I move to spin off a new thread about AEIOU. @PaulWeiler sent me down this rabbit hole, and I'm happy to see that I'm not the only one who's been down it, because I get the sense I'm nowhere close to the bottom of it. @VViews that song you linked to adds a fascinating piece to the puzzle. Speculative exegeses of traditional folk songs, nursery rhymes, schoolyard rhymes, and clapping games is another hobby I've dabbled in, and reading the words and the proposed background to "Lo Boièr" gives me this same kick. Much appreciated!
JKP thanks for clarifying what you meant. I think it could be said that Voynichese vord composition (graphotactics, if you'll forgive me coining a word) strongly prefers final characters with tails over characters without tails.
I think I've figured out why your initial statement set off my tautology meter: There just aren't many, if any, Voynichese glyphs that meet all of the following conditions:
- Non-negligible occurrence in vord-final position
- Never contains a final stroke that could be construed as a "tail". (I define "tail" as any oblique stroke that moves leftward and crosses either the baseline or the midline.) (see Brian Cham and David Jackson's "Curve-Line System")
- Cannot be morphologically derived, in any straightforward way, from the addition of any type of tail to a glyph that does not contain one. (see Emma May Smith's theory of the equivalence of EVA=a and EVA=y)
EVA=[d] and the gallows characters are the only ones I can think of that could arguably fulfill all three of these conditions. And even those are questionable. Can any of the gallows characters be said to appear non-negligibly in vord-final position? Can the idea that EVA=[g] is just a tailed final variant of EVA=[d] be safely ruled out?
EVA=[y] could arguably be derived from the addition of a downward tail to [a], [c], [e], or [o]. Meanwhile, EVA=[l], [n], [m], and [r] make up the bulk of final glyphs that aren't EVA=[y]. EVA=[l] and [r] already contain a tail in all positions, and don't need one added, while EVA=[n] and [m], which both occur only vord-finally, can be derived from the addition of a tail to EVA=[i] (or, quite likely, [aii] and [aiii]. But that's another whole bucket of worms.)