LINGUIST List 2.748

Sat 02 Nov 1991

Disc: R-Linking

Editor for this issue: <>

Directory

1. Jim Scobbie, Re: r-linking
2. , Re: 2.720 R-Linking
3. Jim Scobbie, Re: 2.738 R-linking, Invariance, Gemination

IANSMITHVM1.YorkU.CA writes:
>There is some phonological evidence for r-insertion, however:  derived from
>V-reduction can also trigger r-insertion in many non-rhotic varieties. As in
>"The wind[] [r] isn't broken" [hypothetical] or "See ya [r] Ian" [attested in
>natural speech]. We don't want to claim window and you have underlying /r/,
>nor to we want to have vowel reduction produce [r], so here there seems no
>viable alternative to r-insertion.
I do. I want to claim that there is a close phonological relation between
the feature definitions of /schwa/ and /r/ in non-rhotic dialects. I bet
the [] before the [r] is different from a phonetically reduced /ow/. But
we've got no data on that here, so its just a claim.
--
James M. Scobbie: Dept of Linguistics, Stanford University, CA 94305-2150

Message 2: Re: 2.720 R-Linking

Alexis Manaster-Ramer has attempted to point out a "fundamental flaw" in David
Stampe's reasoning on R-linking. He alleges Stampe's position to be the
following:
 (a) r-insertion is not phonetically natural and (b) it is automatic
 (a process rarther than a rule in NP terms), yet (c) according to NP
 only phonetically natural developments can be processes (i.e., automatic)
I have gone over Stampe's contribution, and I can't find where he raised the
issue of automaticity. He did take the position that R-linking was not a
phonological process, but he linked the prima facie case for processhood
to other criteria--e.g. the subjective difficulty that speakers feel in trying
to suppress "R-insertion". Stampe's happy solution was that the apparent
problem resolved itself in his framework when you took /r/ to disappear by
process (derhoticization). He offered corroborating evidence, such as the
fact that so-called "intrusive r" only appeared in de-rhoticizing dialects.
I think that Stampe has long recognized the existence of morphonological rules
that are fully automatic. Unfortunately, the best know publications by him
and Pat Donegan do raise the issue of automaticity as a "Process" property.
But this has more to do with the fact that they never defined Rules very care-
fully. The problem was that they tended to attribute properties of derivation-
al morphonology to the entire class of Rules. (In fact, the distinction
between derivational and inflectional morphonology was practically nonexistent
in those days, and still is for many linguistis.) So David and Pat tended to
say things like "Rules don't necessarily apply in speech performance" and
"Processes, but not Rules, interact with tongue slips." These claims are
simply wrong when you look at a/an suppletion, liaison, contraction, etc. I
would maintain that derivational morphonology does apply during speech
processing--but not to very familiar words like 'electricity'. Their raison
d'etre is vocabulary augmentation.
Now, Ian Smith points out that expressions such as "The wind[] [r] isn't
broken" and "See ya[r] Ian" are counterexamples to Stampe's claim on the
grounds that the alleged intrusion occurs after phonological vowel reduction.
That would be an easier argument to make if reduced vowels weren't easily
confused with the phoneme // (or /r/ under Stampe's analysis). It is
perfectly permissible to have allomorphic variants of words in most theories.
I think that lexical phonologists might dislike the idea of having lexical
and post-lexical operations that cover the same ground, but no such parsimony
exists in Natural Phonology. You can achieve the same effect as vowel reduc-
tion by replacing vowels consciously with the schwa phoneme. English speakers
know this and even incorporate it in so-called "eye dialect" writing--e.g. "Ya
know, doncha?" Note that non-neutralizing allophonic variation tends not
to be represented in eye dialect. You can also get there by attempting to
pronounce the regular unstressed noncentral vowel and letting vowel reduction
take you the rest of the way. Both paths are allowed in NP.
					-Rick Wojcik (rwojcikatc.boeing.com)

Message 3: Re: 2.738 R-linking, Invariance, Gemination

I wrote:
>... We seem to spend quite a lot of
>time in this group bandying relatively trivial stuff around.
Well thankfully no one so far has sent me the obvious email message...
I must have been in a grumpy mood or something, because this gives totally
the wrong impression of what I think about this forum. So I'd like
to get in quick and retract my moan.
--
James M. Scobbie: Dept of Linguistics, Stanford University, CA 94305-2150