polka(dot) pronunciation?

[Dennis R. Preston]

Of courses post-vocalic /l/ is not "optional" in the US. The variety
of /l/ which occurs in this position is extremely complex as are the
surrounding features which condition it.

First, there are "clear" and "dark" /l/s. The first has the entrire
body of the tongue raised and fairly broad contact between the apical
and post-apical areas of the tongue and the dento-alveolar area. In
the latter, only the apex of the tongue is raised (the body is kept
quite low in the mouth, even depressed or "cupped"); contact, which
is at the dento-alveolar area for clear /l/, tends to be retracted.

That said, there is a tendancy for "dark" /l/ to be "vocalized" (or
made more vowel-like). Contact between the tongue and any part of the
alveolar (prepalatal) area is lost and an /U/ like vowel ("good") is
heard; the "roundness" of this vowel (glide) is usually, but not in
all varieties, apparent.

Some poeple think that /l/ simply disappears (as it does in my
speech, for example in "wolf" and in that of others in "help").
Please note that these cases of disappearing /l/ all invoilve
clusters, not simply post-vocalic /l/.

Finally, conditioning is very complex and certainly dialect-specific.
For example, in my South Midland (Louisville, KY area) speech, if the
preceding voeel is tense (peel), I get an /l/ somewhere between light
and dark but usually articulated with contact in the post-alveolar
(prepalatal if youm prefer) area. If the vowel is lax (pill), I get
vocalization, with the above-mentioned /U/ glide (and accompanying
rounding). Folk south of me get this vocalization even after tense
vowels.

Of course, the "near loss" of /l/ has historically resulted in many
misperceptions. The one I encounter most frequently in my work on
folk linguistics is the reanalysis of the word "drawl" as "draw"
(with the accopmanying folk etymology that drawlers are people who
"draw" out their words. Such misanalysis has also resulted in
hypercorrection (now often severly stigmatized) in such form as
"ideal" for "idea."

Since /l/ and /r/ share features, this whole bidness is not unrelated
to the story of postvocalic /r/, but this is too long awready.

As usual, when dictionaries say something like "optional" thereby hangs a tale.

dInIs (whose Milwaukee wife still sarcastically barks at him when he
says "wolf")

>A quick glance at some dictionaries (OED, Cambridge, AHD, RH, Macquarie, 4
>M-W's) indicates that British and Australian pronunciation tends to keep
>the /l/, while it is optional in the US. Of these, only the M-W 9th
>Collegiate goes so far as to show "polka" only with /l/, "polka dot" only
>without. My own perception/experience is that both are pronounced both ways
>but that the /l/ is much more often pronounced in "polka" (dance) than in
>"polka dot".
>
>-- Doug Wilson

--
Dennis R. Preston
Department of Linguistics and Languages
Michigan State University
East Lansing MI 48824-1027 USA
preston at pilot.msu.edu
Office: (517)353-0740
Fax: (517)432-2736