LINGUIST List 12.2228

Thu Sep 13 2001

Sum: American Pronunciation

Editor for this issue: Karen Milligan <>


  1. Larry Trask, American Pronunciation

Message 1: American Pronunciation

Date: Wed, 12 Sep 2001 16:21:23 +0100
From: Larry Trask <>
Subject: American Pronunciation

Re: Linguist 12.2166

A few days ago, I posted a question about the transcription of
American pronunciation. This arose out of the decision of the editors
of the Oxford English Dictionary to transcribe the American "tapped
/t/" ("flapped /t/") with the symbol /d/, thus making the
transcription of 'atom', for example, identical to that of 'Adam'. I
expressed my disapproval of this decision, and asked for the views of

I received replies from 50 people, most (but not all) of them
linguists or phoneticians, and most (but not all) of them native
speakers of American English. In addition, two respondents consulted
friends or spouses who had no linguistic training, giving a total of
52 respondents.

First, I have tried to classify the respondents according to their
opinion of the OED's decision -- with a little difficulty in a few
cases. Here are the numbers.

 37 Utterly hostile to the use of /d/
 7 Skeptical about /d/, but not wholly hostile
 1 Sympathetic, but still skeptical
 3 Sympathetic to the use of /d/
 4 I can't tell

 52 Total

Now, I can't pretend that my sample is representative of American
opinion on this point, but it is clear that there exists a great deal
of dissatisfaction with the OED's decision. Accordingly, I propose to
pass the comments on to the editors of the OED and to ask them to
reconsider their decision.

A number of points were raised by the respondents. Here are some.

1. For many (most?) American speakers, the tapped t is neither /d/ nor
[d]; it does not sound like /d/ or [d]; it is not perceived as /d/ or
[d]; and the use of /d/ is widely perceived as just plain wrong. A
few respondents find the suggested use of /d/ to be patronizing or
even offensive: it looks to them like an Englishman's confused idea of
American pronunciation. And, quite apart from bewildering Americans,
such use of /d/ will badly confuse non-Americans about what's going

2. Pairs like 'atom' and 'Adam', while phonetically homophonous for
many speakers in casual speech and for some speakers even in careful
speech, are not phonetically homophonous for everybody. Even in the
most casual speech, some speakers report that they consistently use a
voiceless tap for /t/ but never for /d/, and some speakers report that
they consistently make the familiar distinction of vowel length, with
a longer first vowel in 'Adam' than in 'atom'.

3. Even for speakers for whom 'atom' and 'Adam' are phonetically
identical in very casual speech, they are often not homophonous in
even slightly careful speech. And basing the transcription wholly on
maximally casual styles, with no attention to other styles, is out of

4. The most salient phonetic feature of the tap is not, as the editors
suggest, its voicing -- which need not even be present -- but rather
its shortness.

5. The tapping (flapping) of /t/ is entirely predictable by rule, and
it does not need to be represented in a transcription, since native
speakers will apply the tapping automatically where it is normal.

6. Transcribing, say, 'atom' with /d/ but its derivative 'atomic' with
/t/ will give the impression of an alternation which does not in fact
exist. The same goes for inflected forms like 'write' and 'writing'.
And, of course, even 'write' gets a tapped /t/ when followed by a
vowel, as 'write a letter'.

7. Transcribing both tapped /t/ and tapped /d/ as /d/ will produce
chaos with all those pairs of words which are phonetically distinct
for perhaps all Americans in perhaps all styles, such as
'waiting'/'wading' and 'writer'/'rider' -- unless further steps are
taken to represent the phonetically distinct vocalic nuclei here.

8. Of all the many phonetic details typical of American and other
accents, there is little reason to single out this one for explicit
representation. What about the many other details, such as the nasal
tap used by many Americans for intervocalic /n/, and the interesting
(and varied) American treatment of the cluster /nt/ in words like

9. If explicit representation of the tap is considered to be
essential, then either the IPA fishhook or the established American
stopgap [D] would be preferable to the misleading /d/.

10. Merriam-Webster's Third International dictionary used /d/ for
tapped /t/, and this decision was received with some hostility.

That's about it. My thanks to Matthew Baerman, Donn Bayard, Gordon
Brown, Susan Burt, Karen Chung, Jodie Clark, Douglas Dee, Jeanette
Denton, Ivan Derzhanski, Dorothy Disterheft, Dirk Elzinga, Yehuda
Falk, Paul Fallon, Daniel Faulkner, Antony Green, Charles Gribble,
Jorge Guitart, Carlos Gussenhoven, Rob Hagiwara, Clyde Hankey,
S. J. Hannahs, Kirk Hazen, Robert Hoberman, Mika Hoffman, Alan
Huffman, Jim Jenkins, Roger Lass, Katalin Mady, Mark Mandel, Bart
Mathias, Joyce Milambiling, Corey Miller, Keith Miller, William
Morris, Geoffrey Nathan, Chad Nilep, Marc Picard, Elizabeth Pyatt,
Harold Schiffman, Robert Trammell, Martha Tyrone, Karen van Hoek,
Josefina Vitale, Chris Wallraff, Natasha Warner, Dick Watson, Adam
Werle, Stanley Whitley, Lynell Williams, and Richard Wright.

Larry Trask
University of Sussex
Brighton BN1 9QH

Tel: (01273)-678693 (from UK); +44-1273-678693 (from abroad)
Fax: (01273)-671320 (from UK); +44-1273-671320 (from abroad)
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue