Harvard physicist's take on the VMS

[RadioFM]

But a complex gibberish generation method that was stable and produced complex but well-defined word structure, with many statistics compatible with those of a natural language (Zipf's law, lexicon size, entropy per word...) would be even more anachronistic.  Statistics and probability theory were some 150-200 years in the future.  Devices that could implement such a method with stable probability distributions at a convenient speed were 500 years in the future.

Zipf's Law, lexicon size and entropy can arise naturally in human gibberish. Gaskell and Bowern measured this in their Gibberish paper.  Well-defined word structure can be a byproduct of wheels, or (IMHO) a Curve-Line stylistic choice, flourishes at word boundaries and a reduced alphabet; the bulk of the text is written using fewer than 23 symbols ( Modern alphabet - { j, u, v} ).

Why would the author choose those illustrations -- which are unrthodox but mostly "boring" -- instead of illustrations that would make the book more attractive, like a man making gold or walking over fire or in bed with a dozen nymphs?

Let's suppose it's an actual herbology/medicinal book. Where would the author(s) get their knowledge of herbology from? Self-taught, a quack? Hard not to draw from other sources, especially since there are a couple that were well cemented in medieval academy. A good deal of the plants should be matchable to known herbs of the time, even in your case, you should be able to match herbs to East Asia or whatever.

Marginal notes on a book are usually added by later owners.  The VMS probably left the Author shortly after it was written.  And he may well have been unable to read those tiny letters himself (like Marci and many elderly people of the time) because of failing eyesight.  (Which could also help explain why he recruited an incompetent Scribe to create the vellum version, instead of doing it himself).

I won't go into this because we can't falsify the VMS being a copy of a previous manuscript of East Asian tradition, just as we can't falsify the VMS being written by Wilfrid Voynich using 1900 sources and old materials.

I gather that punctuation was still somewhat of a novelty at the time.

If I tell you the 1430s author(s) were most likely familiar with capitula and rubrication and the sort, you'll probably say "Oh but it's actually a copy on vellum of an older manuscript, and the original authors weren't", so there's that.

Relatively short amount of text (when entropy-corrected) 

What do you mean? Short compared to what?

Conditional character entropy depends, as you said, on the encoding of text. When contracting the verbosity of Voynichese to match conditional entropies of European texts, the outcome is a shorter text. I'm drawing a comparison between the effort of making all the hundreds of nymph drawings, all embellished with headgear and clothes (which thankfully your copyist dilligently kept when making the replica on vellum?) versus the amount of text we expect the VMS to contain, were it meaningful.

These observations too do not imply that the text is gibberish.  They may well follow from the contents being not yet another rehashing of some established European lore and manuscript genera, but either is some completely original creation or is drawn from some non-European culture. Like the Rohonc Codex, or the Book of Soyga.  Or like the Rodia towers are to architecture, or the Costa Rica stone spheres are to Mesoamerican archaeology, or those dodecahedral thingies to Roman archaeology...

True. Then again, you have swallowtail merlons on the Rosettes, not East Asian architecture, and European motifs throughout like T-O maps, a cross, possibly the 4 Ages of Man like Koen pointed out, European and not Asian Zodiac...

Namely, even if the Author wanted to make the pseudo-language of his fake book look exotic, he would probably have unconsciously followed some text patterns of European books of that genre.  He would probably have added nonsense tables, distinctive words that looked like numbers, etc.

They actually did. y on word boundaries drawn from latin con- and -us, letterforms like ch, r, a, o. None of which would read outloud like an East Asian language when taken at face value, mind you.

confidently dismissing research without glancing at it just because the conclusion isn't to one's liking is not how it works.

It is not how it should work in an ideal world where everybody has infinite time and stamina.  But in practice that is just how it works, and must work.  I don't try to check the arguments of articles that claim to prove that Earth-is-flat or vaccines-cause-autism or btcoin-is-better-than-money.  I am quite convinced that those conclusions are wrong, so my expectation of benefit-minus-cost for reading them is strictly negative.  You too will do the same with articles that make those or other claims -- don't you?

Yes of course, I have my own set of biases and I draw a line. Personally I would love for the text to be meaningful, but honestly I've taken a look at the gibberish evidence and to me, it seems plausible. The VMS probably wasn't expensive to produce, and hoaxers most certainly existed. If it doesn't seem plausible to you, then let's agree to disagree, but I dispute the implicit comparison of the VMS Gibberish Hypothesis to Flat Earth, for instance, as a fair one.  The latter would require extraordinary evidence, leaps of logic to explain astronomical empirical data and massive, long-time running conspiracy theories to been proven true, whereas the former would not. 

Anyway, as you may have noticed, I found a book that matches the Starred section surprisingly well, including all the "bizarre" statistics and word structure, the apparent absence of grammar and function words, no punctuation, no numerals, no reference to humors theory,  repetitiousness, etc etc.  But that is the topic of another thread...

Sounds good.

---

[JoJo_Jost]

Register or login to view the content

[Jorge_Stolfi]

Register or login to view the content

[RadioFM]

Register or login to view the content

[RadioFM]

Register or login to view the content

[JoJo_Jost]

Register or login to view the content

[Jacob Carolus]

Register or login to view the content

[Jorge_Stolfi]

Register or login to view the content

[ololololo]

Register or login to view the content

[ololololo]

Register or login to view the content