Re: VMs: Why VMs spaces behave like normal word spaces

[Nick Pelling <incoming@xxxxxxxxxxxxxxxxx>]

Hi Rene,

> This very property is
> exploited (in a way) by Sukhotin's algorithm to
> separate vowels and consonants. Interestingly,
> the space character has a variation in preceding
> and  following characters which is marginally
> higher still than that of vowels.
>
> That last property is also observed in the Voynich
> MS. I made the calculation many years ago.
> I made a 'scatter plot' (every point representing
> a character of the alphabet) plotting conditional
> entropy of the preceding character vs. that of
> the following character. The character that
> appeared furthest to the upper right was the space
> character, both in Latin and in Voynichese.
>
> So, I don't see anything particularly suspicious in
> the VMs spaces.

Unless you can compare data-sets across a time-dimension, statistics talk 
of correlation, not of causality. As I said (towards the start of the 
thread), I'd be extremely impressed if anyone could find a way to prove VMS 
spaces are either real or artificial - for now, the jury is out (and have 
ordered sandwiches for the next six months - but luckily they're in 
Scotland, so they have the option of "not proven"). :-)

Alternatively, the postmodernist (well, Derridan rather than Lyotardian) 
view would be to say that anything so close to both signal and noise would 
have to be some kind of function of both - but why would a language (or 
transcription) need to include noise? For a postmodern analyst, then, I 
think the VMS would have to be a combination of code (signal) and hoax (noise).

Just my $0.02 worth (or should that be euros now?)

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


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