VMs: Re: VMs word models --> state machines...?

["Michael Winkelmann" <DasWinkelmann@xxxxxx>]

Hallo to Nick and hallo to everyone,

> I'm planning to extend my JavaScript VMs code to handle various
well-known
> VMs "word" models, to measure how well they actually work relative to
each
> other - as my code already uses a state machine, it should just be a
> question of converting each paradigm to its implicit state machine,
letting
> it run, & gathering statistics.
> (...)

this is such an easy and elegant idea. It should be my idea, but it
isn't...

> For example, I just tried to do this for Michael Winkelmann's recent
model,
> but my poor German comprehension has left me a little bemused - can
someone
> get me properly started?
> http://elias.weltretter.de/index.php?txt_id=45

Sorry again for my german text. I wanted to translate my text in
english, but the last week I had the influenza and couldn't do it. And,
of course, before posting it to the list, I planned to do some own
analysis and get "hard data". There _are_ exceptions from my very simple
rules.

But it is time to write the most important parts in an (hopefully)
easy-to-decipher english.

My text was originally written as a german e-mail to Jonathan. While
thinking about patterns of perception I remembered some of my old
findings and took a look in my diary. My diary looks a little bit
similar to the VMs on some pages, which are full of strange diagrams,
voynich glyphs and highly abbrieviated text written by a fluent hand,
but heavy to read even by its writer. ;-)

In the last summer, I noticed on a hike thinking about the VMs, that the
glyph sequence "qoteody" would form a possible word, but the sequence
"qoteidy" would surprise me. And indeed, "qoteody" appears 12 times,
"qoteidy" never. In transcription, these words look very similar. What
is the difference between the words, and which rules of word composition
I noticed intuitively?

I found the rules not by looking at transcriptions, but by looking at
the images of the manuscript. (Mostly Herbal A folios, some herbal B
folios.)

It was my strange observation that there are rules for "harmonic" glyph
sequences, which are followed by most of the "words" (but not by all
"words").

These are the rules which make the script look so elegant, which is not
caused by the shape of the glyphs only. If you format a text in a
"normal" languge with the EVA Hand font, the result is far away from
being elegant, it looks rather ugly.

It is strange, because the writer does not care for the "layout" of the
text. There are no predrawn lines (except for the circular diagrams) and
the "text" appearance is a little irregular on most pages. The usage of
a writing system with "glyph harmony" rules generating aesthetically
appealing "words" seems to be contradictionally in this context.

By explaining this to Jonathan, I was "forced" to write it more
systematically than in my own notes. After I'd done it, there was a
simple scheme. So I put the (fast written and incomplete) text on my web
site and announced it to the list member who can understand german.

(For all following text examples I use the transcription of T. Takahashi
in the version extracted from J. Stolfis interlinear archive of Voynich
manuscript transcriptions.)


THE GLYPH CLASSES:

> I-class i n r l m

I-class: All glyphs starting with the straight i-stroke and optionally
with a "flourish".

> E-class e o s y g

E-class: All glyphs starting with the rounded e-stroke and optionally
with a "flourish". Of course, the EVA d is in the E-class - it was my
mistake.

> K-class (gallows) k t f p

Gallows: We all know those strange glyphs.

> L-class ih ch cfh [+ sh ?]

ch and sh starts with the e-stroke and are members of the E-class. (To
me they look like a special kind of ligature. I name them ligatures, but
of course they can be glyphs of it own. So I prefer to take some caution
and describe the ch, sh explicitly.) ih is a kind of "control sequence"
which allows the change from I-class glyphs to E-class glyphs.

> X-class (exceptions) a q

Exception glyphs:

"a" is a "control glyph" which allows the change from E-class glyphs to
I-class glyphs. In its visual appearance it combines the round
"e"-stroke with the straight "i"-stroke.

"q" is nearly always at a "words" start. (But there are rare exceptions,
as in f93v.P.6 "sheqokam".)


THE HARMONIC RULES:

> Law#1: E follows (E / ch / K / a)

#1: After an E-class glyph follows a E-class glyph, ch, gallow or a.

> Law#2: I follows (I / L)

#2: After I-class follows I-class or ih.

> Law#3: I (except i) usually terminates a word

#3: A glyph of the I-class which is not "i" or "l" usually terminates a
"word".

Example: "daiin" is "harmonic", but "daiiny" is not (and looks a little
weird to everyone who spent his sleepless nights on the VMs). Some
transcribers did read "daiiny" in f47r.P.11, last "word". I would
suspect that there is a space before the "y"-glyph, which is smaller
than normal to avoid overlapping the glyph with the "plant".

> Law#4: E (except e  [or "o" too?] ) usually terminates a word

#4: A word does not end in EVA "e" or EVA "i". It is nearly always an
"i"-stroke or "e"-stroke with a flourish (as n, r, l for I-class or s,
d, y, o for E-class).

Example: The word "che" (f105v.P.10, third word) is rare (this is its
only occurence), but the word "ches" is rather frequent, even "chs".

> Law#5: [Not sure about this one - any suggestions?]

#5: The "l"-glyph is a kind of "joker". It is by its appearance (starts
with the i-stroke) a member of the I-class, but can be used as E-class
glyph.

Example: The word "olol" occurs 18 times in the T. Takahashi
transcription. There are many words with a sequence of E-class glyphs
interrupted by "l"-glyphs, and some are very frequent ones.

> Law#6: [Not sure about this one either - any suggestions?]

#6: If a glyph of the I-class is followed by a ih-ligature (with an
optionally embedded gallow), the word will be continued with glyphs from
the E-class.

> Law#7: K-class is nearly always surrounded by E-class, or as c[K]h

> The absence of "d" is puzzling: should the final E-class "g" actually
be
> "d"? This could just be a typo in the table.

The reason for the "d"-absense is that I typed my text faster than I
thought. There is no thing so easy that I can't do it wrong...

> I'm far from sure about "ih"


STRANGE "IH":

The ih-ligatures are strange. It took me a while to notice them, they
look very similar to the ch-ligatures. The first occurence on f1v (title
of the first paragraph) is hard to identify in the manuscript, and it
does contradict my simple rules in the T. Takahashi transcription. But
if there were a space before "a", it would fit. (The whole page f1v is a
transcribers nightmare.)

The second ih-occurence on f2v (first word on the bottom paragraph)
isn't easy to recognise on the image I used. As transcribed, it does
contradict my rule too.

The other ih-ligatures in Herbal A with embedded gallows are:

f24r.P.2: okaiikhal  - i-ih-a, okay
f24r.P.19: otaiphy - a-ih-y, okay
f30v.P.3: chtoithy - very strange

This looks _weird_ in the manuscript. There seems to be a horizontal
line from the "o" to the "h", the "i"-stroke is embedded. "o" and "h"
looks less faint than the other components of that weirdo, so I would
suspect it as a result of later retouching. On my image, the first word
of the pargraph looks "very retouched" too - if somebody has access to
high resolution images, please check it.

f45r.P.1: shaikhy a-ih-y, okay
f51r.P.14: daiiithy - i-ih-y, okay
f90v2.P.1: cphdaithy - a-ih-y, okay
f90v2.P.2: ikheeos, ih at beginning, okay
f90v1.P.7: etodaithey, a-ih-e, okay
f93v.P.6: qochoithy - that's the reality...


In the last word, if the "o" were an "a", everything would be okay. But
it seems really to be an "o".

> as well, & the whole issue of c[K]h seems not to fully match how they
> appear in the text IIRC... etc etc.


EXCEPTIONS FROM THE HARMONIC RULES:

There _are_ exceptions from the rules, but approx. 90% of the words in
the manuscript are "harmonic". For the other 10%, most exceptions occur
for the last glyph. Approx. 3.5% of the words shows exceptions in other
positions, the most frequent exception is a "oi" seqence (as in
"choiin", "koiin" etc.)

This could be an eighth rule, the oi-exception. But in my mail to
Jonathan I wanted to make clear the "I-class follows I-class and E-class
follows E-class and a class change is done with 'a' or 'ih'-glyphs"
paradigm, which is valid for most of the words.


WHAT DOES IT MEAN?

The rules, which creates an elegant script, could not be created by
accident. They are clearly a property of the encryption or writing
system.

I don't know much about other languages than german. It the text is not
encrypted, it is possible that there exists a language with equivalent
harmonic rules in the phonetic structure, and such a language should be
easy to identify. By the appearance of the illustrations, the VMs was
written somewhere in europe, and so the language is most probably an
europeaen one.

But after decades of research, no one is able to identify the Voynichese
language.

If the text is encrypted, the used encryption scheme must create
sequences of glyphs with these properties. But we all know, that it is
not a simple character substitution cipher...

And all the "nothing than gibberish" theories must explain the harmonic
rules. Gordon Rugg does not, except he creates his underlying tables
with greatest care. But this is far away from a "fast created fake" for
getting money for a bogus source of secret knowledge.


IS IT A TEXT?

Everyone who want to solve the VMs do believe, that there is a
"meaningful text" which has to be read or deciphered.

But in the case of a "text" in the ordinary meaning of this word, which
is a sequences of phonetically notated words: Can there be a kind of
notation or encryption, which creates a optical harmonic script?

This question does not mean, that the VMs is "meaningless" or
"gibberish". It could contain a meaningful message which is not a text
in the ordinary sense.

This is my fast hypothesis of a useful notation of a meaningful content:
It could be a musical notation system for a kind of music composed after
rigid rules. Such a message is totally meaningful, and the writing of
the message is perfectly useful for musicans. But it were impossible to
"read" the "text" in this case, we should better try to "sing" it...


THE MUSICAL NOTATION HYPOTHESIS:

Some properties are easy to understand, if the VMs contains a kind of
musical notation:

- The structures in a line, which shows that every line is a "unit of
meaning" could explain the line as a musical unit of meaning.

- The frequent repetitions in the text and the repetition of "words"
with minimal changes are ordinary musical style. A short theme is
repeated some times, perhaps in different transpositions or with a
little change. You'll find it in children's songs and in symphonies.

- The special statistical properties of the first line of a paragraph or
side could represent some peculiarities at the begin of a musical
composition or a part of it.

- The suffixes of the "words" could represent musical or rhythmical
phrases or typical rhythmical patterns.

- The remarkably high order of the whole text were a normal feature of
every kind of "well-sounding" music. Even, there were always some amount
of surprise, without it, the music tends to be "boring".

- The two Currier languages could represent the musical scales "major"
and "minor".

- A notation system, which creates "nice looking" glyph sequences for
"well sounding" elements of musical style would be completely
reasonable.

- The frustrating failure of all attempts to read an assumed text of the
manuscript would be explained. It were caused by a wrong assumption on
the nature of the text.


PROBLEMS:

Of course there are problems with the "fast hypothesis"...

- Why should be there no other manuscripts with the same or a similar
kind of musical notation?

- Illustrations and labels are completely unexplained.


FINALLY:

After decades of research, with the mental power of hundreds of
enthusiastic people and with the cheap power of computation available
nowadays the VMs is still "unread". For most of us and for me it is
clear that the VMs is far away from being "nothing than gibberish". It
seems, that it is time to check the assumptions on which the research
was done. We all want to hear a mediaval voice from a sequence of
puzzling glyphs, the voice of the author, who wrote and drawn this
unique and enigmatic work. And why should this voice not sing?


Michael

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