24-09-2026, 08:43 AM
As I said, I won’t be investing any more time in the discussion about the Chinese theory, but for many reasons that have come up in my research (some of which aren’t even related to the Chinese theory), I’m no longer convinced, since much of it doesn’t align with “normal” language. The problem is, I’m not familiar with the specific peculiarities of the Chinese language, and that simply makes me uncertain about applying these statistical oddities to Chinese.
However, when I see how many assumptions you need just to establish your theory in the first place, I’m still not convinced, too.
For example, you use a VMS passage that belongs to the passages in the VMS with the highest aiin/air ratio. 40 percent (!!!) of the tokens contain an aiin or an air. That makes such comparisons much easier. “If you also take into account how much leeway you allow for the position of each individual crib, a match is by no means unlikely—especially if several different Voynich forms are accepted as matches.”
This is also evident from the fact that earlier (if I understood correctly), you initially used “dair” in line 34 and have now switched to the following word, “okaiin.” So I’d say there’s a lot of variation!. But even without the “air” variants, you still have 25 percent of the tokens that contain “aiin.” But u say: the different “aiin” variations are all errors - the VMS text actually consists almost entirely of errors, which can then be used to justify absolutely ANYTHING. Explanations like that just leave me with a very uneasy feeling. And you’d feel the same way if the tables were turned!
I also think the selection of which of these “aiin” words belong and which don’t has been retroactively adjusted and justified. Tricky...
But, to avoid getting bogged down in further discussions: If you can produce plain text consistently across multiple paragraphs without making too many new assumptions, that would be much better proof. So I’ll just wait for that… it’s less time-consuming than continuing to discuss this here.
-----
As for the backward direction: There’s a difference between having a letter in the forward direction that influences the first letter of the next token across several identical glyphs, and having a letter in the backward direction that influences the next letter and the one after that, creating a completely normal, linguistically coherent chain from which this connection becomes apparent. (In addition, in many everyday languages—and presumably in Chinese as well—the last letter or letters do have an influence on the next word. In any case, this influence is greater than that of a letter at the beginning of a word on the next token. But u dont need that in backwards!!)
The example of “qo” clearly illustrates that there really is a difference between forward and backward:
In the forward direction, “qo” is an atom because “q” is followed by “o” nearly 100 percent of the time. In this respect, one must group “qo” together as a single atom in one’s analyses.
In reverse, there are the variants “o” at the end of a word and, alongside that, the variant “oq.” As far as I can recall, the ratio is roughly 60 to 40. In this respect, the “o” does not belong 100 percent to the “q,” but is essentially a variant of the “o” ending. That is, how should I put it, easier to work with. And according to Ockham’s Razor, I prefer these simpler things.
...
In reverse, the structure of the VMS is also much more uniform and fits together better - see here: You are not allowed to view links. Register or Login to view.
However, when I see how many assumptions you need just to establish your theory in the first place, I’m still not convinced, too.
For example, you use a VMS passage that belongs to the passages in the VMS with the highest aiin/air ratio. 40 percent (!!!) of the tokens contain an aiin or an air. That makes such comparisons much easier. “If you also take into account how much leeway you allow for the position of each individual crib, a match is by no means unlikely—especially if several different Voynich forms are accepted as matches.”
This is also evident from the fact that earlier (if I understood correctly), you initially used “dair” in line 34 and have now switched to the following word, “okaiin.” So I’d say there’s a lot of variation!. But even without the “air” variants, you still have 25 percent of the tokens that contain “aiin.” But u say: the different “aiin” variations are all errors - the VMS text actually consists almost entirely of errors, which can then be used to justify absolutely ANYTHING. Explanations like that just leave me with a very uneasy feeling. And you’d feel the same way if the tables were turned!
I also think the selection of which of these “aiin” words belong and which don’t has been retroactively adjusted and justified. Tricky...
But, to avoid getting bogged down in further discussions: If you can produce plain text consistently across multiple paragraphs without making too many new assumptions, that would be much better proof. So I’ll just wait for that… it’s less time-consuming than continuing to discuss this here.
-----
As for the backward direction: There’s a difference between having a letter in the forward direction that influences the first letter of the next token across several identical glyphs, and having a letter in the backward direction that influences the next letter and the one after that, creating a completely normal, linguistically coherent chain from which this connection becomes apparent. (In addition, in many everyday languages—and presumably in Chinese as well—the last letter or letters do have an influence on the next word. In any case, this influence is greater than that of a letter at the beginning of a word on the next token. But u dont need that in backwards!!)
The example of “qo” clearly illustrates that there really is a difference between forward and backward:
In the forward direction, “qo” is an atom because “q” is followed by “o” nearly 100 percent of the time. In this respect, one must group “qo” together as a single atom in one’s analyses.
In reverse, there are the variants “o” at the end of a word and, alongside that, the variant “oq.” As far as I can recall, the ratio is roughly 60 to 40. In this respect, the “o” does not belong 100 percent to the “q,” but is essentially a variant of the “o” ending. That is, how should I put it, easier to work with. And according to Ockham’s Razor, I prefer these simpler things.
...
In reverse, the structure of the VMS is also much more uniform and fits together better - see here: You are not allowed to view links. Register or Login to view.
![[+]](https://www.voynich.ninja/images/collapse_collapsed.png)