Michael Everson wrote:

> >But what is the nature of the correspondences? That is the aspect of
> >"featural" that I have not seen clearly explained, and I
> would really like
> >to know what people are meaning when they say that.
>
> It could be anything. It depends on the script. [...]

But, then, one could conclude that almost every script is "featural"...

Even Gaelic consistently uses an "h" on the right of consonants (or a dot on
top of them) to show aspiration, and an acute accent to show vowel length.

So if you consider "á", "bh", "ch", "dh", "é", etc. as single "letters", you
could say that the Gaelic script is featural. And English uses the same "h"
to show palatalization in "ch" and "sh", and so on...

My understanding was that the term "featural" applies to a writing in which
the main graphic units denote "features" (or "traits" or "phonetic
properties"). So, if I want to write "can", I should write a sequence such
as "<velar><aspirated><stop>, <vocalic><open>, <dental><nasal><stop>", where
each expression like "<velar>" would be a single graphic symbol.

But my understanding is clearly wrong, because I don't see how this would
apply to Hangul.

BTW, Evan Kirshenbaum's notation to represent IPA in ASCII
(http://www.hpl.hp.com/personal/Evan_Kirshenbaum/IPA/faq.html) is an actual
example of symbols representing features.

_ Marco