The Constructed Language Hypothesis

[Torsten]

Is it possible to do a word-by-word substitution of the Voynich MS and come up with a meaningful text?

I see more reasons why the answer should be "no"  rather than "yes".

Indeed, there are numerous compelling arguments against a word-by-word substitution.

The strongest "pro" argument is that "the pages that are nearest neighbors in topic modeling tend to be adjacent to one another in the manuscript" [Bowern & Lindemann 2021]. However this observation only means that the text is structured in some way, it does not imply that Voynich words are indeed words or that the text possesses a linguistic structure. 

Moreover, words used on the same pages together "share similar morphological patterns, either in their prefixes or suffixes." [Montemurro et al 2014]. A well-known example of this observation is the prevalence of final EVA-edy in Currier B but its rarity in Currier A. Less recognized is the possibility of describing similar patterns throughout the Voynich text. For instance, EVA-qo rarely occurs as part of labels; in Herbal in Currier A You are not allowed to view links. Register or Login to view. is less frequently used than in the Pharma section in Currier A; You are not allowed to view links. Register or Login to view. only frequently occurs on some Herbal A-pages, and so on. "The closer two words are (with respect to their edit distance), the more likely these words also can be found written in close vicinity (i.e. on the same page)" (Timm & Schinner 2019, p. 6). It seems as if the distribution of words is caused by the distribution of glyphs combinations, rather than the other way around. 

Upon closer examination, even this "affirmative" argument essentially transforms into a "negative" argument.

There are also numerous counterarguments:

Tiltmann wrote in 1976 "My analysis, I believe, shows that the text cannot be the result of substituting single symbols for letters in the natural order. Languages simply do not behave in this way. ...  And yet I am not aware of any long repetitions of more than 2 or 3 words in succession, as might be expected for instance in the text under the botanical drawings" (Tiltmann 1976).

D'Imperio wrote in 1978: "Also the strange lack of parallel context surrounding different occurrences off the 'same' word as shown by word indexes. In the words of several researchers ' the text just doesn't act like natural language'" (D'Imperio 1976, p. 30).

Timm & Schinner 2019: "In natural languages there will be frequent words distributed equally over the entire text, the so-called function words (like conjunctions, articles etc.). They do not appear contextual, but rather serve to implement grammatical structures, and they normally do not have co-occurring similar words of comparable frequency. In the VMS frequently used tokens differ from page to page. With the exception of repetitive prayers or poems, words in natural languages are chosen because of their meaning, and not of their similarity with previously written words" [Timm & Schinner, 2019, p. 6].

Currier wrote in 1976: "The frequency counts of the beginnings and endings of lines are markedly different from the counts of the same characters internally." (You are not allowed to view links. Register or Login to view.). Because of Curriers observation even Claire Bowern concludes: "All of these observations lead to generalizations that seem typographical rather than linguistic in nature. ... A comprehensive linguistic analysis needs to take seriously the possibility that, for example, paiin, saiin, aiin, and am are all positional variants of the same word." [Bowern & Lindemann 2021].

Stolfi: "In fact, the word length distribution matches almost perfectly a binomial distribution ... This coincidence suggests that the length of a word chosen at random ..." (You are not allowed to view links. Register or Login to view.)