Re: VMs: RE: An Experiment in Retranscribing the Voynich Manuscript (long)

[Koontz John E <John.Koontz@xxxxxxxxxxxx>]

On Wed, 1 Sep 2004, John Grove wrote:
> 	I've been through some of this before and somehow convinced myself
> that it wasn't going anywhere when I couldn't account for the extremely
> limited set of consonants this approach leaves you with. I also tried
> the reverse... the ligatures were the consonants and the vowels were the
> starters, throwing in a default vowel; 'h' were diphthongs...

Objections noted and agreed with.  The other stuff sounds familiar, too!
I'm obviously interested to hear that others have carried things this far,
though the basic ideas all seem to have been around for a long time.  The
availability of things like the Takahashi transcription file (which
depends on the existence of the advanced EVA transcription scheme) and
easy access to simple text manipulation tools make it easier to follow up
on these ideas than it was for the wartime and postwar investigators.

> 	As in your example, the consonants are limited to trmhlp and
> you'll see that 't' is used far too often.

And in the first whack at this example I reversed p and t.  As it is, it
does sort of come across as a Finnish tongue-twister, doesn't it?  Of
course, there's a good chance that the repeated sho-forms really do
involve repeated roots, always supposing that the VMs isn't the world's
longest lorem ipsum text and that the concept of root makes some sort of
sense.  A possibility that occurs to me as meeting the root structure of
last bit would be something along the lines of "king of kings enters unto
his kingdom," though I don't mean to suggest this has anything to do with
the actual content.

> What we need is a way to increase the number of consonants. Vowel sets
> could obviously be as short as 3, but consonants will need at least a
> well-balanced distribution of '7' letters.

Well, the consonants of Latin are really just pbfm tdsn cg hrl, which is a
pretty sparse system.  Q and x are kw and ks.  V/w and y are written with
u and i.  H might be regarded as silent by some later writers.  Some
additional letters are used to write Greek loans.  Medieval and modern
Western European languages tends to be rather more complex and prolific
phonologically.  We're pretty close to having enough symbols, but the
frequency distributions do seem to be a problem.  It makes you wonder if
some subtle points have been missed in the transcription, though I
rather doubt that.

I've done some tentative experiments with trying to render Latin into
Voynich, trying to put my finger on what's wrong, but haven't gotten far
with them.

> unfortunately this is the result of tossing in arbitrary consonants without
> any rules, but the idea of assigning vowel/consonants to the glyphs has this
> major (in my opinion) flaw.

Yes, I didn't put any care into selecting substitutions here, of course,
since I just wanted to provide something that was a bit easier to grasp
mentally than CVCVCV.  The latter also tends to hide the sameness of
consonants by exagerating it.

> I'd love to hear a workable rule that helps increase the consonant set.

Well, if you leave o as o, then you get oae+ and oai+ sets, etc., but I
didn't think the distributions were remarkably better, and o(o) seemed to
be pretty freely occurring on the initial boundary of the a?(e+|i+)
sequences, so I figured it was more likely to be another CV syllable.

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