LINGUIST List 2.754

Mon 04 Nov 1991

Disc: R-Linking

Editor for this issue: <>


  1. , 2.748 R-Linking
  2. Jacob Hoeksema, Re: 2.738 R-linking, Invariance, Gemination

Message 1: 2.748 R-Linking

Date: Sat, 2 Nov 91 23:03:54 EST
From: <>
Subject: 2.748 R-Linking
Rick Wojcik suggests that I was wrong in my critique of Stampe's
arguments for "underlying" /r/ in linking-r dialects since according
to Wojcik, Stampe does not raise the issue of automacity, which I
refer to in my critique.
OBVIOUSLY, Stampe does not raise this issue, because there is no
issue. If r-linking were NOT automatic, there would be no difficultu
fitting it into the framework of Natural Phonology. The point is
rather that, unlike the classic examples of automatic processes
it is difficult to believe that r-linking arises as a phonetically
natural process, for reasons I have already stated: it is only
found in languages (dialects) which have previously lost r. We
do not find children in Chicago or LA who spontaneously use linking
r and then "suppress" it, as befits a genuine "natural" process.
My point was and is that Natural Phonology ceases to be a testable
theory as soon as we start claiming that a process is "natural"
the moment we discover that it is automatic, as in the case before
us. The original, and interesting, claim of this theory was
precisely that automatic processes are exactly those which can
be INDEPENDENTLY shown to be natural (phonetically motivated,
found in the speech of children who ultimately lose them, attested
in many different languages and at different times, etc.).
R-linking (and a number of other examples) have been cited for
years as counterexamples to this universal. There has been no
response, beyond the current proposal to emasculate the theory
by saying that we can set up whatever underlying representations
we want, so as to get the process in question to come out looking
natural. But, even if the authors of the theory insist on doing
this, it seems to me that it will remain an interesting question
why these particular examples work the way they do. As I suggested
in my last posting, the fact that we are dealing with external
sandhi might have something to do with it.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: Re: 2.738 R-linking, Invariance, Gemination

Date: Sun, 3 Nov 91 14:46:59 MET
From: Jacob Hoeksema <>
Subject: Re: 2.738 R-linking, Invariance, Gemination
Re: R-linking
>From a posting by Manaster-Ramer:
> This is something that I
> have pointed out since 1981 at least, arguing that it is precisely
> those dialects which lost /r/ in this position that then insert it
> (or, if you accept Stampe's analysis, generalize it to all underlying
> post-central vowel environments). Likewise, it is precisely those
> English dialects that lost final /l/ that then exhibit a linking-L
> phenomenon. Likewise, as I have pointed out since 1981, Korean
> lost initial /n/ before /i/ and /y/ (y means yod not a front rounded
> vowel here). Subsequently, this /n/ gets reinserted even in cases
> where it does not belong etymologically in sandhi environments. As
> a result of which, one can hear Korean speakers rendering English
> 'not yet' as /nannyet/. The /nn/ arises, apparently, because
> of the reinserted /n/ and then the assimilation (by a regular and
> well-known rule) of the final /t/ of 'not' to that /n/.
One might add here n-linking in (Holland-) Dutch. In standard Dutch,
/n/ is deleted word-finally after shwa, but inserted in hiatus,
cf. e.g. Geef me et boek ==> geef men et boek. In dialects which
do have have n-drop, this linking n does not show up (e.g. my own
dialect of Groningen).
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue