscript as
> > yet, nor can I provide
> > this in EVA, but here you will find a list of word
> > pairs for the first six
> > quires represented by their font glyphs.
>
>  [...]
>
> > http://voynich.info/vgbt/xcrptn/word_pairs.pdf
>
> Very interesting!
> That has a very nice Zipf shape about it.
> Really, this should be compared with similar
> results based on known-language texts of a
> similar length, in order to draw conclusions.
>
> Cheers, Rene
>
>
> __________________________________
> Do you Yahoo!?
> SBC Yahoo! DSL - Now only $29.95 per month!
> http://sbc.yahoo.com
> ______________________________________________________________________
> To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
> unsubscribe vms-list
>

-- 
*********************************************************
John Elliott
Centre for Computer Analysis of Language and Speech
University of Leeds.  http://www.comp.leeds.ac.uk/jre/
and Computational Intelligence Group, School of Computing
Leeds Metropolitan University
email:  jre@xxxxxxxxxxxxxxxx  or J.Elliott@xxxxxxxxx
Home: 0113 286 6517 john.elliott@xxxxxxxxxxxxxxxxxx
*********************************************************



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