+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Voynich Talk (https://www.voynich.ninja/forum-6.html)
+--- Thread: Choice of glyphs (/thread-2729.html)
RE: Choice of glyphs - -JKP- - 08-04-2019
(08-04-2019, 02:11 PM)Koen G Wrote: You are not allowed to view links. Register or Login to view.I think it's almost impossible there's a one-to-one correspondence of Voynichese "words" to Roman numerals. Some tokens would be much more common, those standing for common low numbers. So even if it were something like Roman numerals there should be some clustering going on.
You mean like ai aii aiii aiii? or c cc ccc cccc cccc? I'm looking at a line of vms text and these patterns occur in every single token on the first line except for the last token. IF the last token wraps on to the next line (it looks like it might, based on the pattern), then EVERY token has these characters.
RE: Choice of glyphs - Koen G - 08-04-2019
No, what I mean is that then most VM words would represent a number over 10. I randomly opened You are not allowed to view links. Register or Login to view. and there's just some short words like ot and ol. The rest would, given the Roman-like system, easily run in the hundreds or thousands.
That's no problem per se, but it would mean that we're looking at a string of variable large numbers.
RE: Choice of glyphs - bi3mw - 08-04-2019
I experimented with alphanumeric encoding. You can use any text (here Regimen Sanitatis) and turn it into You are not allowed to view links. Register or Login to view. . Statistically, this encrypted text should have similar characteristics as the Voynich manuscript. That have to be checked.
RE: Choice of glyphs - -JKP- - 08-04-2019
I gave a name to the most common biglyphs a few years ago. There is a particularly important subgroup that is frequent and can also stand alone.
I call them Janus Pairs. Some examples include al, a, ol, or, and others. For simplicity, I'll keep gallows chars out of this example and simply represent them as || (one or two pipe symbols, more on gallows chars later).
Pretend a Janus Pair is represented by the glyph * (asterisk). Pretend (also for the sake of simplicity) that the 9 symbol is a null.
EVA-ch is represented as cc (two EVA-e). This is not merely for convenience. If you LOOK at how EVA-ch occurs and note its companions, you will see what it might be reasonable to consider it as cc, at least for the purposes of examining these important patterns. EVA-ch is FREQUENTLY followed by c, cc, etc., in much the same way as the minims follow the "a" shape.
Here are a couple of lines of VMS text:
|*||cc*l cccc ||ccc9 l*c*iii *c* *ccc*iii *cccc* 4*c9 *iii * *iii ccc* *||ccc9g l||9 ccc*r cccg||ccccg r*c* *c* *cc*ii ccc*9 ll* *iii cccg cccccg *ii **r ccccc**ii cccc9 ccccc*iii *occc*o*iii *
Compacting the Janus pairs into asterisks allows one to see the ccc and iii patterns more clearly. Notice also that EVA-r and EVA-l do not always function as part of the pairs (biglyphs), just as the 9 (EVA-y) seems to have its own role both within and without Janus pairs.
What's important to note is that either c-shapes or i-shapes (and occasionally both) occur in ALL the groups interspersed between Janus pairs for these two lines, and for the rest of the paragraph.
RE: Choice of glyphs - -JKP- - 08-04-2019
(08-04-2019, 02:49 PM)Koen G Wrote: You are not allowed to view links. Register or Login to view.No, what I mean is that then most VM words would represent a number over 10. I randomly opened You are not allowed to view links. Register or Login to view. and there's just some short words like ot and ol. The rest would, given the Roman-like system, easily run in the hundreds or thousands.
That's no problem per se, but it would mean that we're looking at a string of variable large numbers.
It doesn't matter how big the numbers are, it matters how they are processed. Maybe they are broken into groups of two. It only takes a couple of dozen to represent a whole alphabet. It only takes THREE glyphs to represent the alphabet in Roman numerals. The rest might be stop-markers, modifiers, or nulls.
The VMS chars can basically be organized into four groups, four basic kinds of patterns.
RE: Choice of glyphs - Mark Knowles - 14-04-2019
Without wishing to be a broken record, in diplomatic cipher keys the character set used is pretty arbitrary, so why can't that be the case with the Voynich?
Often the characters used for letters tend to be simpler broadly speaking that those used for other items like nulls or glossary etc. So there tend to be some patterns in the choice of symbols for the different things, but there is no reason for there to be.
RE: Choice of glyphs - Anton - 14-04-2019
(08-04-2019, 03:08 PM)bi3mw Wrote: You are not allowed to view links. Register or Login to view.I experimented with alphanumeric encoding. You can use any text (here Regimen Sanitatis) and turn it into You are not allowed to view links. Register or Login to view. . Statistically, this encrypted text should have similar characteristics as the Voynich manuscript. That have to be checked.
h0=5.21
h1=4.61
h2=3.29
h1-h2 = 1.32, h0-h1 = 0.6
Not bad!
RE: Choice of glyphs - bi3mw - 15-04-2019
@Anton: Thank you for taking the time to review the features of the encoded text.
I wanted to make a cipher without the need for a codebook that comes as close as possible to the characteristics of the VMS. The problem is that the word lengths do not fit.The result is also a very "mechanical" chart, but not the same picture as with the VMS.
I suppose that sometimes "filler letters" have to be used to make the result look more VMS-like.
RE: Choice of glyphs - RenegadeHealer - 17-09-2019
(07-04-2019, 01:33 PM)Koen G Wrote: You are not allowed to view links. Register or Login to view.What does this tell us? It appears that the glyph set is mostly composed of numbers and scribal conventions. Those few that are found in the alphabet often double as a number or abbreviation. So the glyph set simultaneously:
- Avoids unambiguous characters from the alphabet.
- Uses characters familiar to the Latin scribe.
The final point might also be of interest. The selection of known symbols likely meant that the writing system could be learned easily.
But why construct a glyph set that relies so heavily on numbers and abbreviation symbols?
Ease of learning and familiarity definitely seem like reasonable motivating factors for the design of a new character set, for any purpose. This was even more true in the days when at least one person was going to need to write a large amount of it by hand even once. Hangul, again, is a great example for comparison: designed entirely from scratch by a single individual, very logical and consistent, and clearly designed for, and by, people who learned all they knew about writing from the Chinese. (But decidedly not Chinese characters, despite the clear stylistic influence.)
It's a temptingly fun thought experiment to consider why someone would design a character set that maximized, rather than minimized, ambiguity. To blend in unnoticed to the majority of people who give it no more than a cursory look — to hide in plain sight — is the answer that comes most readily to me. I can think of several examples of communication that is designed to purposely fall into the uncanny valley of things we're not sure if we should have understood, from any of the 3 major theories regarding the VMS:
- Meaningless filler designed to give the impression of being language-ish: I'm sure many non-native English speakers here recall hearing the song You are not allowed to view links. Register or Login to view. before they learned English, and later being surprised to learn it's not real English. As a native English speaker, though, it's a creepy song; I feel as though I should understand it, and if I listen hard enough, I will. But no. Adriano Celentano mimicked the phonology and prosody of English quite well, even though he didn't speak it at all.
- Cipher: Irish Travellers and carnival folk use a lot of short cryptic phrases to communicate things that are for insiders' ears only. Most are so briefly muttered and phonologically similar to English words, that listeners mistakenly think they merely misheard something unimportant. The term kayfabe, associated with the fictitious backstories and rivalries of professional wrestling, started as a signal for one carny to subtly remind another to stick to the story they'd agreed upon.
- Natural language: The Cherokee alphabet was designed by Sequoiah, at a time when he could not speak, let alone read, any language written with the Roman alphabet. But wanted his people's language, when written, to blend in with the examples of English writing he first saw.
I'm not cheering for the number hypothesis, as interesting and defensible as it may be, because it's as much if not more of a dead end than the meaningless-filler-hoax hypothesis. Even if the numbers could be parsed without ambiguity, unless we find something that strongly hints at a key, what could these numbers possibly refer to? Playing fill-in-the-blanks with a half-finished crib and Google Translate at least produces some cheap laughs at the sheer ridiculousness of the output. (A pox upon all of our livers, indeed!) So we decide the VMS is nothing but numbers. When this forum dwindles to a bunch of crackpots listing a verse of the King James Bible or a line of Dante's Inferno for every vord, I'm not sticking around.
RE: Choice of glyphs - Koen G - 17-09-2019
(17-09-2019, 05:08 AM)RenegadeHealer Wrote: You are not allowed to view links. Register or Login to view.I'm sure many non-native English speakers here recall hearing the song You are not allowed to view links. Register or Login to view. before they learned English, and later being surprised to learn it's not real English. As a native English speaker, though, it's a creepy song; I feel as though I should understand it, and if I listen hard enough, I will. But no. Adriano Celentano mimicked the phonology and prosody of English quite well, even though he didn't speak it at all.
Heh, I only heard this song as a kid and I had no idea it was nonsense
The "all right" really sells it.The "looking like language" strategy would only fool the illiterate though. Wouldn't someone used to reading manuscripts raise an eyebrow at this strange glyph set? In their eyes it would be comprised of numbers, ligatures and abbreviations, which might be more jarring than we suspect.