29-04-2016, 07:31 PM
Yulia,
I see three problems so far:
1) This probably ties in to what JKP says: if you encode a Latin text into Voynichese using your method, the result contains a very large amount of highly exceptional internal word structures. Many exceptions can be found in the MS, but they are exceptions exactly because generally Voynichese words follow a certain structure. To put it in a more intuitive way, the sentence you created doesn't look like one that would be likely to appear in the MS.
2) My personal main problem is that there is little to no clear meaning to the resulting text (which is also spread across the labels then?). You could argue that it's a nonsense text or a highly metaphorical, "poetic" type, which is a possibility. In that case though, the fact that someone went through an immense amount of trouble to encode this already untransparant text, seems very unlikely. If you want to write nonsense, then why encode real words? And if your text is so unclear that nobody understands it anyway, why go through the trouble of encoding it?
To me there appears to be a certain contradiction between deciphering the text and looking for the best words on the one hand, and still turning up with something that makes little sense on the other.
3) I'm not sure yet to what extent there would be information loss when encrypting and/or decrypting using this system. It could perhaps work for cohesive, simple texts without too many "unknown" words like proper names. People can make up for some information loss when they can guess from context. But whether or not your method retains enough information on the kind of text you propose in your readings, I don't know.
I see three problems so far:
1) This probably ties in to what JKP says: if you encode a Latin text into Voynichese using your method, the result contains a very large amount of highly exceptional internal word structures. Many exceptions can be found in the MS, but they are exceptions exactly because generally Voynichese words follow a certain structure. To put it in a more intuitive way, the sentence you created doesn't look like one that would be likely to appear in the MS.
2) My personal main problem is that there is little to no clear meaning to the resulting text (which is also spread across the labels then?). You could argue that it's a nonsense text or a highly metaphorical, "poetic" type, which is a possibility. In that case though, the fact that someone went through an immense amount of trouble to encode this already untransparant text, seems very unlikely. If you want to write nonsense, then why encode real words? And if your text is so unclear that nobody understands it anyway, why go through the trouble of encoding it?
To me there appears to be a certain contradiction between deciphering the text and looking for the best words on the one hand, and still turning up with something that makes little sense on the other.
3) I'm not sure yet to what extent there would be information loss when encrypting and/or decrypting using this system. It could perhaps work for cohesive, simple texts without too many "unknown" words like proper names. People can make up for some information loss when they can guess from context. But whether or not your method retains enough information on the kind of text you propose in your readings, I don't know.