Re: VMs: Word Length Distribution

[Dennis <tsalagi@xxxxxxxx>]

knoxmix@xxxxxxxxxxxxx wrote:
> 
> Maybe this has been exhausted on the list in the past but I will
> bring it up again anyway. What could account for the binomial
> distribution of vocabulary words as shown by Jorge Stolfi?
> 
> http://www.dcc.unicamp.br/~stolfi/voynich/00-12-21-word-length-distr/
> 
> I think this is an acid test for any scheme that someone might
> devise. 

	I agree.  This is one of the most significant,
leaps-off-the-page-at-you, as-obvious-as-a-sore-thumb
things I've ever seen about the VMs!  

> Does it show that the Currier transcription is pretty close
> to the mark or is it a function of the transcription? 

	In fact I see Stolfi used EVA.  I wonder what you
would see if you converted it to Currier, as one can
easily do with BITRANS?  Could this be related to the
fact that EVA represents some common glyphs with more
than one letter?  Currier doesn't do this.  

> I ran a check
> on a long section, almost 8000 tokens, of an unmodified EVA
> transcription that I have been working with and it showed almost
> identical results. It will  be interesting to see whether this holds
> with shorter sections and if not, where and how it breaks. 

	Stolfi used the entire VMs; look at his tables.  

> Does it
> vary from one section to another?  

	Good question, but I doubt it does. 

> Is it consistent with a real
> vocabulary in the writing about certain subjects or with any known
> specific cipher? Who would have such a vocabulary? Could (the)
> elimination of certain (sometimes) unessential parts of speech or
> letters explain it? (Shorthand, Nick?) Pidgin? Maybe there was a
> European Pidgin that became obsolete.

	He did compare it with English and Latin texts, so it
must not be specific vocabulary.  

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