Re: Revised Zharranh page

Benct Philip Jonsson wrote:
> I've looked at your page now (as much as it was on-screen readable
> with Gentium -- how come Acrobat Reader renders unhinted fonts so
> much better than MSIE?) and it looks OK except for three things:
> (1) the comma below consonants for palatals.  I for one keep
> seeing 'retroflex' dots, no doubt conditioned by the standard
> Indologist transliteration practice.  I suggest acute above as a
> palatalization diacritic -- _t' d' s' z'_ etc. for alveopalatal
> and _k' g' x'_ etc. for mediopalatal.
The comma below idea is based on the spelling of Latvian, and I've also
seen it used for palatalized Russian sounds in phonemic transcription.
T-acute and d-acute are impossible due to the ascenders; there are
precomposed k-acute and g-acute (at least in specialized fonts like
Gentium), but n-acute could be either palatal or alveopalatal according
to this scheme (not that it matters much). There's even an l-acute
(Slovak long l). But I'd run into a problem with the palatal lateral
fricative, which would logically be a slashed l with acute accent.

>  For letters with ascenders
> I use the Greek number sign U+0374 after the letter, since it IMO
> spaces better -- is more narrow -- than spacing acute in Gentium,
> which is the font I prefer for this kind of thing.
I'd prefer to avoid spacing diacritics (although the line between
spacing diacritics and spacing modifier letters is a bit vague).

> (2) Mutatis mutandis the use of IPA retroflex letters.  Mostly because
> _t_ and _t`_ are dead hard to keep separate in handwriting (in fact
> most IPA characters are troublesome in handwriting!) I prefer the underdot.
> Also the IPA letters have no uppercase.
I think the underdot is too easily confused with the comma below. I'd
prefer using the dot below as well, but then I'd need an alternative
scheme for palatals. Using the retroflex letters also leaves the
underdots available for the unlikely possibility of languages with
Arabic-type emphatic consonants. Handwriting isn't a concern (if these
languages are ever written by hand, it'd be preferable to use the
original alphabets, not a Latin transcription).

> I of course prefer thorn and eth any day of the year.  The Saami t/d
> with stroke certainly made sense when haviing to make do with
> mainland Nordic typewriters, but I hope (probably in vain) that they
> will switch over to proper thorn/eth.  For dental stops/nasals etc.
> I think the IPA bridge below or IMHO even better the Americanist
> inverted breve below makes excellent sense -- again thinking of
> handwriting convenience.  FWIW Dravidianists use a macron below to
> distinguish alveolars from dentals, {_r} (essentially = [4]
> being the only such character to occur with any frequency.
But the IPA bridge requires special fonts, while the crossed t and d are
available in the ordinary Windows fonts. Depending on the font, the
bridge might also get obscured by underlines. A thorn or eth would be
okay for dental fricatives, but it'd be confusing to use them for stops.