Re: Reflexive encoded by case-stacking?
|From:||Keith Gaughan <kmgaughan@...>|
|Date:||Monday, January 16, 2006, 19:50|
Henrik Theiling wrote:
> Keith Gaughan <kmgaughan@...> writes:
>> Henrik Theiling wrote:
>>> Does any conlang (or natlang) encode reflexives by adding two
>>> case-endings? E.g.
>> My Térnaru does that, however, it's a kinda-sorta-trigger language. I'll
>> give an example. 'An' is the patient trigger and 'a' is the actor
>> an-a-Lídu ták _or_ Lídu an-a-ták.
>> Lidu hit himself.
> Ah, nice, that's about what I was searching for. How does the marking
> work? Why can the two prefixes be with both words? 'T?k' means 'to
> PS: This mailer cannot handle utf8 as you might have noticed...
Not to worry. Yup, 'ták' means 'hit'.
The reason why the agent and patient particles can be affixed to either
'Lídu' or 'ták' is that there's a subtle difference in meaning between
the two. In 'an-a-Lídu ták' the focus of the sentence is 'ták', whereas
in 'Lídu an-a-ták', it is 'Lídu'.
As I said, it's a kinda-sorta-trigger language. They started out as case
particles, as in Japanese, and mutated from there. The realisation that
I could stack the case particles to encode reflexivity is what lead me
to accidentally give it triggers.
Térnaru has a small number of particles that indicate roles in a
sentence. These indicate how the various constituents of the sentence
relate to one another. The language doesn't really have verbs as such;
though 'ták' appears to be a verb, it's really just a noun expressing
the concept of hitting.
That's pretty much the size of it.