(07-05-2023, 11:01 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.The first is from D'Imperio. Figure 30 on page 108 lists the 'houses of the moon' commonly also referred to as the 'mansions of the moon'. According to traditional astrology there are 28.
Now the astronomical / cosmological figure on You are not allowed to view links. Register or Login to view. shows 28 words each covering a segment of a circle.
This is as good a crib as one can get.
And the uneven width and spacing of the "logs" suggests that the "Artist" started drawing them at about 04:00, with the
saral one. At first he left clear gaps between them. But, as usual, he did not plan, and the widths of the logs got random but generally larger. After drawing the first 20 logs, up to
okeol at 01:15, he realized that he still had 8 logs to draw but only ~75 degrees of space left. So he started making them thinner and closer together. But still he had to draw the last three right next to each other, with no gaps.
Since the labels are well aligned with the logs, it seems that he wrote each label after drawing the corresponding log (or maybe the other way around).
So the list of labels that he got from the Author presumably started with
saral,
saiir, ... and ended with
okeod. What is the usual order for the Mansions of the Moon?
Quote:But it even gets better. D'Imperio's figure shows that the majority of these names start with an 'a'. In You are not allowed to view links. Register or Login to view. the majority starts with Eva-o or Eva-y.
Like most star names, they are probably derived from Arabic names, and the initial "a" is from "al-", the definite article.
Quote:There are three or so that start with 's' and there are also three or so in the MS that start with Eva-s.
In the spoken language the article mutates to "ar-", "az-" or "as-" depending on the first letter of the noun. So those Mansion names that start with "s-" may have been "as-sXX" in Arabic.
All the best, --stolfi
Another partial crib (apologies if it has been discussed before) is the "primeval scream" on f70r1,
........ooooooooolar...sara....
It suggests that the o on its own stands for a specific sound (rather than being a modifier, silent marker, or encryption symbol); and it is a vowel, or at least a sound that can be sustained like 's', z, 'sh', 'zh', 'rr', 'l', 'm', 'n', 'f', 'v', 'h'. As opposed to a plosive like 'k', 'g', 't', 'd', 'p', 'b', 'th', 'dh', 'r'...
Beware however that the "scream" may be just hallucination by a BEEP! sorry.
All the best, --stolfi
I have read the older thread about cribs and well... none of them worked so far.
I was also reading a bit about Chinese astrology. I don't know Chinese astrology, I don't know Chinese language but when
I saw one picture I immediately noticed something:
You are not allowed to view links.
Register or
Login to view.
[
attachment=13701]
In one minute I got some Chinese numerals and made bigger progress with Chinese than people with Voynichese in 100 years
I give it as example. Natural languages have patterns. Diagrams which are structured have even more patterns. Even constructed languages like the one from Rohonc Codex have a lot of patterns. These patterns are cribs.
But Voynichese doesn't have similar patterns. It has some very "inhuman" patterns like similar words or word parts appearing in clusters in rows and columns:
f49r
[
attachment=13702]
These patterns aren't cribs. They are a problem. A serious problem that needs to be explained.
(28-01-2026, 01:34 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.It has some very "inhuman" patterns like similar words or word parts appearing in clusters in rows and columns:
As for sequential repetitions, many languages, use You are not allowed to view links.
Register or
Login to view. of "words" (or syllables) for certain grammatical functions, such as indicating plural or collective (Chinese
天 tiān "day" -> 天天 tiān tiān "every day"), repeated action (Vietnamese
nói "to talk" -> nói nói "to keep talking and talking"), turning adjectives into adverbs, etc.
As for "vertical" patterns, one should always check whether a supposedly "non-natural" pattern occurs in ordinary text just by chance:
[
attachment=13708]
Perhaps these chance alignments are more common in Voynichese than in any random language. But that needs to be measured, and one must take into account the fact that there are more similar-looking words in Voynichese than in French or English, and the frequency of use of each word is probably higher since the lexicon is rather small.
All the best, --stolfi
Quote:But that needs to be measured, and one must take into account the fact that there are more similar-looking words in Voynichese than in French or English, and the frequency of use of each word is probably higher since the lexicon is rather small.
Well, we could do such a test but it wouldn't convince you of anything.
There is really a huge lot of vertical patterns in VM. I could invent some formal definition of them and count them in the entire manuscript. It would be a long, tedious and boring job. Then I could do the same for some manuscript in Latin. Then I could do some formal statistical test and compare the results.
But if they were different and VM had more of these patterns you would say
it just means that VM has a different structure from my chosen Latin manuscript and nothing more. And VM is most probably not in plain Latin anyway so it is an irrelevant result.
--------------------
And I would like to add some more general observations.
People tried really hard to read VM and haven't read a single word. There are surprisingly few potential cribs and the ones we have aren't working.
Actually we know something. We can safely exclude several possibilities. Voynichese for 99% isn't:
- simple substitution of European language
- simple transposition cipher
- homophonic cipher
- syllabary
- abugida
and possibly several more.
So the options left and usually mentioned are:
1) gibberish
2) nomenclator
3) some megacipher
4) an exotic language
I would also add
5) an exotic language written in cipher or megacipher
If 5) would be true then I am out
I believe we should try in the future to discuss these options and maybe try to prove if some are more possible than others.
But "exotic language" is a solid, concrete bunker with thick walls which is not necessarily a good thing. It is very hard to prove it as proving would basically mean cracking the manuscript but disproving is equally hard.
(29-01-2026, 12:30 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.There is really a huge lot of vertical patterns in VM. I could invent some formal definition of them and count them in the entire manuscript. It would be a long, tedious and boring job. Then I could do the same for some manuscript in Latin. Then I could do some formal statistical test and compare the results.
I don't think the number of vertical patterns in VM is higher than expected given the properties of the text (small alphabet + CLS or grammar rules). So, while the number of vertical patterns seems high, I'm not sure this requires a separate explanation.
Quote:So, while the number of vertical patterns seems high, I'm not sure this requires a separate explanation.
Yes, it could be as well a result of low entropy of Voynichese words. They are schematic, similar to each other, 50% of words end with "9" and so on.
But here I can imagine a good test for it.
1) Count the patterns
2) Mix the words
3) Count the patterns again
4) Compare the results
But I won't be doing it any soon
(29-01-2026, 01:09 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.1) Count the patterns
2) Mix the words
3) Count the patterns again
4) Compare the results
Mixing the words will destroy the LAAFU properties of the text, and some of the vertical patterns can certainly appear by chance just because some character combinations prefer ends of lines and some other combinations starts of lines.
I think a better test would be randomizing the order of the internal (not the first and the last) lines of paragraphs. Also, we don't have a reliable way of counting the patterns, so maybe we just need to find some people who haven't yet memorized the manuscript to any significant extent and show them the scrambled and unscrambled versions of the same page and just ask which one appears to have more vertical patterns.
Quote:we don't have a reliable way of counting the patterns,
We could leave the job to people or computers. I believe it could be automated
1) word A is above word B if X coordinate of A is close enough to X coordinate of B
for example
ABS(Xa - Xb) <15 pixels
2) There is such thing as
Levenshtein distance You are not allowed to view links.
Register or
Login to view. to count similarity of words
We could use some modified version of it
(29-01-2026, 12:30 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.Well, we could do such a test but it wouldn't convince you of anything. Then I could do some formal statistical test and compare the results. But if they were different and VM had more of these patterns you would say it just means that VM has a different structure from my chosen Latin manuscript and nothing more. And VM is most probably not in plain Latin anyway so it is an irrelevant result.
I am not
claiming that the "vertical patterns" in the VMS are chance coincidences. I am only
warning you that they
may be. So you should not assume as a fact that "the VMS has an unnaturally large number of vertical patterns".
Quote:I could invent some formal definition of them and count them in the entire manuscript. It would be a long, tedious and boring job. Then I could do the same for some manuscript in Latin.
This can actually be done by computer with a modest amount of work. I could do it myself if I didn't have a dozen other computer analysis projects going on...
Quote:So the options left and usually mentioned are: 4) an exotic language ... But "exotic language" is a solid, concrete bunker with thick walls which ... disproving is equally hard.
It is hard to absolutely disprove "Latin with a complicated cipher", "gibberish", or "nomenclator". But there are only a limited number of plausible candidates for the "exotic language". If that is the case, it will eventually be solved. Even without knowing the language, one could identify key words and common sentence structures, and maybe match those with one of the candidate languages. It would be more like deciphering Hittite, Minoan Linear B, or ancient Maya, than cracking Enigma or the Zodiak Killer letters...
Quote:I believe we should try in the future to discuss these options and maybe try to prove if some are more possible than others.
Well, I am quite convinced that it is an exotic language in the plain, written phonetically.
Possibly with 20% of errors. Check the diary of the Lewis and Clark expedition, a military mission that explored the Western US in the 1800s, before the Pioneers went there. It has an amazing amount of spelling errors, and totally inconsistent. I once counted 4-5 different spellings of the word "buffalo", sometimes two in the same paragraph. If the VMS is like that, it will make decipherment quite a bit more, ahem, "fun"...
All the best --stolfi