Harvard physicist's take on the VMS

[Jorge_Stolfi]

And I agree! I don't think wheels or any complex device was involved in the generation of the text, I prefer self-citation to that. Again, despite the limited amount of text used for the paper, Gaskell & Bowern's results providence evidence for the gibberish case as a natural tendency of some individuals; we have actual empirical data that is consistent with the gibberish hypothesis, and I can't stress "empirical" enough.

What can I say? I am not impressed by those experiments...

when you start grouping combinations like those you mentioned and parsing them as single characters (thus increasing the alphabet), the text naturally shortens.

Now I am not sure what you mean for "shorter text".  Do you mean that the words become shorter?  The number of words in the text obviously stays the same.

The main problem with this approach is that there are some many instances of ambiguous parsing, that there's no way to systematically group characters together and get sensible word-to-word ("phrase-level") syntax. We have instances of chol, ol, lol, dy, dchdy, choldy. How do you group characters in dcholdy? dch-ol-dy? d-choldy? d-chol-dy? dch-oldy?

That is a problem for us, but the Author obviously knew how to parse the words!

But my word model is ambiguous only when there are three or more e in a row: cheeedy could be {ch}{eee}{d}{y} or {che}{ee}{d}{y}.  But again, that is a problem for me -- the Author would know which is the right parsing.

The next-character entropy of Voynichese is lower at the start/prefix of the word and then rises, and the per-word conditional entropy is somewhat comparable to European languages, but I'm not sure if one of those two metrics induce the other.

It depends on how you define the conditional character entropy.

If you use only the previosu character -- that is, you compute the entropy from the probabilities Pr(x[k+1]|k,x[k]), tthe sum of those entropies wil over-estimate the information contents of the text.  

If you use the whole prefix of the word -- that is, the probabilities Pr(x[k]|x[0..k-1]) then the sum of those entropies should be the entropy of the whole word. 

I don't know about East Asian languages

I computed that decades ago.  My recollection is that it was on that ballpark.  But I used the wrong knd of text.  I should recompute it from the Shennong Bencao.

Most Voynichese glyphs are constructed from pairs of simple pen strokes: a left half which is either i or e, combined with one of 5-6 different strokes at right in almost all possible ways.(...) Thus the similarity of some glyphs to Latin letters is a not-at-all-unlikely coincidence. 

Oh come on! They even let a -que abbreviaton slip in there! There's no way it's all 100% pure stochastic coincidence it just so happens to look like Latin letterforms lmao. ... some of the glyphs are also seen on ciphers from ca. 14-15th century, perhaps before.

Just one weird p that looks like a "-que" abbreviation in the whole VMS? and in the middle of a word?

And please make up your mind: did he borrow from scribal abbreviations, or from cipher symbols?

The Armenian alphabet has letters that look like S, U, L, t, n, q, j, etc.  But their sounds in Armenian are totally unrelated to their sounds in Latin or other European languages.  Ditto for Georgianand may other scripts. Even Ciryllic has "P" with sound "r", "C" with sound "s", "N" flipped with sound "i", "H" with sound "n", "m" with sound "t"...

There are only so may ways to make a symbol with 1-2 pen strokes.  Coincidences are more than likely, they are practically inevitable.

Anyway, the way those glyphs are use in Voynichese implies that their phonetic values is not at all like those of the matching Latin letters.  Taking these at face value is a common mistake of beginner voynichologists.

Why would a European trying to transliterate an East Asian language devise a glyph system, clearly inspired by letterforms of his time/place yet for some reason phonetically not at all related, is beyond me.

I answered this question many times.  Once more:

• Maybe for the same reason that people invented and re-invented You are not allowed to view links. Register or Login to view. scripts: because writing Latin letters is too slow for recording normal speech in real time.
  
• Maybe because the language had phonemes that would be messy to Romanize. Like six different consonants with sounds similar to "sh", or clicks, or You are not allowed to view links. Register or Login to view., or tones...
  
• Maybe because he was a native You are not allowed to view links. Register or Login to view.  (or Georgian, Jew, Persian, Turk, Hindu ...) and thus would not have thought of using the Latin alphabet as model for his Burmese script?
  
• Maybe for the same reason that the Klingon alphabet does not resemble the Latin one: because he wanted to make it maximally cool and different?
  

All the best, --stolfi