Re: Referent Tracking

>SWITCH REFERENCING is a part of the language, but is confined to relative
>clauses. A set of relative conjunctions go before the relative clause. The
>choice of which conjunction should be used depends entirely on what object
>(subject, direct or indirect object) is shared between the two clauses. The
>only reason I chose to implement the system is because I wanted my relative
>clauses to follow their main clause rather than be embedded in it. Examples:
>
>Jone (s) yuush maey (o) wiekfase (v)
>John cooks the rice
>
>Jone (s) yesh thoel (o) tokase (v)
>John likes birds
>
>
>
>>Jone yuush maey wiekfase zhek yesh thoel tokase
>>John, who likes the birds, cooks the rice
>>
>>
THis is interesting, because it seems very different to me to other
switch reference systems (which are often used on verbs in complements
or on clause chains). It's also a very interesting (and slightly
strange) rule that bans relative clauses from occuring inside main
clauses and forces them to occur afterwards... but on the other hand,
the rule doesn't seem unnatural to me. One question: how do you handle
subordinate (complement) clauses like for instance "she was going home"
in "she said she was going home"? Are they mid-clause or shifted to the
end (or the beginning)? And if they're shifted to the end, do the
relative clauses occur before or after them? I do you say (roughly) "the
woman said she was going home who I met in the shop" or "the woman said
who I met in the shop that she was going home"?

>Moving onto FOCUS - by which I mean identifying the most relevant piece of
>information in a clause - this is achieved by word ordering, with the word
>appearing directly in front of the verb being the most focussed word. I have
>to mention focus because it clashes with the TOPIC-COMMENT system, which in
>Gevey is shown by all "new" information going in front of the verb and all
>"old" information after it (in direct contradiction to what most other
>languages do). The new-old rule tends to overshadow the focus rules, so when
>old information is more important to the speaker than new information they
>run into a problem - which I solve by deploying three focus markers (a
>promoter, a demoter and an intensifier) when necessary.
>
>
I believe that Hebrew (perhaps not Modern Hebrew, I'm unsure) has a rule
where (generally only one) NP that's new information occurs preverbally,
and old information occurs after the verb. I don't know of any language
with exactly the same rule as yours though. I'd be interested to hear
more detail about it.

Reply