Re: The Language Code (take 4)

En réponse à Roger Mills :


>Dirk Elzinga wrote:
> > M       morphology
> >         h       head-marking (+/-)
> >         d       dependent-marking (+/-)
>
>I'm sure we've gone into this before, but please explain. How does it differ
>from---
Head-marking and dependent marking indicates whether it's the head or the
dependent part of a grammatical relationship which is morphologically
marked for that relationship. For instance, in a genitive construction, the
head is unmarked, while the complement receives a mark of genitive
(apposition, case affix, etc...) to indicate the genitive relationship.
That's dependent-marking. When such a construction is rendered with a
construct state however, it's the head which receives some mark indicating
the relationship, while the complement doesn't. It's head-marking.

Same for the relationship between a verb and noun phrases. If the nouns get
case marks indicating their function in the clause but the verb stays
untouched, it's dependent-marking. If the function of the different
participants is indicated by agreement patterns on the verb, it's head-marking.

But note that most languages have a mix of head and dependent marking. For
instance, Latin has both case endings (dependent-marking) and agreement of
the verb with the subject (head-marking). Classical Arabic has both
construct state (head-marking) and genitive (dependent-marking). Japanese
leans towards the mostly dependent-marking side, while I guess some
polysynthetic languages that encode most of the participants' functions in
a big verbal complex are leaning rather towards the head-marking side. And
a language like Basque, with both cases and verbal complexes agreeing with
the subject, object and even dative object, is quite in the middle. And
when languages lack overt marks, we could say that they are not marking at all.

Christophe Grandsire.

http://rainbow.conlang.free.fr

You need a straight mind to invent a twisted conlang.