@ Stolfi, since you’re basically not saying anything really new anymore, and I can’t really add anything new to this either—except to throw even more statistics into the mix, which you’ll again only partially acknowledge—I’m calling it quits on this discussion. It’s a waste of both our time.
I was less interested in whether and how this peculiarity could be woven into your theory than in asking the chinese participant in the Chinese thread whether such a phenomenon exists in Chinese. But I think we’d need a Chinese linguist here who’s also knowledgeable about older languages.
It also bothered me that you’ve now simply declared your theory to be proven—even though there are still significant weaknesses—especially in the assumptions you need to make to calculate your probability. You’re equating words that, based on their contextual behavior in the VMS, very likely don’t correspond to the same Chinese word. But that would open up yet another can of worms with many probabilities, which wouldn’t lead to anything else anyway. I’d much rather get back to working on my own theory.
What I find particularly interesting is that the oddities mentioned here simply disappear when the VMS is read backward, just like many others—and a much clearer structure then becomes apparent.
And this happens in such a simple way that I now want to explore this approach more intensively and further, because it has really convinced me.
But for me, a theory is only proven if it can explain these peculiarities of the VMS AND make the plain text readable. Until then, it’s all just—how do you say it?—mere theory.
Have fun 
I was less interested in whether and how this peculiarity could be woven into your theory than in asking the chinese participant in the Chinese thread whether such a phenomenon exists in Chinese. But I think we’d need a Chinese linguist here who’s also knowledgeable about older languages.
It also bothered me that you’ve now simply declared your theory to be proven—even though there are still significant weaknesses—especially in the assumptions you need to make to calculate your probability. You’re equating words that, based on their contextual behavior in the VMS, very likely don’t correspond to the same Chinese word. But that would open up yet another can of worms with many probabilities, which wouldn’t lead to anything else anyway. I’d much rather get back to working on my own theory.
What I find particularly interesting is that the oddities mentioned here simply disappear when the VMS is read backward, just like many others—and a much clearer structure then becomes apparent.
And this happens in such a simple way that I now want to explore this approach more intensively and further, because it has really convinced me.
But for me, a theory is only proven if it can explain these peculiarities of the VMS AND make the plain text readable. Until then, it’s all just—how do you say it?—mere theory.
Have fun
![[+]](https://www.voynich.ninja/images/collapse_collapsed.png)