IPA as Latin [was: Re: Fixing Two Unicode Asymmetries in case conversion]

From: Kenneth Whistler (kenw@sybase.com)
Date: Mon Nov 16 1998 - 14:44:19 EST

John Cowan replied to Michael:

> Michael Everson scripsit:
> > Not at all. Of course IPA is Latin -- it simply doesn't make _use_ of case
> > distinctions, which is a matter of practicality to ensure simplicity and
> > legibility of transcriptions.
> I can't agree. IPA is shape-sensitive in a way that Latin is not:
> the letterforms are (in principle) fixed, a bit like the Zapf
> Dingbats block. Unicode kludges around this, but doesn't really
> satisfy it. Xerox Character Code, OTOH, had a completely separate
> IPA block, caseless and with fixed glyphs.

I have to agree with Michael. IPA is a particular variant of the Latin
script, with many constructed extensions not in general use in
many orthographies for particular languages (but increasingly so),
with no use of case, and with particular constraints on rendering
occasioned by the need for precision and reproducibility of transcription.

It is still recognizably Latin in its use of the core Latin alphabet
and in its use of accents, and in its general layout rules.

For Southern California English, to describe the fact that orthographic
"Spahn" = orthographic "spawn" = broad phonemic /span/ = IPA
phonetic [span], the orthographic conventions of quote marks,
slashes, brackets, etc., should be sufficient. In Unicode, it would
be unreasonable then to have to *recode* [span] to new characters
for it to be correctly represented IPA. Such disunifications tend
to create far more mischief and problems than they fix. What becomes
of broad phonemic transcription, which is often based on IPA but
isn't, for example, or of orthographies which adopt one or more
IPA characters as standard parts of their orthographies -- using them
in ways that typically do not accord exactly with the narrow
transcriptional practices of IPA proper?

Besides, there is little point in flogging this one further. The
decision to treat IPA as Latin was made back in 1989, with the
active participation of Joe Becker, who I think we can agree was
in the position to understand the shortcomings and mistakes of the
Xerox Character Code Standard (as well as its good points).


This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:43 EDT