Print

Print


On Jul 6, 2009, at 8◊39 AM, Jim Henry wrote:
>
> Grouping abjads, abudidas and alphabets into one set, it seems the
> three major sets of writing systems are based on phonemes, syllables
> and words.   David J. Peterson, in his talk on orthography design at
> LCC3, said if I recall correctly that no natlang writing system is
> based on morphemes per se (which I think he took as evidence for
> morphemes being a theoretical construct with no real basis, at least
> in natlangs).

Ahem: As *further* evidence.  I still don't get what their allure is-- 
for
conlangers, at least.  All conlangs end up looking the same, like a
series of toy choo choo trains...

> So a morphographic or mixed writing system where at
> least some glyphs represent bound morphemes which are neither
> syllables nor stand-alone words would be something not occuring
> naturally (modulo an appropriate definition of "natural" for writing
> systems, which would not be the same sense of "natural" we apply to
> spoken languages).

And the funny thing is I did something like this once:

http://dedalvs.com/tantyls/orthography.html

This is kind of an abjad, in that only important segments are written
(what I call X's, which are all the consonants and the main vowels).
The language operates that way, too, so that there are no constraints
on what X's can come next to each other.  After the word is formed,
though, the little vowels (reduced vowels that harmonize) fill in the
gaps to make the word pronounceable.  There are some words that
are five vowels in a row.

Here, though, the various affixes are represented by individual glyphs.
The only part that would be problematic for a DM morphologists
would be the circumfix and the...well, I don't really have a name for
it.  A combination of an infix and a suffix.

Though naturally unattested, this type of system wouldn't be all that
interesting.  What would be interesting is to come up with a new  
analysis
of English and then come up with a writing system for *that*.  Say,
instead of morphemes, you analyzed language as comprising
linguons.  Each phoneme comprises a certain number of posiguons,
neguons and neuguons, and the number of posiguons and neuguons
determined the phoneme's weight in the Periodic Table of Phonemes,
which, I imagine, would look something like this:

[h] = 1 posiguon
[T] = 2 posiguons
[f] = 2 posiguons, 1 neuguon
[p] = 3 posiguons, 1 neuguon
[i] = 5 posiguons

And so forth.

At this point, you could come up with simple glyphs with the number
of posiguons and neuguons necessary to create each phoneme, and
you'd come up with, basically, an alphabet.

Or, you could take a word and combine them all.  So let's say you
had a word like "heath".  That comprises [h], [i] and [T], which is
eight posiguons.  So what you'd have here is a number system for
writing the number of posiguons and neugons, and then some way
to distinguish words with the same number (let's say if [hihh] were
a word).  You might have something like this:

VIIIp0n@
VIIIp0n#

And, of course, the @ and # would be the same for all words (you
might need...I don't know, ten of them?), and then words could be
grouped by those artificial classes.  That'd be fun!

-David
*******************************************************************
"A male love inevivi i'ala'i oku i ue pokulu'ume o heki a."
"No eternal reward will forgive us now for wasting the dawn."

-Jim Morrison

http://dedalvs.com/