From: Kenneth Whistler (kenw@sybase.com)
Date: Wed May 12 2004 - 14:16:46 CDT
Patrick Andries wrote, regarding the script identity issues:
> Doug Ewell a écrit :
>
> >Peter Kirk <peterkirk at qaya dot org> wrote:
> >
> >>Because each such case has to be judged on its individual merits,
> >>according to proper justification and user requirements. There can be
> >>no hard rules like "always split" or "always join".
> >
> >Nobody, neither Michael nor anyone else, ever advocates such a rule.
> >
> >
> I certainly don't think there is one rule or one definition, but I can
> well imagine several factors and thus guidelines.
Yes, we can all imagine several factors, and Mark Davis has indicated
a desire to draft a set of guidelines which will help focus the
discussion in the future regarding possible edge cases.
> I just hope there is
> some regularity in the justifications and one does not eschew
> questioning (as above) by saying every case is different.
I don't think anyone wants to eschew questioning; it is, however,
indubitably true that every case is different. There *are* no
hard and fast rules which apply, because a) the history of
writing systems is too complicated, and b) the history of
IT implementation of writing systems is too complicated.
> This sounds
> sometimes a bit too arbitrary to me.
History is funny that way -- it seldom unfolds in a neat,
categorical manner.
>
> Do no intrinsic objective qualities of scripts (number of symbols,
Not objectively measurable.
> recognizability of the symbols (?),
Not objectively measurable.
> contextual handling,
Not objectively measurable.
> directionality,
Well, this one at least, we have a pretty good chance of pinning
down, but even for this, the edge cases result in irreconcilable
arguments: is Etruscan left-to-right or right-to-left or both?
> etc.) explain the spliting/joining
No, such measures are, typically, used to *justify* the splitting
or joining of scripts after the fact. They may not *explain*
the decisions at all.
> and, in the last resort, are the
> reasons to be found in the way the user communities feel (as interpreted
> by the proposers)
Often. And this is typically not a matter of *last* resort. The
political linguistic and social issues may *predominate* over any
supposedly objective graphological analysis.
> and acceptable genealogy (this unencoded script is a
> "too distant relative" to any encoded script or it is an ancestor to
> more than just the scripts already encoded and we need to cover those
> other scripts used).
This is clearly a factor. Certainly one guideline we could stick to
pretty firmly is that historically unrelated scripts should *not*
be unified.
> Well, those would be guidelines at least.
>
> No hard rules, but guidelines.
Fine. And I'm not opposed to the preparation of such guidelines.
But people should realize that each borderline case is going to
devolve into argumentation similar to that we have seen for
the Phoenician case, no matter *what* guidelines you put in
place.
--Ken
This archive was generated by hypermail 2.1.5 : Wed May 12 2004 - 14:18:14 CDT