[split] Torsten's criticism of Bowern & Lindemann

[Torsten]

The parsing of words does not affect the overall word length distribution. It merely shifts the mean, as every word is now shorter. 
By the way, if you parse daiin as [da-iin?] and chor as [ch-or?] how do you parse vords like aiin, okaiin, cho, or chol?

One could parse like d-aiin and cho-r. In that case, the examples you give would be parsed aiin, o-k-aiin, cho and cho-l. I agree that this would be more of a shift than a redistribution though, but the point remains that we actually do not know word length in Voynichese.

Again Zipfs law of abbreviation is about the distribution of the word length. If the word lengths are unknown, as you now claim, it is especially misleading to state: "The mix of word lengths and the ratio of unique words to total words were similarly language-like." 

Out of curiosity, if you parse "chor" as [cho-r], how would you then parse "char," "cheor," and "cheom"? Similarly, if you parse "daiin" as [d-aiin], how would you parse words like "daiir," "daiis," or "daiim"? And how do you account for the increase in the number of letters resulting from this parsing method?