[Article] The Voynich Manuscript: A Non-Stochastic Glossolalia

[nablator]

4.  There is no mention of peer-reviewed Voynich publications that could be a problem for this theory (e.g. Lisa Fagin Davis' work on scribes comes to mind first).

The apparent variety of scribes is not really a problem: the transcription could have been done by anyone, and later it was re-written to the VMs by the same or different people.

Maybe LAAFU effects? I do not expect an outpouring of glossolalia (which is like a religious experience) to take lines into account at all.

Yes, all line properties are proof that Voynichese is not simply a transcription of speech.

I was thinking of something more difficult to achieve on the fly without a written record and some kind of procedure (like marking long sequences of reused vords somehow) or algorithm to limit the number of re-occurrences of any sequence of 3 vords to just zero or one in Currier A.

I am not sure that this property is 100% true because of the many mistakes and ambiguities in transliterations. There are 27 re-occurring sequence of 3 vords in paragraphs of Currier A pages of the ZL transliteration, including sequences of vords than span two lines.

The probability of this happening by chance, with the known affinities of some patterns across vord breaks, and the known line properties making re-occurrences more likely than in a uniformly random draw of vords, is difficult to evaluate: it could have happened non-intentionally.