Route: spelling and pronunciation [and IPA vs. Orthographic respellings]

Dennis R. Preston preston at PILOT.MSU.EDU
Sun Jun 30 13:23:04 UTC 2002

I am amazed at this very basic linguistic confusion.

Spelling (usually phonemic - and even morphophonemic, as SPE taught
us - rather than phonetic) is clearly abstract, if by that we mean
"divorced from the acoustic reality." A phonetic representation is by
contrast concrete (whether it is IPA or the broad phonetic
representation represented by respellings, e.g., "rawt" to indicate
how some folk pronounce "route" - pronounced /route/ by nobody except
those who read English but do not known it, although I had a friend
in grad school who once pronounced "creature" /kreature).

Of course almost all spelling systems have their their roots (sorry!)
in "sound," but they almost certainly have their roots in phonemic
(abstract) representations, not representations which represent
acoustic-articulatory reality, the goal of phonetic representations.

There is indeed a contrast here (and I do not want o suggest that the
mental representations of phonemes are not "concrete" in some CogSci
way), but surely the words abstract and concrete are reversed here so
far as their ordinary meanings are concerned.

dInIs (pointing to the "spelling" of his name)

>The point Rudy is trying to make seems to me a question of abstract vs.
>IPA is abstract, which is as it should be since it is supposed to
>represent the sounds obtaining in most human languages. Abstract
>representations call for the use of a key in order to make the
>transition from real to unreal (or abstract) and vice versa. It serves
>purposes all right.
>It seems to me it is also somewhat self-serving (and intellectually
>stimulating for
>what it is worth) because it has a make-work quality to it. It supports
>teachers, textbooks, etc.; sorry if I sound like a patron! That is why I
>lovingly refer to the COBUILD
>dictionaries as "the collected teachings of the lexicographers of
>Birmingham." So much time and space seems wasted.
>On the other hand, the popular "phonetic" transcriptions based on
>English orthography are "concrete" and represent the real world. They
>have no need for a key. Having to use a key is like hiring an
>interpreter who is not really needed. All that is needed is reading
>knowledge at the elementary (vaguely understood as Grades 6 - 8) level,
>if we can
>forget for the moment about illiterate undergrads and such. It is a
>process of
>going from the known to the unknown, but not starting from zero
>knowledge. IPA use involves learning of a higher order, but the question
>is, Is it cost-effective for general (not scholarly) purposes? A side
>issue is, no amount of IPA can produce
>the real sounds of a language, only an abstract representation of them.
>Actualization of sounds as heard is still up to the user, given the
>imperfections of one's personal sound equipment. Sometimes rough
>justice, as in popular phonetics, is better than justice in the
>Re "root" and "rowt" to represent <route>, I see no problem if you
>consider the realities of the situation, namely, sounds in context.
>Words like <bellow, blow, flow, hallow, show> normally occur at the end
>of words.  If the normal pronunciation can be accepted as the regular
>pattern of English orthography, then <show> and other words become the
>exception. (Add also words like <acknowledge>) These, in my view, should
>be dealt with separately and represented as (SHOH),(ak.NOL.ij) etc.
>which are phonemically quite exact. The syllable "ak," for example,
>cannot be pronounced by a sixth grader as anything but the IPA ae
>digraph + k.
>Being unambiguous and consistent is important, but at what price, is the
>question for me.
>Just a few private thoughts.
>(lexicographer who used IPA in the Sixties, IPA-cum-diacritics in the
>Seventies, and now uses a popular phonetic system; cf.
> - still in preparation, the
>phonetics awaiting insertion)
>P.S. I am still casting about for printable symbols that are compatible
>WORD or the actual processing software (Notepad?) used for text
>input in website creation. Maybe there isn't any. Cf.
>"lek-s&-'k”-gr&-fE" (M-W's "lexicography").
>Scanning in of the text is one alternative,
>I suppose). I have been using WordPerfect since before WORD came on the
>scene, which is how I seem to be handicapped. Help requested.
>Rudolph C Troike wrote:
>>  One of the problems in trying to use English orthographic re-spellings to
>>  indicate pronunciation is that regional pronunciations vary considerably,
>>  and of course there is considerable difference cross-linguistically in the
>>  pronunciation values attached to particular letters or combinations. This
>>  is why the International Phonetic Association was formed in the 19th c.,
>>  and why linguists use more-or-less consistent transcriptions to represent
>>  pronunciations. Americans for over 150 years have been shielded by
>>  dictionary-makers from confronting and learning a consistent phonemic
>>  representation, but British dictionaries have for some time utilized a
>>  version of the IPA system, and presumably British school-children learn
>>  this system, and know how to interpret it in using a dictionary. When I
>>  taught in Taiwan 30 years ago, I found that even 8th graders who were
>>  studying English as a foreign language, and struggling with the very
>>  concept of an alphabetic system for writing the language, were also
>>  expected to learn a "broad" IPA phonemic transcription at the same time.
>>  American innocence in this area, even after over 50 years of efforts by
>>  linguists, remains regrettable, and a reflection on our educational
>>  system.
>>          Representing the different pronunciations of <route> as "root" and
>>  "rowt" runs afoul of the problem that the word <root> can be pronounced to
>>  rhyme with <foot> or <boot>, and the respelling "rowt" can be interpreted
>>  as reflecting the vowel of <rote> or of <rout>. A phonemic transcription
>>  of /ruwt/ (or /ru:t/, as some might prefer) vs /rawt/ (or again, some
>>  might aesthetically prefer /raUt/) has the value of being unambiguous and
>>  consistent (the variant transcriptions are equally clear and consistent).
>>          Rudy

Dennis R. Preston
Professor of Linguistics
Department of Linguistics and Languages
740 Wells Hall A
Michigan State University
East Lansing, MI 48824-1027 USA
Office - (517) 353-0740
Fax - (517) 432-2736

More information about the Ads-l mailing list