Re: Evolution - was VMs: Inks and retouching

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

On Sat, 31 Jul 2004, GC wrote:
> John, we're in agreement on the <in/ii>.  Can you give me a
> folio location for your '@'?  Is this EVA?  I still can't find it.

Might I also ask, as a latecomer, what the b is, structurally?

The formation of the @, if not its location, seems clear, and the logic
of the analysis attracts me strangely, but I missed the formation of b.

I have a terminological suggestion, if I might venture it.  The basic
constituent elements that John Grove recognizes might be called graphs, or
elements or sub-glyphs.  These are in some cases, of course, characters
under the line of analysis that currently culminates in EVA, and in other
cases sub-characters.  Any larger elements made up of one or more graphs
are glyphs.

In essence Grove proposes that EVA approach misanalyzes the glyph level of
the Voynichese script, but not looking hard enough at the way in which the
graphs are patterned.  It stops short, fooled by the lack of connections
in some cases.  To support his view he offers some glyphs that employ the
standard graphs, but do not appear in recognized glyph lists.  He also
adduces the weirdos, words exemplifying anomalies in the presently
recognized syntax of glyphs, because a misanalysis of the graph-structure
of glyphs can lead to a single glyph being analyzed as as several glyphs
in sequence.

Of course, even EVA and its predecessors recognize that there are
characters made up of other characters, e.g., the i(0-3)R (R = lrnm) set,
or the gallows sets (when embenched).  And it appears that some earlier
analyses have seen the iR set as i(1-4)R', with R' being the "twiddles" or
endings, as Grove calls them.

As far as that goes, here's a weird item from f100r, that I noticed on one
of the possible mandrake pages suggested by Dana Scott.  The file
reference is f100r.p1. The oddity is in the fifth word.  This is variously
transcribed as cthdaoto, ???? ?oto, tedaoto, and cthedatoto, but, as the
file comments "The two <t>s in <cthedat> below are actually a single <t>
with one leg in each spot."  What I would say is that we have cth, i.e., a
t with a ch circumfix, followed by oto, but that the ch actually embenches
only the first leg of the t and that the elements da follow the h within
the t.  If you think of a t as being two glyphs - call them t1 and t2, a
cth is c t1 t2 h, and what we have in this word is c t1 h d a t2 o t1 t2
o.

This is depicted at
http://beinecke.library.yale.edu/dlxc/brbldl/oneITEM.asp?pid=2002046&iid=1006248,
for the curious who skipped the Mandrake discussion.

The reason I'm interested in Grove's analysis is that it tends to fit in
with my abugida suggestion.  Perhaps not directly, but the foundation of
that query was my noticing the analogy between the internal structure of
the iR glyphs or graph sequences and Tolkien's Elvish script, which
happens to be an abugida.  What puzzled me about EVA Voynichese, however,
was that only some of the glyphs in EVA behave in this systematic way.
Then I noticed Stolfi's word syntax which shows some similar "cross-glyph"
behavior, in EVA terms.  I postulated that the iR forms were bundles of
sub-character marks representing some of the actual character + vowel sign
pairs of an abugida or abjad, and that some additional bundles of
sub-character marks were not recognized as bundles in the EVA, but are
shown by Stolfi's analyses to behave as such.  (Stolfi's Note 17 goes even
further in this direction than his Word Syntax, in some ways, though I'm
still digesting both and I don't know their relative chronology.)

What Grove's analysis suggests is that the fine structure of the text is
at the graph level, and that premature systematization of the glyph level
is to some extent leading us down the garden path.  This is in line with
my postulated bundles of sub-character marks, but he offeres concrete and
(to me) reasonable suggestions about how to make the analysis into marks.
I haven't yet pursued this to the point of wondering what the bundles set
actually is, but if we substitute graph sequences into Stolfi's syntactic
analyses we will know.

Moving along, I was going to ask, did anyone think that the j and g
characters (which I gather are rare) looked like i + X and e + X, where X
was a twiddle that looked perhaps most like the twiddle added to an i to
make an m?  That would, of course, make j just a somewhat constipated m.
(And g would be a variant of d, as the logic unfolds below.) Along these
lines, would p and f and their embenchments t and k be something (q?) plus
this same twiddle?  In that case, and Grove says as much, s is e plus this
same twiddle.

Following the EVA itself for i and Grove's suggestions for e, then, I see
four twiddles or endings or diacritics (or banners?) at least, as Grove
does, to wit:

- Up hook attached to top of base (r-ending), so that EVA r and s seem to
be the top uphooks of base i and e.

- Up hook attached to the bottom of the base (n-ending), so that EVA n and
o are the bottom uphooks of i and e.

- Downhook attached to the top of a base (l-ending), so that l and y are
the top downhooks of i and e.

- Loop over attached to the top of a base (m-ending), so that m and d are
the loopovers of i and e.

One possibility would be that these endings are vowels, though in that
case I would expect a fifth one.  Hence my interest in b, which Grove
identifies as the fifth element in the sdyob set of e + ending
combinations.  He didn't mention a fifth element in the rmln sequence of i
+ ending forms.  The only candidate I see is the cross-bar linking ch and
sh.

Under this logic one would have to wonder, with Stolfi, whether ptfk were
in contrast.  And why is the loopover or m-twiddle the only one attached
to q (and possibly a high i, in the case of f and k).

Well, enough for now.
______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list