18.567 Phonological Interpretation (Acronyms, Markup)

From: Humanist Discussion Group (by way of Willard McCarty willard.mccarty_at_kcl.ac.uk>
Date: Sat, 5 Feb 2005 08:13:52 +0000

               Humanist Discussion Group, Vol. 18, No. 567.
       Centre for Computing in the Humanities, King's College London
                     Submit to: humanist_at_princeton.edu

         Date: Sat, 05 Feb 2005 08:07:07 +0000
         From: Alexandre Enkerli <aenkerli_at_indiana.edu>
         Subject: Phonological Interpretation (Acronyms, Markup)

One way in which "Internet acronyms" are *conceived* differently from the=20
usual acronyms is that they aren't necessarily linked to a phonological=20
representation and, in fact, are often used without direct connection to=20
the string of words their initials originally referred to. In fact, most of=
them aren't lexicalized per se but stand for a broader concept than what=20
the string means. "IMHO" isn't just a way to say "In My Humble Opinion," it=
represents a way to negotiate the value of an argument. At least, these=20
acronyms (along with "emoticons" and abbreviated style) are a significant=20
part of "digital culture" (remember that one?), which revolves around=20
*written* communication.
Yes, the lexicalization of "snafu" is a similar *process* but the type of=20
word "snafu" is seems, to me, to be remarkably different from "IMHO,"=20
"AFAIK," "IIRC," "YMMV," or "IANAL." While "SNAFU" has (AFAIK) a military=20
origin, its use in colloquial speech has little or no military connotation.=
The same might happen with "Internet acronyms" but I, for one, have yet to=
hear some of them *pronounced* in oral communication. In fact, they're=20
phonologically ambivalent as people are unclear whether they should spell=20
them out or pronounce them as words. Even "URL," which is semantically=20
uncomplicated as opposed to the other acronyms, is officially spelled out=20
but often pronounced as a word. Its phonological status isn't clear.

An argument there, challengeable as it is, is that the fact that people now=
write a lot on the Internet is accompanied with a further separation of=20
text and speech. Acronymed, abbreviatied, smilied IM writing styles are=20
tied to their medium. Contrary to many ideas about writing, they're as=20
time-based as speech but not unlike "graphic poetry" and other creative=20
uses of the written domain, they imply elements (e.g. emoticons) which=20
can't really be transmitted into the oral domain.

A similar argument might be used in this distinction between metadata and=20
"character data." Most theories current in linguistics department would=20
stop at the sentence level and give primacy to spoken language. Linguists,=
as opposed to scholars in other "language sciences," limit their subject to=
the well-known levels of analysis from phonology to syntax via morphology=20
(semantics might be left out in some of these theories). In languages using=
writing systems based on the representation of phonemes (i.e. an=20
"alphabet"), the non-metadata content *could* be defined as those=20
characters which contribute to a lexeme's pronunciation. This goes with the=
perspective that the primary content is the one which can be analyzed=20
through sentence-level grammars. A challengeable perspective but one which=
certainly had an influence on computer science.
The IPA has no capital letters or punctuation. It only represents phones=20
(though it's used to represent phonemes) which have, by themselves, no=20
metadata value. A minimal writing system for computer-mediated=20
communication could have IPA characters for phonemes (more significant than=
phones) and other characters (punctuation, numbers, whitespace...) for=20
metadata. The distinction is artificial in the abstract but couldn't it=20
work concretely?
The advantages of distinguishing metadata from "lexical content" are in=20
terms of computer processing.

As a summary of responses to Dino Buzzetti, diacritics have an impact on=20
phonological value ('=E9' or "ai" for /e/, '=E7' or "s" for /s/) while a=
originally had a prosodic value and may now have a purely structural value.=
Simplistic? Possibly. But perhaps simplification is in order when trying to=
redefine text through its computer representations.

Sincerely yours,


Alex Enkerli, Teaching Fellow, Visiting Lecturer
Department of Sociology and Anthropology, Indiana University South Bend, DW=
1700 Mishawaka Ave., South Bend, IN 46634-7111
Office: (574)520-4102
Fax: (574)520-5031 (to: Enkerli, Anthropology)
Received on Sat Feb 05 2005 - 03:22:01 EST

This archive was generated by hypermail 2.2.0 : Sat Feb 05 2005 - 03:22:02 EST