Re: A quick question

Christopher Wright wrote:


>A sound change affected final /k/. Now, one letter represents /k/ or,
>when final, /x/. Could the /x/ pronunciation be preserved regardless of
>inflections?
>
>For instance, arc- "bear" is pronounced /arx/, but would arca "bear
>(nom)" be /arxa/?
>
(OK, so you've mixed up phonemic and phonetic representations, as we all do
from time to time.  Still, your intent is clear.)

There are two possibilities.
(1) the Phoneme /k/ > phonetic [x] in word final, but remains [k] elsewhere.
In this case you would have /ark/ [arx] vs. /ark-a/ [arka]. (This could
likely continue to be written "arc ~arca")

(2) Phonemic /k/ has shifted (entirely? need not be necessarily true) >
[x]-- in that case you'd have [arx] ~[arxa].  Whether this new sound [x] is
phonemic, or still just a variant of /k/, would depend on whether there was
also a phonemic /x/ in the language with which the new [x] < k could merge.
Even so, it could continue to be written "arc ~arca", though that would now
require some explanation.

Consider:   /k/ > [x] under some conditions, not others-- (allophony)-- [k]
and [x] are a single phoneme, call it /k/. A single symbol could be used to
write both.

or:  /k/ > [x] in all environments, but there is no other /x/ in the
language-- a phonetic change, it creates a new phoneme /x/ (and eliminates
/k/) but doesn't  create any new contrast, just new alignments.  Say the old
system had / p t k, f s /, the new system has / p t , f s x / .  The old "k"
symbol might continue to be written, but now represents [x].

or: (a) /k/ > [x] in some environments, and there is an /x/ phoneme--
partial split of /k/ and merger with /x/---
    (b) /k/ > [x] in all environments, and there is an /x/ phoneme--
complete merger of older /k/ with /x/. Changes the system by eliminating a
contrast.
In these last two,  it is now impossible for a speaker to say whether  a
given [x] derives  < /x/ or old /k/, unless the orthography continues to use
the old "k" symbol.  But too many changes of this sort, and you'll end up
with Eng.-type  spelling or worse!!

It's not always necessary to resort to analogy, though Christophe's Lat.
honos > honor is a good example (note that not all such cases underwent the
change-- flos ~floris, tempus ~temporis)

Clear?  I hope. :-)

Reply