Differences that people make but can't hear

Damien Hall djh514 at YORK.AC.UK
Sun Dec 28 13:42:56 UTC 2008

I don't have many books by me, which is why I didn't reply to this question
immediately. But, since analyses by Labov and others, the phenomenon of
people making consistent differences that they can't hear has been known as
a near-merger (of the two phonemes between which this slight difference is
made). The classic example is _merry_ and _Murray_ in Philadelphia; many
natives of the City of Philadelphia pronounce both so that they sound like
_Murray_ (which has led to the Philly stereotype 'Murray Christmas').
People who have this near-merger are often shown not to be able to tell the
difference between the two words, even when they are recorded saying both
words and their own pronunciations are played back to them. And yet a
non-negligible number of these speakers make a very small but consistent
phonetic difference between the two words, obviously too small for them to
perceive consciously, but there nevertheless when you do an instrumental

I deliberately specify that it's usually people from the City of
Philadelphia who have this near-merger: even people from the rest of the
Philadelphia urban area (the towns of the surrounding counties of PA and
NJ) have been shown not to have it (up to now). This is different from
other features of the Philadelphia accent (its short-a system, and so on),
which _do_ also appear in the surrounding urban area.

A good description of this particular near-merger, the very inventive test
cooked up to diagnose it (the Coach Test), and a discussion of how the
phenomenon could be accounted for), can be found in:

Labov, William. 1994. _Principles of Linguistic Change, Volume 1: Internal
Factors_. Oxford, UK and Malden, MA, USA: Blackwell.

Happy New Year!

Damien Hall

Damien Hall

University of York
Department of Language and Linguistic Science
York YO10 5DD

Tel. (office) 01904 432665
     (mobile) 0771 853 5634
Fax  01904 432673

The American Dialect Society - http://www.americandialect.org

More information about the Ads-l mailing list