VMs: Re: Weirdoes, ligatures (_short_)

[Jacques Guy <jguy@xxxxxxxxxxxxxxxx>]

06/10/02 18:39:20, steve ekwall <ekwall2@xxxxxxxx> wrote:


> We have the question of the <i> strokes in middles of words.  Is
> <ii> in the middle of a word simply an <in> without the end
> stroke?

You cannot tell without studying the frequency distributions,
and even so, you may not be able to tell.  There are many
similar cases in "real" spelling systems. In Malagasy for
instance the sound of /i/ is spelt <y> word-finally, <i>
elsewhere. You can guess that if and only if you do a frequency
count. Then you will see that <a>, <e>, <o> (no <u> in that language)
occur in all positions, but <y> does only word-finally, and
<i> _never_ occurs word-finally. That is strong evidence that
<i> and <y> represent the same grapheme (are environment-
conditioned allographs--to use the jargon--occurring in
mutually exclusive distributions). 

I did such a frequency analysis for <ee>, <a> and <y> 
a long time ago, and concluded that they were allographs
of the same grapheme, too. (It's in the archives).