Theiling Online    Sitemap    Conlang Mailing List HQ   

Re: Back to conlanging

From:Mark J. Reed <markjreed@...>
Date:Monday, March 20, 2006, 13:28
I just happened to pick a nasal example, btw.  I was envisioning a
more general applicability, where manner of articulation is all that
distinguishes affix consonants from each other, while place is
determined by harmony.

I'm also contemplating something a little more complicated than simple
assimilation, though, where the affix consonants are determined by the
root consonants according to rules originally based on aesthetics, as
modified over time in the natural tendency toward ease of
articulation...

On 3/20/06, Chris Bates <chris.maths_student@...> wrote:
> I've looked it up, and if you got your information from Trask then he > seems to only claim that roots (rather than stems) have only one or the > other of t(s), t(z)... affix harmony is not the general rule (and > inflectional affixes never show any harmony), although sometimes it does > occur in fossilized root + affix combinations and/or old compounds. And > palatals don't take part in the harmony at all... > So basically, as I said, Basque may have had active harmony at some > point, but the changes are mostly fossilized now rather than forcing > changes in things like inflectional affixes (unlike, say, the nasal > harmony of some Amazonian languages). > > >> Basque has sibilant harmony. Sibilants or affricates > >> within the same word must be all apical _(t)s_, all > >> laminal _(t)z_ or all palatal _(t)x_. Thus Spanish > >> _Francés_ is borrowed as _Frantses_ rather than > >> **Frantzes. > > > > > > See eg. txistu (flute) for a counterexample. Also sortzen (creating) > > and other words... I have to admit that roots with mixtures of (t)z > > and (t)s are not exactly common, but if Basque did have sibilant > > harmony it's no longer an active process, in the dialects I know at > > least. Since I know practically nothing of the French dialects, things > > could be different there... > > > > >
-- Mark J. Reed <markjreed@...>