Re: Verb-second ... verb-penultimate languages?

staving Eldin Raigmore:

>I am thinking of constructing one or a few conlangs in which the "unmarked"
>order is verb-initial, and the "marked" order is V2 -- with
>the "pragmatically most salient" constituent moved before the verb.
>
>This would be a lot like what Ray just described for Germanic V2 languages,
>but with the "unmarked" order being verb-initial instead of SVO.
>
>How realistic is that?  Is it ANADEW?
>
>I am contemplating a situation in which the "unmarked" order would be
>noticeably less common than the aggregate sum total of all of the
>various "marked" orders, although the "unmarked" order would be noticeably
>more common than each individual "marked" order.
>
>Is that realistic, or not?  Is it ANADEW?
Over on artlangs.com, we've been working on a language (Khacveq) whose word
order is TVSO. It reduces to VSO if the topic is omitted or understood from
context, but in other ways it might resemble a V2 language. I personally
think that this could easily arise if a VSO language adopted a
topic-comment discourse structure, and that it might evolve into V2 over
time. I suggested this syntax because Khacveq is triconsonantal, and I
thought that it would be good to have something similar but not identical
to the word order typically found in triconsonatal natlangs.

This raises a question - how closely are Celtic (VSO) languages related to
Germanic (V2) languages? I know that the Germanic peoples acquired their
exonym because in Caesar's time they were allies (germani) of the Celts,
and that Celtic culture originally appeared around the headwaters of the
Danube, which is now a German speaking area.

Pete