Cribs in the Voynich MS

[Mark Knowles]

It might be interesting for these 28 words to calculate the sum of the Levenshtein distances to every other word in this list.

Sharp discontinuities are not evident... but it still might be interesting.

Yes. That is problematic. It means that either there is no real basis for partitioning the set of words into two. Or one needs to think of another different, though maybe related way of doing it.

Looking at them one thing I do wonder about is length.

A typical Levenshtein distance calculation would find the two words "okol" and "okolokol" fairly different however in many ways they could be seen as much more similar.

For example I would be inclined to the view that

oar alys

Is mich more distinctive than

okchy sar

However clearly the standard mean Levenshtein distance doesn't agree.

In fact if I was manually sorting the list in order of what I think are the most distinctive I would come up with a fairly different order. I think this comes down to a bias towards longer words having greater mean Levenshtein distances. I am not quite sure how to modify the calculation to account for the word length question effectively.

I would regard

oeesa

And

ochoyk

As more distinctive.

Similarly the group of words starting something like

okchy ...

don't seem so distinctive.

Of course there is some degree of subjectiveness. Nevertheless it would  be nice to find a function that better represents my assessment.