Ideas for Moving Voynich Research Forward

[Mark Knowles]

At the same time, there is no reason to believe that these words behave differently from any others, since they represent a representative sample. 

I question this statement. How are they a representative sample of the whole text? Surely by the process of selection of any kind you introduce a bias.
I want to agree with you, but your research method seems very problematic.

I said that they form a representative subset for the word types and not that they are representative for the text. 

Since I checked all the words occurring 7 and 8 times in the Takahashi transcription there was simply no degree of freedom on my side to introduce any bias in selecting specific words. Per definition the words occurring 7 and 8 times are overrepresented in this sample. But this is known and if you believe that words occurring 7 and 8 times in the Voynich text behave differently you can repeat the sampling method by choosing different numbers. Keep in mind that the sampling method was only used to search for patterns. In the end I checked the patterns found against the complete set of words.

I question whether "they form a representative subset for the word types".  By selecting only those occurring 7 and 8 times you introduce some bias. I believe that words occurring 7 and 8 times in the Voynich text behave differently from words occurring at some other frequencies. I would be curious as to what your results say for words that occur 1 time in the Voynich text.