RE: Evolution - was VMs: Inks and retouching

["John Grove" <4groves@xxxxxxxxx>]

	GC,

	For a few years at least we've both been arguing on the same side but
from different perspectives and terminology - and in my case with no real
theory; just observation. What I've referred to in the past regarding the
'stroke' order concept is pretty much exactly what you refer to as separate
glyphs as c, cc, ccc, (unfortunately interchangeable with 'e', 'ee', 'eee')
or i, ii, iii. One argument I recall from Gabriel was that if a word
contained
an 'eee' how could we tell if it was supposed to be 'ee' + 'e', or 'e'+'ee',
or
'eee', or 'e' + 'e' + 'e'. Which of course missed the point I had tried to
make
that in my view - both the 'i' and 'c' stroke characters/glyphs require a
closing
ligature to separate them; thus, an eee isn't finished until the next piece
of
the glyph (which is where your glyphs and mine go there separate ways -
sometimes).

	See, what is referred to as 'in' in my view is actually the 'ii' glyph with
a final ligature and relates in design to what is rarely found in the 'ee'
stroke
as 'eb'. The '@' is problematic and isn't really represented well - is a
wierdo -
but of the same construct as many other compound glyphs: Complex because the
seem
to alter the glyph construction process by adding more than normal ligature
'endings'.

	The wierdo's that make of a small set of 'ch' or 'sh' type glyphs without
'end-stroke'
attached to the right of the 'h' to form a shortened 'chy' or 'cho' etc...
These wierdo's
are explained by the stroke-concept better than just being labelled as
wierdos in general.
The '@' is basically the same thing with an 'a' that isn't followed by an
'n' but has the
'n' 'end-stroke' attached directly to the right side of the 'a' which seems
to suggest that
the 'a' itself is an 'e' beginning stroke followed by an 'i' end-stroke.

	Taking that a step-further means that there really isn't an 'i' beginning
stroke 8-) and
all characters (less gallows perhaps) start with an 'e'. This doesn't work
ofcourse because 'l'
is a stand-alone character as well and words like oiin which cause some
heartache to this observation,
so I'm not firm on saying a glyph has to start with an 'e' - yet.

	As I continue to babble onward...

		Okeey then consists of 'o'+'k'+'eey' The 'eey' is made up of the glyph
'eee' and a signifier
to close the glyph. There is no way you could read that as
'o'+'k'+'e'+'e'+'y' this way. My rule of thumb
is therefore... open with an 'e' and wait till you hit a closing ligature of
type 'r/s','n/b','j/d','l/y' or o.

		Now, this means that daiin is in my view only two glyphs 8-) 'd' + 'aiin'
with the 'aiin' being built
by using an opening 'e' with the 'iiii' glyph directly attached to the 'e'
and followed by the closing stroke that
makes it an EVA 'n'. Which means that we would have a lot more glyphs than
we thought because 'a' isn't a glyph
by itself anymore but specifies a secondary set of glyphs to read.

	e + end-strokes = 's,d,y,o,b'
	e + mid-stroke 'i,'h' = 'a,ch'
	complex glyphs: 'sh' = e+h+ (s,b)
	multiples of 'e' + endings create above five finals but preceded by
'e,ee,eee,eeee' ('es','eeo','eeey',etc...)
	multiples of 'e' + mulitples of midstroke i + " (air, aiin, aiiin, etc)

	Zero-beginning set:
	(no 'e' starter')
	mid-stroke ('i') + end-stroke ='r,j,l,n'
	multiples of 'i' + " (so stand-alone 'iin' is different than 'aiin', but
both are single glyphs)

	Weirdo - complexities:

	@ = 'e'+'i'+'n-final' is different from 'an' which is constructed by
'e'+'ii'+'n-final' > again, I like to consider
the '@' and the 'an' and the 'ain' and the 'aiin' and the 'aiiin' as 5
separate characters/glyphs - perhaps related to one another, perhaps not.

	The same for the wierdo 'chy' (not sure of the EVA weirdo ident number),
but it follows the same construction method as the '@' above:
'e'+'h'+'y-final' and differs from the normal 'chy' which is constructed by
'e'+'h'+'e'+'y' (two glyphs in this case 'ch'+'ey'...

	Okay, back to the @ for a second... if 'i-stroke were beginning characters
as well as end-characters':

	'e'+ending 'i' make an 'a' by itself, but 'e'+ ending 'i' +
'i-beginning'+'n-ending' creates a two glyph word with 'a'+'n' and follows
the same construct as the 'chy' wierdo etc...

	Occasionally, there are glyphs with multiple 'h' endings as well =
'e'+'h'+'h' = 'chh'.
	Other wierdo's simply place the 'end-stroke' above another 'end-stroke'
(those 4-o's with a 's-type ligature above the o')

	Well... I guess that's enough babbling.


	John.
-----Original Message-----
From: owner-vms-list@xxxxxxxxxxx [mailto:owner-vms-list@xxxxxxxxxxx]On
Behalf Of GC
Sent: Saturday, July 31, 2004 5:31 AM
To: vms-list@xxxxxxxxxxx
Subject: Evolution - was VMs: Inks and retouching


John,

I hope I'm not testing your patience by carrying this discussion one level
further in clarification, specifically on the 'evolution' of theory.

I think I've made relatively obvious my thought processes on glyph
identification.  The first step was to notice (as so many others before me)
that there was space between symbols.  It's jumping to a conclusion, but the
rational path is to follow the familiar in this instance, based on the
western appearance of the manuscript.  That western-based conclusion would
be that the spaces between one glyph and the next delineate a unit of
writing.  These units compose words, separated by larger spaces, then lines
and paragraphs.  This assumption is validated by the fact that so many
thousands of glyphs fit this observation, and only a relative few do not.
This is the core logic - recording all connected strokes that exist between
the spaces, even Nick's {4o}, as a glyph.  I've said many times that the
first approach is to record what is written as faithfully as possible, then
use statistical and other means to formulate further assumptions.

My last post described the next step in the process, the formulation of
further assumptions, specifically relating to similarity of form and
grouping.  One of the things I left out is the necessity of having the
recorded information at the ready for cross-reference and study.  Properly
formatted information cuts down on processing time, and precise indexing in
several forms - folio, paragraph, line, word, glyph, as well as bifolio and
Currier "Language" - allows for speedy recovery of data and the literally
thousands of cross-checks necessary when formulating the working assumptions
that when tested and proven lead to the creation of hypotheses and theories.
Tedious doldrum, but absolutely necessary.

Once in awhile I pass on observations, such as the one where I said that not
all Currier[A] pages were written in what passes for Currier [A].  I usually
don't offer supporting information, as it has always been my hope that the
reader would investigate these matters for themselves.  But since this
discussion is on the evolution of theory, I'll offer a bit of detail.

A glyph based transcription in VMS studies has a decided advantage over any
other, since its form and function are rooted in the observed writing style
of the manuscript itself, and therefore most closely mirrors what is written
on a page.  Going beyond this is adding noise to the signal, and in certain
cases, the noise can drown out the signal entirely.  I've seen the color
charts that demonstrate certain overlaps in folios.  In some cases my own
data clarifies, and in others it doesn't recognize these overlaps at all.
It may be arrogant of me, but I consider them as photographs taken with a
camera out of focus.  Fuzzy charts with fuzzy edges.  The "cameras" used in
these charts cannot resolve data down to the paragraph level, and can often
misinterpret an entire page.  These kinds of things demonstrate a confidence
relative to the information fed into them.

Here's a snapshot taken from my information, and for purposes of general
discussion I will only focus on a single glyph, and not the other supporting
features of this page.

In the contiguous herbal section, the glyph {c} (EVA <e>) makes up 5.2% of
the glyphs in this section.  EVA probably counts over 800 more occurrences
of {c} than by my methodology.  By my count {c} is present in about 1 in
every 5 words in this section on average, with a heavier bias toward [B]
pages.  My word count for f18r is 77, not counting half-spaces, while the
EVA word count is 83.  Addition of half-spaces would bring our counts equal,
though the half-space is a feature that should be recorded.

EVA counts 4 {c}'s on f18r, but my count is 3.  A quick scan of the page
indicates that the EVA transcription of f18r.11.3 is faulty, and that the
real count should be 3, not 4.  A little math says that there's something
way out of whack here.  If you think that a bit odd, go to the verso, f18v,
and by my count there is no {c} glyph at all on the entire page.  What we do
have are {c} glyphs in the groupings I described earlier.  Providing my own
definitions for these glyphs would be meaningless, but pictorially they
would be -

{ccc} f18v.2.1
{cc} f18v.5.1
{ccc} f18v.6.4

There is more to paragraph and page transition than meets the EVA eye.
Tracing the flow to its source is important, and this cannot be broken down
into generalized groups like Currier's [A]/[B], or even dealt with using
basic Currier transcription or the odd European fad that occasionally pops
out of the hat.  Tracking glyphs like these, seeing where they occur, what
comes before that is statistically different, what comes after, and what the
neighborhood looks like, is very important stuff.  When these occur, what is
present, what is always missing?  This doesn't grab the headlines or
imagination like dark painters and Chinese copyist-retouchers, etc., but
it's the only way to reach the top, one rung at a time.  Each rung gained
gets you a little higher so you look down at a larger part of the puzzle.

GC





----- Original Message -----
From: "GC" <gc-@xxxxxxxxxxxxx>
To: <vms-list@xxxxxxxxxxx>
Sent: Friday, July 30, 2004 1:47 AM
Subject: Re: VMs: Re: Re: Inks and retouching


> John wrote:
>
> > Sometimes your 'colour' is just a little too colourful. 8-)
>
> What the heck, it was only Wednesday, and I'm not due to visit Earth again
> till next week.  So much of conversation is simply too serious, maybe I do
> get carried away - I'll try to reign in my id. :-)
>
> > Anyway, in reference to the oddity of the word ending in 'i' I don't
think
> > anyone decided that and EVA <i> couldn't end a word - it just didn't in
> all
> > but
> > this one circumstance (I think). The fact that you agree that this
> assertion
> > is
> > correct doesn't do much in the way of making a point. The possibility
that
> > the author
> > ran out of room seems reasonable since we don't have much else to go on.
>
> A couple examples, you're right, not much to go on here:
>
> f29r.5.6 - here you also have my alternate {8} and my alternate {a} in the
> same word, unusual.
>
> f55r.10.5 - again we have the alternate {a} in association, and the <i>
may
> or may not be separate from the 'second a'.
>
> We have an example of the double <i> ending a word at f32v.7.7.
>
> > However,
> > I think that 'm' is normally the expected
> > 'oops-I-ran-out-linespace-so-guess-the
> > rest-of-the-word-when-reading' ending to a voynich word.
>
> My reverse sort lists 12 1/2 pages of words ending in {m} for the herbal
> section alone.  (Reminder  - Gotta make one of these for the rest of the
> manuscript).  I haven't gone into my database to tally them up, but I'd
put
> the figure at roughly 12%.  So if the {m} is some sort of abbreviatory
> signal, then what is the {n}?  This would add about another 3% to the
tally.
>
> > >> No one seems to have differentiated here on the variations of this
> > pattern.
> >  If as many of us 'assume', the <in> and <iin> are indeed {n} and {m}
> > respectively,
> > what is the EVA <n> by itself?
> >
> > Don't forget that sometimes the 'n' is part of an '@', just like
sometimes
> > a 'y' or an 'o' is part of a 'ch'...
>
> Okay, I'm a little lost.  I pulled up the dreaded EVA chart to locate an
@,
> but couldn't find it.  I opened the EVAhand1 font file and the @ location
is
> blank.  I did notice that EVA for my @ is <u>.  EVA <u> *looks* like an @,
> so it was a ringer for the the job, even though it occurs in the herbal
> section one time, at the end of f35v.6.3.
>
> > >> These are glyph patterns, one-two-three strokes, and all three can be
> > found in the
> > middle of words, only without the distinguishing flourish that makes an
> <i>
> > an <n>.
> >
> > Interesting, and how exactly can one tell that the 'i' in the middle of
a
> > word
> > is actually an 'n' and not an 'r','l', or even a 'j'?
>
> An EVA <j>?  Curious connection.  I would have gone for the EVA <m> as a
> more likely suspect.  Anyway, this all goes back to the discussion awhile
> ago (maybe more than a year?) of observed sets of glyphs and how the human
> mind abhores chaos.
>
> I believe I made the observation that *most* created alphabets appear to
be
> based on patterns.  Those found in Porta were the ones I used as example,
> where a pattern was chosen and only slightly modified from one glyph to
the
> next for each alphabet.  I don't have a degree in logic or psychology, but
> I've been around long enough to notice that people love patterns, and the
> easier to remember, the better.  They also never venture outside their
> sphere of knowledge, and after a certain age do not attempt to increase
this
> sphere, as a rule.  These observations served as one of my primary keys to
> understanding.  My 'logic' went something like this:
>
> Education at this time was not only heavily mnemonic, it also attempted to
> incorporate all sciences into one by connection.  Medical science and
> herbalism for instance, were inextricably tied to astronomy/astrology.
The
> author lived and breathed unseen patterns and connections.  So why would
he
> then create a script where similar patterns had no connection whatsoever?
>
> With this in mind, I took all the glyphs I'd previously recorded, and
began
> setting them out in groups on a page, based on similarity.  Not
> surprisingly, the bulk of these fit into neat little groups, and those
> groups had the same number of glyphs in each.  Your EVA <n> for example,
has
> the <n>, the <in>, and the <iin> as the three most used.  There is an
> occasional <iiin>, but very weak signal here.  These groupings are
identical
> in number and form for your EVA <r>, <s> and <m>.  My connected {c}
> groupings, your EVA <e>, has identical number and structure.  Again the
same
> for the gallows groupings.
>
> There was that little subset that was troubling, the {c} with a tail, the
> {a} with a tail (EVA <u>, my {@}.  <i> glyphs without finals, etc.  They
all
> have one thing in common however that places them not in a subset, but in
> the mainstream.  When a westerner writes an {m} or an {n} in the middle of
a
> word, it would seem inappropriate to give it a final tail - it just looks
> odd.  We don't finish words with {a} or {c} usually, but the same applies,
> only in reverse.  The locations for the modified {a} and {c} are word
finals
> or stand-alones.  The locations for the "unterminated" {m} and {n} are
> always in the middle of a word, and as you pointed out, there are only a
> couple of examples of the EVA <i> as finals, and in two of the three cases
I
> listed above, there was little room to write the hook.  The third case is
> questionable as the actual glyph itself.
>
> Now we go to the extended groups, and each has one or two examples in the
> "this just don't fit" subset.  Each example follows the rule stated above.
> So the herbal section has 39 examples of my "ccc" glyph (proper EVA would
be
> <EEe>?) in the middle of words, and we have 6 examples of this same glyph
as
> word terminator, each with the final "hook" or flourish.  This also holds
> true for EVA <i>, EVA <Ii> and EVA <IIi> in the middle of words.  Out of
> character maybe, but one thing changes - the above holds true for my "cc"
> glyph, but only for the Currier [A] material.  Later on, possibly for
> writing convenience in heavier text sections, the author dropped the final
> flourish in most instances when this glyph ends a word.
>
> The research is a little more evolved in areas of glyph variation and
> intended meaning, but this sums up the basics.  You can't begin to
decipher
> until you've got a grasp on glyph construction and transcription, so this
> has been a very important and involved part of my study.
>
> I hope this cleared things up a bit.
>
> GC
>
> ______________________________________________________________________
> To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
> unsubscribe vms-list

______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list
---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.721 / Virus Database: 477 - Release Date: 16/07/2004

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.721 / Virus Database: 477 - Release Date: 16/07/2004


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