The Voynich Ninja

(10-09-2024, 04:24 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.Some of the problems you point out could be solved by glyph variants. For example if only the minim matters, not what type of extra part is attached to it. (I don't know if this particular change would solve anything, it's just a made up example). This would turn the burden of multiple codes into a matter of esthetic scribal preference.

Sorry, I'm not sure I fully understand what you're saying. Are you suggesting that only the presence of a minim is relevant, and that we should assume [daiin] has the same meaning as [dain] or [daiiin]? Or are you referring to something else?

The order of glyphs in Voynich words is very predictable. Most words start with either a group like [d-], [ch-], [ok-], or [qok-] and end with a group like [-aiin], [-ol], or [-dy]. The following tables gives an example of how the most common prefixes (d-, ch-, ok-, and qok-) could be combined with the common suffixes -aiin, -ol, and -dy, respectively. Note that this table, e.g., reproduces the most frequent tokens [daiin], [ol], and [chedy]. 

The

Quote:            -aiin      -ol       -dy
none      aiin       ol        dy
 d-        daiin     dol       dy
 ch-     chaiin   chol   chedy
 ok-     okaiin   okol   okedy
qok-  qokaiin  qokol qokedy

Note: It is also possible to replace the prefixes by similar shaped glyphs; e.g. [s-] instead of [d-], [sh-] instead of [ch-], and [ot-] instead of [ok-]. Similarly it is possible to replace the suffixes by similar shaped glyphs; e.g. [-ain] instead of [-aiin].

If you remove some glyph variations or even parts of the words, the text inevitably becomes more repetitive. However, this doesn't solve the issue of problematic word distribution. For example, removing all prefixes would make already repetitive sequences even more monotonous. Take the sequence "shol chol shoky okol sho chol shol chal shol chol chol shol ctaiin shos odan" from f42.P3.20—after removing the prefixes, it would reduce to "ol ol oky ol o ol ol al ol ol ol ol aiin os an," increasing the repetition without addressing the word distribution problems.

(09-09-2024, 09:55 AM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.One can formally describe rewrite rules to take Voynichese closer to a hypothetical "underlying structure" which the writing system doesn't always represent directly. E.g. Emma did this with her hypothesis of You are not allowed to view links. Register or Login to view..
Emma's post tries to make sense of [o] and [a/y] as corresponding to two entities with very similar roles in word-structure. Though there may be other explanations, an obvious parallel is phonetics and how vowels are largely interchangeable.

Another way of analyzing this would be to observe that [o], [a], and [y] all begin as [e], but are then each closed off by a different concluding stroke.  That is, if we compare word pairs such as [daiin] and [dy], or [qokaiin] and [qoky], which Emma found appearing in remarkably consistent ratios, it may not be that [a] as a whole is alternating with [y] as a whole, but that the "roots" of these words are actually [de] and [qoke].  The word [daiin] could then be decomposed into [de], plus [\\\\] (four minims), plus the upward curve that forms an [n] with the final minim.  Or, if [n] is just a terminal variant of a minim, we could further reduce the structure to [de] + [\\\\].  But either way, the heuristic advantage of treating [a] and [y] as ligatures in this context would be that the structure Emma describes would arguably be represented directly in writing.

Indeed, the shape of a glyph must be compatible with the shape of the previous one, and is also influenced by its position within a word or a line. Currier wrote: "There seem to be very strong constraints in combinations of symbols; only a very limited number of letters occur with each other in certain positions of a 'word'" (see You are not allowed to view links. Register or Login to view.). Following Currier, the glyphs used to write the VMS are mostly based on a curve or on a line as first quill stroke: "We have the fact that you can make up almost any of the other letters out of these two symbols [i] and [e]." (Currier 1976). The glyphs [e], [y], [g], and [s] are based on a curve-stroke, and the glyphs [i], [l], [m], and [r] are their counterparts based on an a line-stroke. The glyph [a] is a combination of a curve and a line stroke ([a] = [e] +[i]). 

Furthermore, Schwerdtfeger has described in 2008 the following four design rules for the VMS (see You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view.): 
(1) line-glyphs can follow line-glyphs or [a]; 
(2) curve-glyphs and [a] can follow curve-glyphs; 
(3) the [l]-glyph can be used as a curve-glyph or as a line-glyph; and 
(4) gallows glyphs count as curve glyphs.

An example of this behaviour is the sequence [che]. The ligature [ch] consists of two [e]-glyphs connected by a dash. [ch] is obviously similar to [sh]. The common Voynich word [chol] therefore can be seen as the combination of a ligature <ch> and a ligature <ol>. Consequently, obvious transformations would be [ch] into [sh], [ol] into [al], [or], and [ar]. Therefore, [chol] could be the origin of words like [shol], [shal], [shor], [shar], [chor], [char], and [chal]. At the same time [ch] is a sequence of two [e]-glyphs. And since [e] can appear repeatedly, it is also possible to add an additional [e]-glyph after [ch], leading to words like [cheol] and [sheol] (see You are not allowed to view links. Register or Login to view.).

(14-09-2024, 04:27 PM)Torsten Wrote: You are not allowed to view links. Register or Login to view.Furthermore, Schwerdtfeger has described in 2008 the following four design rules for the VMS (see You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view.): 
(1) line-glyphs can follow line-glyphs or [a]; 
(2) curve-glyphs and [a] can follow curve-glyphs; 
(3) the [l]-glyph can be used as a curve-glyph or as a line-glyph; and 
(4) gallows glyphs count as curve glyphs.

I like this, it's a very brief and elegant way of describing the system. I find the case of a especially intriguing. Is it like the glue between curve glyphs and line glyphs? Is it simply the shape the first minim takes? (In that case, there would always be at least two minims/line-glyphs in a sequence). Is it a e - i ligature, as Patrick mentions? That would imply that there is a free-floating curve available whenever lines will follow.

I’m finding this a very informative thread. I’ve long thought a positional variant might be possible. It just seems so complicated. But maybe it really is that complicated, though nothing else was being used similarly in the time period. 

But then again, it might still be simple but with an extra step involved. I’m a bit obsessed with spaces because of some of the imagery. I’m wondering if anyone has derived a different “alphabet” for the Voynich based on the last character of a word and the succeeding (after the space) first character of each word combined into a binary glyph.   For “fun” some time ago, I went through a dense page that should have at least the majority of the alphabet included. I recorded the combos and found the number of combos did realistically reflect a possible alphabet of about 28 if I recollect correctly. The most frequent binary glyph was Eva -yo. The most frequent letter in Latin is “i”. The only binary glyph on the page repeatedin sequence three times was “yo”. And that made sense because I repeated 3x is iii. So Roman numerals. And thus perhaps a substitution cipher after all that would still account for Torsten’s findings, the rigidity of the text etc. 

I was sure, however, that someone had already thought of doing this but I’ve never seen it so thought I would ask. At any rate, perhaps something like this could be involved? It would mean the actual meaningful  content of the text would be greatly reduced.

I also would love to see some research on that, although to be fair, even with such kind of substitution (e.g. accounting for spaces as possible characters in n-grams), I'm betting the text won't exactly match a natural, romance/germanic language plaintext due to

• LAAFU/PAAFU
  
• reduplication of (long) words
  
• reversibility of vords (there is some evidence that suggests Voynich word order is far more flexible than any natural languages. Talking I'm about this kind flexibility of.)
  

But who knows really. It could very well be a mixture of many ciphers, including substitution (assuming of course, the text is encoded plaintext and not gibberish)

(26-09-2024, 12:48 PM)RadioFM Wrote: You are not allowed to view links. Register or Login to view.But who knows really. It could very well be a mixture of many ciphers, including substitution (assuming of course, the text is encoded plaintext and not gibberish)

While it may well be so that VMS is really enciphered real text, I feel we must keep in mind that the system to produce the VMS must have been also easy and fluid enough for them to produce anything with any reasonable effort. With the understanding and tech of in the 14th-15th centuries. 

Sometimes I feel many of the proposed systems do not take in account the persons doing the job. nor the general situation of technologies of that period. With some   exceptions of course, like Torsten's autocopy of wheel based systems, grilles to some extent, which would have been doable in the period and fluid enough to be able to produce something with reasonable effort and knowledge. Complexity is not human.

(26-09-2024, 12:48 PM)RadioFM Wrote: You are not allowed to view links. Register or Login to view.I also would love to see some research on that, although to be fair, even with such kind of substitution (e.g. accounting for spaces as possible characters in n-grams), I'm betting the text won't exactly match a natural, romance/germanic language plaintext due to

• LAAFU/PAAFU
  
• reduplication of (long) words
  
• reversibility of vords (there is some evidence that suggests Voynich word order is far more flexible than any natural languages. Talking I'm about this kind flexibility of.)
  

But who knows really. It could very well be a mixture of many ciphers, including substitution (assuming of course, the text is encoded plaintext and not gibberish)

True. I think your examples only support my speculation though. Word length, duplicates, reversibility - none of that would matter. In my example, only the first and last letters of each word would matter, meeting in the space. The space doesn’t have to be decoded for statistical purposes, it can just be a binary glyph.

I've been experimenting with decomposing EVA characters further and running statistics on them (probably something similar to the curve line system), and I discovered the possibility of positional variation while doing so.

Roughly 43.34% of words end in the y/m/g-type curve, 19.07% end in the r/s-type curve, 15.72% end in the n curve, and 15.08% end in l, which adds up to around 93.21%. Around 2.93% of words end in o and 1.58% end in d, making up most of the remainder, leaving 2.28%. If I exclude labels, titles, and shorter words, none of the numbers change much, but added together the remainder shrinks to 1.17%.

Of the remainder, most are either gallows, e, or a.

Related to this topic, has somebody else gotten tentative results suggesting an ending equivalence r ≅ ii ≅ in or otherwise? I've been toying with some unsupervised learning approaches on EVA and some preprocessing, I know I'm putting my own biases and hypotheses in the mix and don't want to fall down a confirmation bias spiral.

PS Q13