Re: Gallows G characters

["John Grove" <John@xxxxxxxxxxxx>]

Glen,

It's been so long since I've looked back at the charts to compare
Currier-et-al versions. I vaguely remember not liking them because of
inconsistencies in how they were applied and sought a new 'clearer'
transliteration.  With EVA, we can build a single character if that's what
we want to call it - while others who don't see the 'same' character we see
aren't forced to adhere to our belief that the character is split or joined
at certain points...

    Take the problem that you've had in developing your own transliteration
alphabet. You state you no longer use some of them, which is
understandable - we've all gone through the same process in trying to get a
grip on that one basic problem -- WHAT exactly is a character.

    Minding your P's and Q's, M's, N's, 6's, and the variations of any
(i-stroked) character preceded by 1-3 i's. There are 'iir', iil, iij type
words out there. Sorry for the EVA - but these would be something like PY,
PE, and PW in your script. There are matching (c-stroked) characters as well
that would be similar to your U8, UD - kind of - but rare, US.

    My long-winded point is that with EVA - I can choose to explain a
'character' as ii in - or as iiin as I wish to use it. In your case daiin
could be either 8aPD or 8am, or (I know you said you don't use 6 anymore -
but 8a6).

    If your statistics always treat the combination daiin as three
characters - then I don't see a problem with your stats being misconstrued
by counting d - a - iin as the three characters. However, is someone else
doesn't think that iin is the whold character and daiin is actually 4
characters - then they could run their stats accordingly - d a ii n -- whose
results are the right ones?

    You are wrestling with the same script problem we all are -- Determining
what is a character should be simple - but it isn't. Just saying that 'iin'
is a whole character in itself isn't simple as you've discovered for
yourself.

    From my personal bias - there are two main forms of characters and I'm
perhaps the most guilty for the 'stroke' count philosophy. The i-form and
c-form characters are very similar, each having a matching form in the
other.

    The i-forms almost always follow an 'a', but are otherwise very similar
in design to the c-form. The i-forms are almost always word final, although
the c-forms are quite frequently word final as well. It's truly great that
we can attack this problem from a world of different perspectives. I for one
enjoy the full scope of this undertaking - I like to hear the points brought
forward by either cryptologists, linguists, herbalists, etc... I don't agree
with them all, but they certainly provide other angles to consider.