Discussion of "A possible generating algorithm of the Voynich manuscript"

[RenegadeHealer]

I'll play lawnchair academic for a few minutes and tell you how I'd go about testing the autocopy hypothesis. I'd put an ad on college campuses pretending to be a props and costumes designer for a movie set in medieval Europe or a fantasy world like it, and looking for arts students who wanted to make a bit of side money helping create a realistic-looking mysterious codex shown in close-up. An exclusion criterion would be any familiarity with the VMs. Teams of five students would be assembled and given a replica of the VMs with all the images but no text, just faintly indicated boxes where text should go. They'd get taught the glyph set and how to write each VMs glyph with a quill pen, given some examples of ways to combine these to make vords, and given a simple explanation and demonstration of how to use the autocopy method to generate lines of vords. They'd then take turns generating fake text until the codex was full. After collecting the work of five such teams from different colleges who had no knowledge of each other and finished the project while following directions, each codex would be transcribed and parsed statistically, with the same metrics for the original VMs and for the output of Torsten Timm's app. If the VMs differs significantly from these human autocopied specimens on at least one metric that the autocopy hypothesis purports to explain, I'd deem the autocopy hypothesis falsified.

Is there any historical precedent for a large amount of pseudo-text generated by analog methods that would qualifiy as variations of the autocopy method? If yes, how do these pseudo-texts break down statistically? If no, why not? Torsten Timm has consistently said that his autocopy method is one of the few practical pre-modern ways to generate large amounts of gibberish. At the beginning of his first paper he also hints (quite rightly) at some plausible motivations for a medieval person to pull such a scam. If both the method and the motivation are rather common — or at least intuitive — then there should be historical examples of such a thing happening, even if only fragments, or none, of the pseudo-text itself survives. Such a find would bolster the autocopy hypothesis immensely.

I would be very interested to hear what an expert on the history of asemic writing would have to say about the VMs. Also any of the people who design fake languages for fantasy worlds for a living. I would fully expect that both would conclude the VMs was meaningless gibberish; when you hammer for a living, everything looks like a nail. The value would be in the small details they noticed as being similar, or dissimilar, to the material they typically work with or create.