Re: VMs: Re: Moot points, getting long

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

On Tue, 10 Aug 2004, GC wrote:
> In this case we do actually have an informant.  One only has to go back to
> D'Imperio to realize that this is a contrived script, not natural, and could
> be contrived for only one purpose, to hide something beneath.  Further
> analysis indicates that the contrivances are numbered in form and variance,
> a further example that the underlying language is familiar, and the method
> of obfuscation is mathematically driven.  What is left in this case is to
> record the glyphs as faithfully as possible and observe how the rarer glyphs
> move, and extend that movement to the surrounding glyphs.

We have an informant of sorts, in that a correct series of observations
and analyses should lead to a deciperment, but I'm not sure I would call
the process you suggest "use of an informant."  It amounts to pattern
matching of a sort.  You deduce by various evidences that the author(s)
were working in certain orthographic traditions albeit innovatively and
encrypting something within a certain range of possibilities and proceed
by trying to deduce their practices.

This is like recognizing a language as Algonquian and deducing from that
what sort phonology it probably has, and so on.  There are a number of
"extinct" American languages known along just these lines as a result of
being recorded at an early date.  Today there are no speakers, but it is
often possible to deduce a good deal of their phonology and grammar from
these texts.  In virtually all cases the meanings of the texts are more or
less understood, though well-glossed cases trail off into those in which
one has to guess the meanings by comparing forms with related languages.
However, the transcription is usually more or less defective and the
grammar may be very imperfectly known.  Nonetheless with a sample of a
reasonable size it is possible to deduce the phonology and grammar even
though no informant is available to confirm various points.  Obviously a
fluent informant - one who "informs" the process of analysis - makes the
process easier and the results more accurate and complete.

> I doubt that a 'significant portion' of us will ever agree on anything, but
> one thing we all agree on is that when a solution is presented, it must pass
> the test.  This is the only convincing evidence that may be presented.

The only absolutely convincing evidence, but in default of a solution,
consensuses as to what is plausible will no doubt continue to form, and I
believe that these consensus, past, present and future, are likely to be
evoluntionarily better, and not simply faddish alternations between
different approaches.

> > In my own fumbling attempts to reproduce the script with pen and pencil
> > ...
>
> Actually, I don't think the author suffered so much.  He would have been
> writing with a quill since at least the age of 8, and it's obvious he was
> familiar with Latin abbreviations and one or more systems of shorthand, as
> demonstrated by the glyphs he chose for his cipher script.

I tend to agree with this, though, of course, there is a certain range of
easily produced signs in any medium and systems using similar media will
probably use similar signs, to the extent the signs are adapted to the
medium.  However, the m-final and the y glyph, though composed of an e and
a final, are probably distinctive enough to suggest that they have their
origins in Latin abbreviations.  As the latter seems to be analytically
two separate graphs or subglyphs in Voynich script, it appears that the
creator of the script probably worked backward from it (and other
elements) to create his system.

> Moreover, the script construction follows the convention of of scripts
> designed for quill writing, in that the strokes are primarily drawn from
> top to bottom right, bottom to top right, and left to right, with the
> exception of closing 'tails', such as those faint traces on the {9}
> glyphs.  Whenever strokes moved in this direction, the ink outlay was
> routinely minimal. These features can be deduced by the fact that
> wherever heavy wet ink crosses it produces a 'draw' that indicates the
> direction of the stroke.  Information on stroke order, direction and
> glyph construction can be extracted from the images, and is therefore
> not a matter of attempting to reconstruct, rather one of simple
> tabulation of observation, ....

I appreciate your descriptions of this, GC, here and previously, and I see
that your counter proposal makes sense in this context.  However, though
my problems with pencil are of a different order and not subject to the
mechanics of using a quil pen, I'd submit that some of the variations in
stroke order you mentioned previously, as well as perhaps some of the
variations in form, could perhaps result from wrestling with a novel
orthographic system, even if the medium was familiar.

Of course, some differences of form might turn out to be allographs, as it
were - insignificant variations in form due to alternative approaches to
production.  Of course, we don't know which, and so I tend to think that
any obvious differences, even if subtle, should be attended to for now.
I know from professional experience (not all of it personal) that it's
easy to overlook length, slight differences in back nasal vowels, pitch
contours, aspiration, etc., that turn out to be fairly important.

> though I happen to enjoy my efforts to write with a quill.  Difficult,
> isn't it?

I'm sorry to say I've never even worked with a flat metal nib.  In
elementary school we were made to blot away with cartridge pens with
slotted nibs for a year or two, but since then it's been ball-point and
felt or fiber tips, I'm afraid!

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