10-04-2020, 02:13 PM
I wonder what scope there is for neatly subdividing words that conform to a given structure or property and words that don't conform to that structure or property. My interest is generally words that don't conform to certain general structures rather than words that do. As I subdivide words into two categories "null words" and "non-null words", which could be called real words or meaningful words, it helps to differentiate between properties that apply to one set or the other, though I can see why this could be regarded as very difficult. Obviously many statistical analyses of Voynichese are applied to the text as a whole and therefore obviously don't distinguish between these "null" and "non-null" words and so may be indicative more of the features of one subset than the other.