Just about every list I've seen of potential solutions to the VMs includes the hypothesis that Voynichese might be a constructed language. But compared to other broad categories of hypothesis -- such as cipher, natural language, and hoax -- the constructed language category doesn't seem to be getting much attention these days. I'm curious why that is -- whether it's because people feel this explanation is unlikely, or because nobody is quite sure how to go about evaluating it.
Friedman and Tiltman apparently took the idea seriously -- see e.g. You are not allowed to view links. Register or Login to view. -- but I don't have a good sense for how far they actually pursued it. Bowern and Lindemann (2020) bring it up too, but mainly to suggest that a conlang would "pattern morphosyntactically with other natural languages, but be anomalous at the root level," referencing Lingua Ignota, Balaibalan, and Enochian as points of comparison. Still, for a constructed language that diverges from expected patterns, the work of Wilkins, Dalgarno, Cave Beck, and so on might be more relevant: even though they're all a couple hundred years later than the VMs, they might help give us some idea of how "weird" an early constructed language might be, and in what general ways.
So how could we imagine a conlang differing from a natural language in the fifteenth century? And how would efforts to figure it out need to differ from efforts to identify a "precursor" natural language for Voynichese?
For what it's worth, I see that our friend Athanasius Kircher You are not allowed to view links. Register or Login to view. a universal language in his Polygraphia Nova et Universalis that looks structurally a bit like Voynichese:
For another fun example of a text in a seventeenth-century philosophical language, see You are not allowed to view links. Register or Login to view..
Friedman and Tiltman apparently took the idea seriously -- see e.g. You are not allowed to view links. Register or Login to view. -- but I don't have a good sense for how far they actually pursued it. Bowern and Lindemann (2020) bring it up too, but mainly to suggest that a conlang would "pattern morphosyntactically with other natural languages, but be anomalous at the root level," referencing Lingua Ignota, Balaibalan, and Enochian as points of comparison. Still, for a constructed language that diverges from expected patterns, the work of Wilkins, Dalgarno, Cave Beck, and so on might be more relevant: even though they're all a couple hundred years later than the VMs, they might help give us some idea of how "weird" an early constructed language might be, and in what general ways.
So how could we imagine a conlang differing from a natural language in the fifteenth century? And how would efforts to figure it out need to differ from efforts to identify a "precursor" natural language for Voynichese?
For what it's worth, I see that our friend Athanasius Kircher You are not allowed to view links. Register or Login to view. a universal language in his Polygraphia Nova et Universalis that looks structurally a bit like Voynichese:
Quote: xxxvii36 xxx21 ii5 xxiii8 xxviii10 xxx20 xxx22 xvii29 xxx28 xiii16 xxix12 xxx22 xii3 xxx28 ii13 xxix5 viii25 xxix20 xxx28 xxiii40
For another fun example of a text in a seventeenth-century philosophical language, see You are not allowed to view links. Register or Login to view..
![[+]](https://www.voynich.ninja/images/collapse_collapsed.png)