Re: VMs: A couple of questions.

["Elmar Vogt" <elvogt@xxxxxxx>]

> --- Ursprüngliche Nachricht ---
> Von: "Smári P. McCarthy" <spm@xxxxxxxxxxxxx>
> An: vms-list@xxxxxxxxxxx
> Betreff: VMs: A couple of questions.
> Datum: Mon, 18 Jul 2005 00:07:38 +0000
> 
> Hi all,
> 
> ...

Hiyas!

> >From what I can see, the VMS doesn't seem to agglutinate, at least not
> regularly (possible prefixing?), ...

>From what we know, I'd say, "Yes and No". There obviously is a class of pre-
and suffixes as opposed to stems/roots (see Jorge Stolfi's pages on the
"crust-mantle-core" paradigm), but it's not clear what their purpose is.

> 
> Regarding the repetitions of words, some languages like Hawai'ian use
> repetitions to make words (adjectives primarily) more important, ...

Note that there seem to be also sequences of *almost* identical words!

> 
> A few questions though:
> 	- has anybody done a comparison with the number of characters in the
> Voynich character set and other Eurasian languages?

There is no universal agreement how many different VM characters there are.
Similar shapes might be variations of one letter, or actually two distinct
characters. Certain shapes could be ligatures of two letters, or one letter
in its own right. (If we knew for sure about the alphabet, I think solving
the VM would be a piece of cake! ;-)

In general people think though that the letter count is fairly low -- about
20 different, with some fairly rare "weirdos". Compare this to 24-30 latin
letters, or somewhat above 30 for cyrillic.

>...
> 	- has the character set been identified as alphabetical, abjadic,
> syllabilic, etc? I see that people are working on the assumption of an
> alphabet.

Virtually nothing is certain about the VM, but the trend is to assume the VM
script is letter-based, due to the low number of identified characters.

> 	- are the letter assignments made by transcriptors based on anything
> other than whim?

Do you mean the transcribed document itself, or the method of transcription?

AFAIK the EVA transcription (the most-widely used) tried to keep a
correspondence between the shape of the transcribed letter and its
equivalent while at the same time maintaining "readability." (Ie you can
read an EVA-transcribed text sample vocally.)

But the transcription makes few assumptions on the plaintext or the method
of encipherment.

> 	- could it be that the words have been encoded using some primative
> method to obscure the meaning further?

Certainly. One idea which keeps coming up is that the VM was actually
encoded by means of a codebook (ie a list of correspondences plaintext word
<--> encoded symbol or a particular number). The number would then be
enciphered in a second step.

This was motivated by the fact that the particular letter arrangement rules
for words present in the VM seem to resemble the Roman numbering system. As
of now, nobody's found a consistent numbering system though. (And it
wouldn't explain why we find three identical words in a row.)

> 	- I have read that the Ziph values are typical Indoeuropean languages,
> but are they close to any other language groups?

I'd be very careful with that statement, because AFAICT Zipf's laws are a
generally fairly poorly understood concept.

Entropy analyses seem to indicate a relatively low information content (high
predictability) for the VM. This would point to --

*) It's a polynesian language,

*) We don't transcribe properly, but miss important details, or

*) It's not a letter-based script at all.

Welcome to the whacky world of the VM! :-)

> 	- What information is available besides that which is on voynich.nu?

http://math.ucr.edu/home/baez/voynich.html

Holds a number of links.

Also, don't forget to check out the Wikipedia page, which is very good,
IMHO!

Cheerio,

   Elmar

-- 
5 GB Mailbox, 50 FreeSMS http://www.gmx.net/de/go/promail
+++ GMX - die erste Adresse f�r Mail, Message, More +++
______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list