Re: VMs: 1st Impressions and baby steps

[Barbara Barrett <barbarabarrett@xxxxxxxxxxxxxxx>]

> >Barbara Blithered;
> >Do I have anything beyond the "gut feeling" that many report when they
> >see the VMS that they ought to just be able to pick it up and reed it?

> Jacques Jzotted; 
> Jim can (not Jim Gillogly, the other one. Sorry, couldn't resist :-).
 
> >Yet obviously the authour wanted its contents to remain
> >obscure except to the initiated.
 
> This is quite common. When I was a not-yet-teenager I kept a "journal"
> (actually a notebook of my "inventions") in various "codes". One I
> remember was just plain French written phonetically in Cyrillic.
> Another again was plain French, but this time written in Latin
> cursive (lifted from one of Lovecraft's novels) with mock
> medieval abbreviations (e.g. a stroke over a vowel to indicate
> a nasal).

Barbara Babbles;
I've come HOME! <g> I did similar in school, although it was to pass
notes with a welsh friend, and latter living in Northern Ireland where
personal searches were then common (you needed a body and handbag search
just to enter the city centre in Belfast) I was rather miffed (and
somewhat embarrassed) every time my diary was taken out of my handbag
and read by the searchers, so in both cases writing systems were
invented by me, with their own orthographic rules and spellings to
baffle the uninitiated and preserve my privacy. Don't know about
"unbreakable" but certainly from an epigraphic viewpoint they'd be
demonically, if not devilishly, difficult to "crack".

ASIDE: If anyone wants a go at my diary "code" I can send a few pages to
them in that script for them to try to break it.

 
> >So in therefore the only "safe" senario I can envision is if the "key"
> >to the VMS is simple enough to be kept in one's head, simple enough that
> >the VMS may be "read off" in a single pass, and the nessesity for
> >written of physical keys, and the need for worknotes etc, is eliminated.
> 
> Sounds sensible.

See my slight revision of that in response to Jeff's post.

> 
> I'll just repeat here my "improved Chinese theory". Even though
> I have posted it here fairly recently.

And I posted my reservations of that idea based on my knowledge of
european beliefs about chinese common until the last century: see
"Chinese" as posted . However my reservations do not make the chinese
idea impossible, just very unlikely.

However, there is a test for the idea's validity. Tone marks. Some
"letters" could do double duty as tone marks. They could be linear
diacritics as in the Yale transcription system. A tone mark would have
to be present for every syllable (-1 of course). Say cantonese was the
language, with it's seven tones, there'd have to be 6 "letters" which
showed a higher occurrence than any others as almost every single
syllable would need one of these letters to provide tone. In Mandarin
with 5 tones only 4 letters would be needed, and the incidence of their
occurrence in the text would be very high indeed.


> 
> I happen to be very probably the only non-native speaker of
> a language called Sakao. When I learnt it, 30 years ago,
> it was spoken by some 1000 natives on the island of Espiritu
> Santo (now Vanuatu, then New Hebrides). The language has two
> dialects, BTW. It is fairly complicated compared to other
> Austronesian languages (internal inflections for instance).
> If I wanted to keep notes which no-one but myself could read,
> I would merely make up an idiosyncratic alphabet, and write
> them in Sakao in that alphabet. To crack my code:
> 
> 1. You need to crack the alphabet. I'd say you can do that
>    only if you have guessed at the language.

Very untrue. In every decipherment involving an unknown script and an
unknowen language the
writing system was always "cracked" first.
Robinson's "Lost languages" (ISBN 0-07-135743-2) gives good examples,
but the pre-mayan decipherment book "The Story of Decipherment" by
Maurice Pope (alas out of print now and pre-ISBN) is better as it
contains several essays on decipherment which are still valid today.

> 2. To guess at the language, you need to know of the existence
>    of Sakao (and its two dialects), AND you need to have
>    sample texts.

Well, as Robinson and Pope (far more eloquently) point out the writing
system comes first, your text would be "deciphered" in the sense that
Etruscan is "deciphered" IE it can be read aloud but what the language
means is still mostly an unknown - however context gives some insights
(in the case of Etruscan the occasional short bilingual has given
meaning to a few words too) but the language and its rules remain in a
very large part unknown. 

Point one, being invalid, invalidates point two.

> 3. I won't elaborate further. I'll just say: this code would
>    be uncrackable, short of an extraordinary stroke of luck.

I believe it would too, but not for the reasons 1 and 2 that you give.
If your writing system was truly idiosyncratic, or even eccentric, it'd
be very difficult to assign values without the "crib" of a bi-script (in
one language) or a bilingual (in two scripts) where one was a known.
As you'd be very unlikely to leave such lying around after the system
had
been worked out and memorized it would be very very difficult to crack
the writing system. But, unless you encipher language in a manner that
is wholly original, never been done before and thus unlikely to be
guessed by epigraphers, then it may be possible to assign consonant and
vowel values in the abstract and create a "pattern" - after that it
would be a case of searching for a language with the same phonemic
pattern. In your case the language is so obscure (Vanuatu has 105 local
languages and the linga fanca, Bislama, is related to none of them,
being a creole of 19thC english) the correct pattern may never be found
unless you were identified as the author and you linguistic skills had
been recorded.

C and V patterns could be assigned in the abstract, but the way to beat
that is to be inventive and use something no epigrapher would ever have
come across. For Example; a syllabary with units of the form VC with
unique orthographic rules for C-start, V-end words and elimination of
vowels for the occasional consonant cluster would baffle epigraphers
because every syllabary and syllabet in existence is of the CV form so
they'd waste time attempting to assign CV abstract values to a VC
system!

> Flash back. I was born in the 16th century, and travelled to
> China. Or perhaps I was fascinated with languages and met
> a Hakka-speaking trader or diplomat in Venice, and set about
> learning Hakka (or Holo, or any of the hundreds of possible
> candidates). So I am a 16th-century Venitian, the only non-
> native Hakka speaker in Europe. It is a pretty sure bet that
> no-one will ever decipher my notes if I write them in Hakka in
> my own made-up alphabet. Even a Hakka speaker. Would he
> imagine that these Latin-looking pages are written in his
> own language?

I don't think this is very likely for many historical reasons, not least
the celebrity of anyone who'd been to the east (native or trader) or
even claimed knowledge of china. The number of charlatans around
claiming such knowlege and adventure were numerous - just one person
with genuine knowlege would have burst their bubble.

But, back to the notion of the VMS is chinese; a "letter as tone" count
could test the idea of a chinese language;
tones would have to be marked in some way because they are so
semantically important they could not be left out - one could have
different vowel glyphs for each tone but the VMS hasn't enough "letters"
for this method, so there would have to be a "tone mark" of some form.
Given that the diacritics in the VMS do not appear frequently enough for
them to serve that purpose then all that's left is a letter doing double
duty as a tone mark.
 
> In this hypothesis, deciphering the VMs amounts to deciphering
> a text written in an unknown language and an unknown alphabet.

This is true of all hypotheses because the script and language are
unknowns!
The VMS doesn't have to be written in an obscure language for it's
language to be an unknown!

> This is was interested me in the VMs. Not the contents,
> not who wrote it, not when, not where, but how to solve the
> problem.

Personally, as a student of writing systems my greatest interest is
discovering how the writing system works, but when, who, and contents,
could provide very important clues - who and when providing the mindset
and probable strategies they could come up with, and the contents (if
related to the illustration) may provide "cribs" if the language
structure is revealed by abstract C & V assignment.

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