Hallucinating distinctions (was New Jersey Dialects)

[Gordon, Matthew J.]

I don't have a label to offer but I do have some data on the construction of distinction between homophones. In a written survey of around 2100 people, some 18% claimed to have a difference between "hole" and "whole". In discussing this with students, I have had some claim to make a distinction but when they're asked to demonstrate it out loud, they usually concede that the pair is homophonous. Still, some hold to their perceived difference and say "they're different, I just can't pronounce them that way right now b/c I'm thinking about it too much" or something like that.

The couch/sofa example is different, I think. It's true that there are many ways of framing the semantic (or at least pragmatic) difference between "couch" and "sofa" (e.g. a sofa seats 4 and a couch 3), but I sense a consensus is emerging at least around here (the Midwest) in which "sofa" is defined as a nicer/fancier "couch." This is easily confirmed by the fact that furniture stores only advertise sofas. I've never seen a "couch" advertised (except in the classifieds) even by the low-end furniture retailers. Of course, it's likely the case that this semantic distinction was contrived originally in the way that Arnold suggests; i.e. by people forcing a distinction when presented with synonyms.


-----Original Message-----
From: American Dialect Society on behalf of Arnold M. Zwicky
Sent: Wed 10/20/2004 11:56 AM
To: ADS-L at LISTSERV.UGA.EDU
Subject:      Re: New Jersey Dialects
 
while i'm asking about labels for common phenomena, how about what
happens when people (often, linguists as well as normals) compare two
items side by side and go on laboring to tease out some difference
(phonetic, semantic, social, contextual, whatever) between them, to the
point of inventing -- i'm sometimes inclined to say "hallucinating" --
such differences.  (i'm soon going to post about this phenomenon in
more detail.)  ask people about  the words "sofa" and "couch", and soon
they'll be all over the map with subtle distinctions.  let usageists
fix on "partly" and "partially", and they'll tease out all sorts of
distinctions, different ones for different analysts, many of which have
no basis in actual usage.  get people focused on small phonetic details
distinguishing dialects, and let them think about their pronunciations
of homophones (especially those with different spellings, like
"read"/"reed" or "sea"/"see"), and they're likely to start hearing
differences.  or, from our recent ADS discussions: compare "sort of"
and "kind of"; or compare the noticeable (to you) idiom "in harm's way"
with "in danger".

the side-by-side comparison invites people to seek out differences.
is there a name for this effect?

arnold (zwicky at csli.stanford.edu)