Re: Lexical Relatedness Morphology (wa Re: [Conlangs-Conf] Conference Overview)

staving David J Peterson:
>Ka kavaka Pete ti:
><<
>Over the weekend, I was thinking about this with relation to
>Khangaþyagon, which has a very Item-and-Arrangement morphology - pure
>aggultination, to the extent that language guesser software has been
>known to identify it as Turkish. Did I avoid that trap, and if so, how?
> >>
>
>Well, of course, Turkish is a natural language.  Some languages
>do more than others with pure affixation.  It wasn't affixation
>that's the problem, but more what you get at in your continuation:
>
>Pete:
><<
>I think I did, because I closed most of my bound morpheme categories
>fairly early on - that is, I don't allow myself to just create new
>morphemes to solve problems, but have to find creative ways of using
>the morphemes I've got, or make the syntax do the work.
> >>
>
>That was my problem with my first language.  I didn't even
>conceive of the notion of closing any of my categories.
That's interesting - it had never occurred to me that categories of
grammatical affixes could be anything but closed. Interestingly, though,
when I look at the page of my notes where I originally worked out
Khangaþyagon's noun system, it looks like in a couple of places I left
space for more segunakar to be added if I needed them - but later decided
not to.

Another thing was that I had always been interested in being able to
combine segunakar to build up fine distinctions of meaning - hence the
system of ranks. This has had some interesting consequences. For example,
my translation for the Primordial Soup relay contained the sequences

ut-     omb
at      around

and

eb-             gri-    am
contact above   destination

I had intutitively thought that the "ebgriam", meaning "to a position on
top of" would be fairly transparent, and that "utomb", meaning "composed
of" was quite opaque. Sally, however, found it much easier to work out
"utomb" than "ebgriam", although I think this was partly due to context.

>Pete:
><<
>Another thing was that a lot of the work in developing Khangaþyagon,
>especially the noun paradigm, was not in creating the items but in
>working out the arangement - the point where I worked out the system
>of ranks for the segunakar was a major breakthrough, and was the
>start of the transition from sketchlang to usable.
> >>
>
>I found a post on this, and it looks interesting.  I also tried to go
>to your grammar page, but it said the host name had expired.
>Do you still have it?  Before renewing, you might consider getting
>in touch with Christophe Grandsire and seeing if he'll get you a
>free.fr page.  They're free, and the service is incredible.
I've got backups of everything, and I'm hoping that Aaron will be able to
reactivate the artlangs.com site - but if he can't, nothing has been lost.

"What a woman says to a passionate lover
Should be written on wind and running water" - Catullus, tr. Bleackley

"Try the web" - Pete

>Pete:
><<
>I must get round to downloading some of the other talks
> >>
>
>They're all good.  For a good talk on case, I recommend everyone
>look at Matt Pearson's talk:
>
>http://video.google.com/videoplay?
>docid=-7308759491555175687&q=language+creation+conference
>
>You may have to e-mail him to get a copy of his handout, though.
>
>Ditto John Quijada and cogsci, John Clifford and semantic primes...
>Pretty much everyone and their topics.
I've now seen Sally's, and I'm agout halfway through John Quihada's.

Pete

Reply