[split] Does Voynichese evolve from A to B?

[Jorge_Stolfi]

But indeed I do believe, now more than ever, that it is meaningful natural language, not encrypted.

Are you saying that you think a gradient is incompatible with the text being a meaningful natural language?

I don't have any argument against the existence of a gradual drift, between pages of a section or between sections.  I simply do not see any evidence of such thing whatsoever.

What I see in those plots (yours, mine, and everyone else's) is 4-5 clusters of dots in random positions of R^n; each cluster corresponding fairly well to one of the classic sections, and being a  more or less roundish cloud of points at random displacements from the cluster's center.

Frankly, I think that your "gradients" are just pareidolia; and the fact that you need a 3rd-degree curve to get the 4 major clusters "aligned" shows that they are not aligned -- not even along a parabolic "horseshoe".  

Moreover, the extent of the whole dataset in the MDS1 direction seems to be only a bit larger than its extent in the MDS2 direction.  Which means that using the MDS1 axis (or any other axis) as the independent variable for curve fitting is an arbitrary choice.  Which means that the order in which the curve visits the clusters is meaningless.

Actually I do have one a priori argument, admittedly not very strong, against the existence of "temporal" gradients within each section.  The VMS was surely composed by the Author as a draft on paper which was then transposed to vellum by the Scribe.  The Author must have completed at least one section of the draft (counting Herbal-A and Herbal-B as two sections) before giving it to the Scribe.  For all sections except Bio and Zodiac, he may well have written the draft in any order and shuffled its pages along the way.  For all sections, including Bio and Zodiac, if he decided at some point to reform the spelling by writing "-eody" simply as "-edy", or decided that "keedy" was a better word than "otaiin" for "herb", it seems very likely that he would have gone back on what he already wrote and made those changes there too.   But then, if there is indeed a gradient in the statistics of the pages of a section, it is unlikely to be due to those changes of mind. And therefore it is unlikely that the parameter t of such a gradient would be correlated with the time of writing of the draft, to the time of scribing onto vellum, or to the Author's intended reading order.  Either it would reflect some other factor (like the toxicity of each herb), or it would be just an artifact of random sampling.

All the best, --stolfi