VMs: Re: Quills & Hands A & B...

[Bruce Grant <bgrant@xxxxxxxxxxxxx>]

Nick Pelling wrote:

> Philip Neal mentioned a related idea: what if the production process
> involved a final encoding stage done by a copyist using a verbose
> (expanding) encoding scheme.

I have always wondered if it couldn't be something as simple as two scribes
copying text from a draft version and inserting some null characters on the fly
as they copy. Differences in their preferences in the choice of nulls and the
frequency of inserting them could account for the difference of the hands. The
intermixing of hands by page, but never (I believe) within a page could be the
result of each scribe grabbing a page or two of the draft, copying them, and
coming back for more when he is done.

If there are a few characters which are always nulls, this could explain the
unusually large number of letters in the alphabet (compared to Latin, for
instance).

One way to test this hypothesis, it seems to me, would be to look for a small
set of characters (the nulls) which, when they are deleted, cause text of both
hands to have similar characteristics (e.g. word length, letter frequencies,
word frequencies, etc.)

I tried some testing along these lines a few years ago, but didn't find
anything conclusive. With today's fast PCs, though,  it should be possible to
automate such a search, I think - maybe I'll take another look at it.

Bruce