Re: Query: Origin of nara(ba).

mirror wrote:
Hello,

Is `nara(ba)' formed as follows -- `na' (same one as the one used
with na-type adjectives, that is, a form of `da') + `ra' (some sort
of archaic generic nominal like `mono' or `koto') + (`ba') (the
ancestor of the particle `wa'?

It seems `nara(ba)' nominalizes the phrase it conditionalizes, so
it seems to follow that this nominalization must modify a generic
nominal, which then can get marked by the optional particle `ba'.

You're very close. '-naraba' is "-ni" + "araba." Meanwhile, the
na-adjective ending is "-ni" + "aru" minus "ru." (Note that "-da" is
historically "-ni" + "-te" + "aru" minus "ru.")

The "-ba" wasn't originally optional (I suppose it's debatable whether
it is historically the same as "-wa" or not; I believe it is); people
just decided it wasn't worth the trouble of leaving on after "-nara" and
"-tara" because the words were still distinct. If you left if off
things like "ikeba," you'd get other valid words such as "ike."

Speaking of "-tara," the relative-past marker "-ta" began as "-te aru"
minus "ru."

Bart
.