Quote:But why the strange alphabet?
Well I don't have a good answer for that. I am
suspecting that symbols in Voynich words are kind of numbers but I don't have any proof.
But I would say that ciphers changed through time. Modern ciphers are most often made of standard letters and numbers. Old ciphers used a lot of fancy, invented symbols. If you were making a cipher in the 15th century, you would probably feel "obliged" to use some symbols created by yourself
(08-01-2025, 07:08 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.Voynich manuscript doesn't have to be a complex cipher.
See for example:
You are not allowed to view links. Register or Login to view.
You are not allowed to view links. Register or Login to view.
It certainly isn't rocket science but if you don't know the source book then you are practically hopeless.
For me the Voynich Manuscript may be something like book cipher - a colllection of numbers pointing to some words on the list. It is extremely hard to break if for example "herb" is 97, "flower" is 520 and "leaf" is 343
If this was the case, with no extra tricks, I suspect VMS riddle would be extremely easy to solve, much easier than Rohonc. There are hundreds of labels in VMS, if, say, "otarly" consistently meant "water" and "okaldy" stood for "hot" (both examples are made up), there would have been multiple convincing reading already. I'm almost absolutely sure that if this is a cipher, it's of a one-to-many type, with several possible ways of representing a single plaintext word.
(08-01-2025, 09:10 PM)R. Sale Wrote: You are not allowed to view links. Register or Login to view.It doesn't take a highly complex system. The combination of two variable systems based on different methods plus an obscure dialect may be sufficient.
It seems that such a complex cipher could protect the owner from religious persecution. Then again, writing in this manner might be considered sorcery. The owner of the VM must have believed it was a powder keg of information.
Quote:If this was the case, with no extra tricks, I suspect VMS riddle would be extremely easy to solve, much easier than Rohonc. There are hundreds of labels in VMS, if, say, "otarly" consistently meant "water" and "okaldy" stood for "hot" (both examples are made up), there would have been multiple convincing reading already. I'm almost absolutely sure that if this is a cipher, it's of a one-to-many type, with several possible ways of representing a single plaintext word.
Well the problem is that there may be not a single "hot" word:
You are not allowed to view links.
Register or
Login to view.
So you may have different forms of "calidus" word in the ciphered text and don't know that they are forms of the same word.
Declension of words makes much harder to spot word patterns. Rohonc Codex seems not to have declension so it has much more repeated patterns than Voynich.
I would like to try once to search for word patterns (taking into account the declension), assign meanings to words and see if it works in other places. As far as I know nobody tried it seriously so far. But it would require time.
This is a good point, but I suppose most nouns in the labels will be in the same case?
Quote:This is a good point, but I suppose most nouns in the labels will be in the same case?
Possibly. By logic they should be the most basic form - "nominativus".
Of course we don't know even the original language used in Voynich Manuscript. Latin is just a guess.
(09-01-2025, 12:31 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.So you may have different forms of "calidus"
It seems to me to be highly unlikely that the manuscript is written in an inflected language. In inflected languages each noun and adjective changes its ending to reflect changes in how the word is to be used. Meanings are determined more by how a word ends and less through any accompanying prepositions. The alternative to a language being inflected is to be analytical, like English. Nouns in these languages do not inflect. Meaning is determined more through prepositions and word order. One consequence of a language being inflected is that the average word lengths are longer, due to the need for fewer short word prepositions and due to the need for extra letters at the end of words. You can see this clearly in languages such as Polish and Russian. Yet words in the manuscript tend to be short. The average word length in the text passages in the Glen Claston transliteration ( v101 ) is only 3.9 ( excludes labels and circular text ). Too short.
If the table for 'calidus' is telling us that you have the hypothesis that the manuscript is Latin then you have some additional problems. Verbs in Latin have passive forms which lead to even less need for prepositions to qualify meaning.
But if you really are welcoming the challenge of trying to find inflections in the manuscript you might like to make things more challenging for yourself by considering the possibility that the manuscript may be in a celtic language like Irish or Welsh. One feature of those languages is the presence of eclipsis or consonant mutations at the start of words. Have fun!
Yes, short words in Voynich are a serious problem.
But I still tend to believe that the language is rather inflected. I did some calculations and may show some stuff soon.
(09-01-2025, 04:54 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.But I still tend to believe that the language is rather inflected.
I had very much the same idea when I first started my own investigations. But after I had generated, tabulated, analysed and correlated many statistics about word suffices I eventually came to the conclusion that there was little evidence to show for an inflected language. I suspect that many other people have tried the same. Latin as a possible language has also been suggested by many people without success.
Morphemes could be decoupled from a stem or lemmatized/canonical form of the word, as in cal ∅idus +Accus sal chedy qokeey (which I'm sure is something many researchers have considered), but if that were the case, we should be seeing a lot more frequently occurring word groups than we actually do. Unless maybe the morphemes are transpositioned away from the stem, say, to the edge of lines. But still, I believe we should have seen enough patterns by now if the process is deterministic enough.