VMs: Stroke harmony. Was: Has anyone been down this route before?

[Jacques Guy <jguy@xxxxxxxxxxxxxxxx>]

31/08/2004 1:06:18 PM, Koontz John E <John.Koontz@xxxxxxxxxxxx> wrote:


>But not always - right?  I see a fair number of ae sequences.

I just had a look and found 14 occurrences of <ae> but 6666
(wow! that's a worry!) of <ai>

> And one
>might argue that there's no sign of this principle with the various
>gallows characters, taking t and k as the straight stroke forms.  And o
>seems pretty independent of this, too.  Am I missing something in the
>characterization?

No. The gallows and <o> do seem to be "harmony neutral".
You will observe similar properties in real languages which
have vowel harmony, Finnish for instance, where e and i are
"harmony neutral". Some languages also have consonant harmony 
(Javanese and some varieties of Mongolian, if memory serves), 
with, again, some "harmony neutral" consonants.

Let me try another count... 
<ea> 449 occurrences, <ia> 2, yes, only two occurrences.

More...

<eo> 3480   <io>  11  huge difference
<oe>  409   <oi> 335  no difference

Interesting, isn't it?

One last count:

<or> 2821   <ar> 3244  no difference

So yes, <o> can be followed by either, curved or
straight. But see how <e> calls for straight.

The terminology (glyph, grapheme and so on) leaves
to be desired. Add to that that the Russians (in Rongorongo
studies and, I believe, in Mayan studies too) use "grafem" 
for graph, or glyph, or grapheme, and that the French have 
followed the Russian usage, and you have a complete mess.
To communicate without confusion you have to redefine 
everything every time. That is one of reasons which
made me write, in my review of Kennedy and Churchill's
"The Voynich Manuscript" that "without a guiding methodology, 
which we lack, which we need to develop, its study leads 
only to follies worthy of Newbold's." Graphs, graphemes,
and so on, are only one very small part of the general
mess.




>My instinct is to see a glyph as a perceptual whole, a real element of the
>orthographic string, whereas a graph is a constituent mark used in writing
>glyphs that might be either part of a glyph or be a glyph in its right. A
>feature in Chomskian terms!
>
>For example the ae ligature is really two glyphs a and e in Latin terms,
>though it tends to develop into a glyph in its own right in writing other
>languages like Old English.
>
>For graphs, in writing a t, there are two graphs, the stem and the
>cross-stroke.  Or, ae, in systems where it is a glyph might be considered
>to consist of two graphs, the a part and the e part.
>
>The use of the term graph in the glyph : graph opposition, however, is
>quite different from its use in the set grapheme, graph, allograph, where
>the meaning of the element graph is more like that of the element glyph in
>the former.  Perhaps it would be best to abandon the term glyph and speak
>of graphemes, graphs, allographs, and sub-graphs or graphical
>constituents?  The alternative is to speak of glyphemes, glyphs,
>alloglyphs, and graphs. I haven't come up with a solution that satisfies
>me.
>
>______________________________________________________________________
>To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
>unsubscribe vms-list
>
>



______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list