On Thu, 16 May 2019, 09:28 [log in to unmask], < [log in to unmask]> wrote: > I personally believe that the Voynich MS is a conlang. Any 15th century > cipher for a natlang could be cracked, and the manuscript has statistical > properties that argue against it being gibbrerish. > Has anyone tried to reverse-engineer a Voynich conlang (as a kind of Shannonian exercise)? I gave it about ten minutes' thought a few years ago and couldn't see any obvious way to construct a conlang to match its properties. But I imagine someone with more ingenuity or more time must have had a stab. (To be clear, I mean not trying to discover the conlang used by the creator of the document, but rather trying to create a conlang that can generate a text that matches the Voynich text.) --And.