Re: I'm back (was: Re: Leaving for three weeks...)

On 8/25/05, Julia Schnecki Simon <helicula@...> wrote:
> 
> 
> Really -- I should probably find some more friends who live a few
> hours away from me, so I can get myself some more quality
> conlanging-on-a-train time... After just one short trip to Frankfurt,
> I suddenly had a budding verbal inflection system. (There are four
> inflectional classes; I already have paradigms for intransitive verbs
> for a "neutral" tense -- I'll call it "present imperfective" for now
> -- for each class and now I'm thinking about what other tenses,
> aspects etc. I'd like to have, and whether a tenseless "aspect-only"
> system wouldn't be much nicer. I also need to figure out how to deal
> with object agreement, because that's something I definitely want to
> have.)

I'm not sure how universally this generalization holds, but 
object-incorporated verb forms seem to tend towards an imperfective, 
habitual, or durative interpretation. It makes sense. "He hoes the/some 
beans" vs. "He bean-hoes", the second meaning something like "He habitually 
hoes beans" or "He's a bean farmer". 

So maybe a *combined* TAM and object agreement system!

Object agreement (person or class) and PAST suffix = past perfective
Object incorporation and PAST suffix = past imperfective
Object agreement (person or class) and NONPAST suffix = nonpast perfective
Object incorporation and NONPAST suffix = nonpast imperfective

Now this necessitates the question of how to deal with aspect in 
intransitive forms. Perhaps just a neutral interpretation, or an indefinite 
object marker can signal the imperfective.

Now I'm looking into noun class systems. I'm envisioning a system with
> a handful of inflectional classes and a somewhat larger number of noun
> classes; each noun would be assigned to a certain inflectional class
> based on its phonetic structure, and to a certain noun class based on
> its meaning. I'm not sure yet whether the noun class will show up
> somewhere on the noun itself (like in the Bantu languages with their
> class prefixes) or not; in any case, I'm planning noun class agreement
> for 3rd-person verb affixes, pronouns, etc. My main problem at the
> moment is that there are so many fascinating noun classes/noun class
> systems out there that I have trouble deciding what *I* want for *my*
> (still unnamed, BTW) language. ;-)

From Totonacan: body-part based noun classes. The following are examples of 
numerical classifiers, these and similar roots also show up in verb 
derivation/noun incorporation, nominal derivation, and locatives. 

qaqa- (ear) handles
chaa- (trunk) sprouts, branches
kilh- (mouth) holes
laq- (face) parts/divisions
pish- (neck) bunches of flowers of leaves
paa- (belly) sacks, containers
kaak- (head) houses, squash

and so on. It's a neat idea in general: take a limited set of semantically 
related nouns (for seasons, colors, tastes or smells, gods) and slowly 
expand each classifier's range. Squashes are artifacts of the harvest 
season; round things are like squashes; big animals are like round things; 
and so on. The moon is classed as silver, and later on, women as well, and 
even later, cats, even if they're orange.

Hmm, thinking seasonally, the Ages of the world. The spear was first 
invented and squash was first harvested in -- or, at least, the current 
words for these things came from -- the Age of the Sun. So they all get the 
same classifier. Basically, classifiers indicating the historical strata of 
the language's words...

I've also started thinking about syntax (probably VSO, with questions
> derived from declarative sentence by adding a question particle to the
> sentence and putting an indefinite pronoun -- or a "pro-verb", as the
> case may be -- in the place of the part of the sentence that the
> question is asked about [umm, was that grammatical just now?]) and
> about number systems (I'm leaning towards something basically decimal
> with remnants of older duodecimal ["dozen" etc.] and vigesimal [think
> Danish -- "halfway-through-the-third-score" instead of "fifty", and
> the like] patterns). These two areas obviously still need a lot of
> fleshing out, though.

Since you started with a Mesoamerican inspiration, maybe overcounting is in 
order.
For 32, not "two and thirty" but "two towards fourty"!

-- 
Patrick Littell
PHIL205: MWF 2:00-3:00, M 6:00-9:00
Voice Mail: ext 744
Spring 05 Office Hours: M 3:00-6:00

Reply