10-06-2025, 11:37 PM
(This post was last modified: 10-06-2025, 11:37 PM by Jorge_Stolfi.
Edit Reason: its ->it is
)
As I wrote in another thread, I don't think it is Chinese specifically, but any language with monosyllabic "words" (even if, as in Mandarin or Vietnamese, most of our words would be compounds of two or three if its "words"). That would include all the languages of at least three families recognized by linguists, Dra-Kadai (Tai-Kadai), Austronesian, and Sino-Tibetan. That is a few hundred different, mutually unintelligible languages... "Chinese theory" is just a convenient nickname for it.
The character entropy (predictability of the next character given the previous few characters) is problematic to compute (because it requires knowing what the actual characters are) and even more problematic to interpret (because it is a feature of the writing system, not of the language; see the discussion of tonal marks above). More interesting is the word entropy (predictability of the next word given the previous few words). IIRC, Rene computed this number some 25 years ago, and came up with ~10 bits per word -- which, IIRC, was quite within the range of other natural languages. I may have confirmed this myself.
The "Chinese" theory in fact can explain several features of the VMs that other theories can't explain, or simply ignore altogether. Mainly, why the Author decided to write the VMs, how he/she chose the contents, why he/she decided to invent a totally new script (instead of using Latin letters or some other existing script), why the characters have that particular look, why the words have that peculiar structure, why doubled words are common, what the "starred parags" section is about and why it has that format... And the motivation for creating a new script was not to conceal the contents of the book; quite the opposite, it was to make the contents accessible, if only to the author himself.
With all due respect for the expertise of the experts who see multiple scribes, I still think that it is very likely that there was a single Scribe (who probably was not the Author!). For one thing, the VMS is quite unlike all the manuscripts on which those experts have built their expertise upon. As Lisa observed, what makes handwriting into a fingerprint of the writer is years of practice writing in the same script and language -- which is very unlikely to be the case of the VMs Scribe (or scribes). Anyway, the observed differences in handwriting between certain pages can have several explanations, and IMHO are dwarfed by the similarities. The shape of every glyph can vary enormously within a single page, so it cannot be used as a criterion for distinguishing scribes. But we can discuss this point in another thread.
The character entropy (predictability of the next character given the previous few characters) is problematic to compute (because it requires knowing what the actual characters are) and even more problematic to interpret (because it is a feature of the writing system, not of the language; see the discussion of tonal marks above). More interesting is the word entropy (predictability of the next word given the previous few words). IIRC, Rene computed this number some 25 years ago, and came up with ~10 bits per word -- which, IIRC, was quite within the range of other natural languages. I may have confirmed this myself.
The "Chinese" theory in fact can explain several features of the VMs that other theories can't explain, or simply ignore altogether. Mainly, why the Author decided to write the VMs, how he/she chose the contents, why he/she decided to invent a totally new script (instead of using Latin letters or some other existing script), why the characters have that particular look, why the words have that peculiar structure, why doubled words are common, what the "starred parags" section is about and why it has that format... And the motivation for creating a new script was not to conceal the contents of the book; quite the opposite, it was to make the contents accessible, if only to the author himself.
With all due respect for the expertise of the experts who see multiple scribes, I still think that it is very likely that there was a single Scribe (who probably was not the Author!). For one thing, the VMS is quite unlike all the manuscripts on which those experts have built their expertise upon. As Lisa observed, what makes handwriting into a fingerprint of the writer is years of practice writing in the same script and language -- which is very unlikely to be the case of the VMs Scribe (or scribes). Anyway, the observed differences in handwriting between certain pages can have several explanations, and IMHO are dwarfed by the similarities. The shape of every glyph can vary enormously within a single page, so it cannot be used as a criterion for distinguishing scribes. But we can discuss this point in another thread.