On Wed, Nov 13, 2002 at 03:26:19PM +0100, Christophe Grandsire wrote:

> En réponse à Amanda Babcock <[log in to unmask]>:
> "my" as an infix? Nice!

Heheh :)  Thanks :)  I was having this issue with suffixes, see, where
the whole "what is a word" question seemed to threaten to make long
agglutinative words sort of, well, topologically indistinct from long
analytical sentences (Japanese can do that to a person :)  I wanted there
to be no question about what is and isn't a word...  Hence the infixes :)

> Funny enough, Possessives in Maggel are prefixes, but
> with nouns derived from a specific kind of verbs (called "strong verbs"), they
> become infixes too ;)))) .

I love Maggel.  (Where's the stress on that?  I wonder if "love Maggel"
would be "mamaggel" or "magaggel"...)  Rather, I love the *idea* of Maggel,
as I'm too timid to get to know the actual language :)

> Vocabulary remark: "topic" typically refers only to new information. What you
> call "topic ... for old information" should rather be called "focus".

Oh dear.  I'm still confused about all that.  I meant topic like the topic
marker "ha" in Japanese.

> Checking on your site, I discovered that the first 'e' is merely a repetition
> of the following vowel.

Yep! :)  I left that out because the infixes would look kind of complicated
if I included place and manner of infixation :)

> And this time it's the consonant which gets doubled.


> Are all verbs infixes?

A few are suffixes or prefixes :)  But yes, they're all derivational
processes applied to nouns.  And they always get derived back into
nouns before they can be used in a sentence.

> In this case, I guess there is only a limited amount of verbal
> infixes, and to make other verbs you create different expressions with
> different nouns and different verb infixes

Yep, exactly.  I'm thinking of things like using "to be a knife" (or maybe
"to be like a knife") for "to be sharp", vs. "to use a knife" for "to cut",

> How is this one pronounced? [j], [Z]?

I mean it to be [j].  Haven't got a [Z] or [dZ] (I do have a [c], or is
it [tS]?  I have a hard time pronouncing a real [c]).

> If they don't have the same reference as the previous comment, how do you know
> which topic it refers to then?

I think it will refer to the original root noun embedded in the previous
comment, which will, depending on the verb derivation that was applied to
it, be the object of an action performed by the topic noun, the instrument
by which the topic noun performed an action, the state that the topic noun
is experiencing, etc.

For example, say, "John was eating an apple and [it/he] exploded".  I don't
have all the words for this yet :), but if it were John who exploded, it
would be something like

        iJohn[AGT] we-explosion.undergo[PAT]
        John eats the apple and explodes

whereas if the apple exploded, it would be

        iJohn[AGT] su-explosion.undergo[PAT]
        John eats the apple and it explodes

At least, this is how I intend to use it currently.  It may turn out that
I need some way to introduce a third party to act on John.  Maybe I just
need passive voice for that.

The original inspiration for this system was what I read about clause-linking
(if that's the right term; my head is fuzzy this morning) languages in
"Language Typology and Syntactic Description: Volume 2, Complex Constructions",
further influenced by Dirk Elzinga's switch-reference system in Miapimoquitch.

> Or do you allow topics (or rather focuses,
> topics can't be dropped since they refer to new info) to be unstated, which
> would make the language pro-drop?

I guess I have to decide this.  I'll say currently not, which will require
some mental adjustment after all the exposure to Japanese :)

> You could introduce a passive, which would transform the verbal infix "love"
> into "be loved", and add the result to "myself" put in comment.

I could, but I dislike the idea since due to the constraints I've put on
how the language derives words, one cannot derive a new verbal infix from
an old verbal infix directly; one could only derive, say, "cat-love" (love
by a cat) from "cat-love" (love of a cat), which does not seem kosher to
me.  It's like the cat went in expecting one thing, and then you suddenly
turn the tables on it :)

On the other hand, if I need to have separate active and passive (and
instrumental, and stuff) verb forms for every verb, I could run out of
phonemes pretty quick :)

> all one of the main use of the passive voice. Of course, since your passive
> wouldn't make the subject unnecessary, it wouldn't be a real voice... You can
> call it "inverse" instead if you want ;))) .

Yep, something like that.  It'll be fun to figure out how to do that.

> Extremely interesting. My Moten uses a lot of infixes too (together with
> prefixes and suffixes), and since it features overdeclination (the possibility
> to add a case marker - usually a combination infix+suffix - to an already
> declined noun, or even a conjugated verb)

Oh, neat!  How does this work?

My automated script that checks for ambiguously derived words has a big
problem with applying "could" and "must" consecutively, since they infix
at opposite ends of a word and there's no trace which was applied first.
I've rationalized that they'd never *be* applied consecutively, so the
ambiguity is ok.  But now I feel I must read up on overdeclination, just
in case it could give me any ideas about why I would *want* to apply
consecutive modals to a verb...

> and various sound changes due to the
> addition of those affixes, words can at times be extremely difficult to parse.
> But you go even further than me in that respect ;))) .

It's a kind of Maggelity :)  I was hoping for a result that couldn't
really be parsed effectively :)

> It's nice anyway. Keep up the good work!