15.2897, Sum: A Definition of Stress

[LINGUIST List]

LINGUIST List: Vol-15-2897. Thu Oct 14 2004. ISSN: 1068 - 4875.

Subject: 15.2897, Sum: A Definition of Stress

Moderators: Anthony Aristar, Wayne State U <aristar at linguistlist.org>
            Helen Aristar-Dry, Eastern Michigan U <hdry at linguistlist.org>

Reviews (reviews at linguistlist.org)
        Sheila Collberg, U of Arizona
        Terry Langendoen, U of Arizona

Homepage: http://linguistlist.org/

The LINGUIST List is funded by Eastern Michigan University, Wayne
State University, and donations from subscribers and publishers.

Editor for this issue: Jessica Boynton <jessica at linguistlist.org>
================================================================

To post to LINGUIST, use our convenient web form at
http://linguistlist.org/LL/posttolinguist.html.


===========================Directory==============================

1)
Date: 02-Oct-2004
From: Vadim Cherny < vacim_lv at center-tv.net >
Subject: A Definition of Stress


	
-------------------------Message 1 ----------------------------------
Date: Thu, 14 Oct 2004 01:08:28
From: Vadim Cherny < vacim_lv at center-tv.net >
Subject: A Definition of Stress

Pertaining to query (Linguist 15.2451).

Perhaps we may formulate the following observations as generally true about accents.
The original accent was a pitch accent. Anthropologically, this is only natural, since animals and
'primitive people' lacking well-defined speech sounds, necessarily relied on variations in the
frequency of the sounds they were producing.

That the same proto-Semitic vowels evolved into different Hebrew vowels depending on the
syllable type also points to pitch variance.

Traces of the former pitch accent may be still evident even in the languages which have practically
lost it in everyday speech. Thus, the Hebrew tradition of semi-singing the Torah may be a rudiment
of a former pitch accent.

Quite probably, the pitch accent included not only increase in the frequency of sounds but also
minor variations in pronunciation. This is illustrated by variations among similar consonants such
as t, th, and teth in classical Hebrew, or pi, pei, and fei in the Tiberian dialect.

Just why languages developed a stress accent remains debatable. Avoiding monotony of speech
sounds, though obvious, is perhaps not the most important reason.

In opera and liturgy, where one is concerned with rationing the flow of air from the lungs in order to
avoid audible heavy gasping for breath, the stress accent often loses its prominence; the same
happens in fast speech.  Perhaps stress is a means of exhaling superfluous air.

Proto-Semitic words possibly followed the pattern CCVC. When a second vowel was added to
make pronunciation easier, turning the word into CVCVC, the interpolated vowel was naturally
weaker than the main vowel. That created a word with one syllable pronounced louder than the
rest.  Later, the habit developed of placing such a stress in every word of more than one syllable.

The stress accent tends to fall toward the end of a word in Semitic languages. It is generally free in
modern languages. This difference may be explained by the Semitic languages lacking
redundancy. Thus, a proto-Hebrew word generally had only a single  unnecessary sound. That is,
the change of a single sound would change the meaning of the word. Therefore, each sound had
to be pronounced correctly. But the sounds after a stress are commonly simplified and lose some
of their distinction, either through a shortage of breath or, more probably, after relaxing the
muscular tension that accompanied the stress. Positioning the stress at the end of a word allows
the best chance for the correct pronunciation of each sound.

Modern languages generally have a high level of redundancy, and hence the simplification of even
several sounds usually allows the listener to recognize the meaning. Thus, nothing prevents the
stress accent from appearing anywhere in the word.

 The stress accent in most modern languages manifests itself in loudness, probably since with the
rise of linguistic redundancy speakers lost their former sensitivity to pitch and became less fluent in
producing and reacting to complicated vocalization.

 Stress based on frequency (pitch) and on amplitude (loudness) both involve prominence of the
vowel (all the commenters agree on this), since both cases require increasing muscular tension.

Amplitudal stress often results in phonetic elongation of the vowel. This is graphically evident,
since increasing the height (that is, amplitude) of the graphic representation of the sound graph
necessarily increases its duration to preserve the visual form that depicts the sound.

There is a somewhat artificial difference between phonetic and phonological length (Steven
McCartney, Bruce Moren). That is, in a similar location within a word a stressed long vowel will be
longer than a stressed short one. I doubt this distinction, and believe it only holds true for vowels
with different pitch. Empirically, both long and short vowels when stressed in the same position
within a word are pronounced with about the same length. The difference supposedly existed in
Old Latin, but it is not clear to me how we could be sure of this, especially since Latin is syllable-
timed, and so the syllable length is more or less fixed.

There is a good argument that short vowels under the amplitudal stress become phonetically long
(Steven McCartney, Bruce Moren). This is not necessarily so with the pitch accent, where a high-
pitch vowel may still remain short, thus saving breath for clear pronunciation of the other sounds.
This may be yet another reason for the ancient preference for the pitch accent.

Stressed vowels may also actually be longer than others without sounding noticeably louder. This
is evident in whispering, where the stressed vowels, though limited in their relative loudness, are
still perceptibly longer.

There seems to be a tendency of the stress accent moving toward beginning of a word. Thus Latin
regularly accents the first syllable of two-syllable words. Likewise, some speakers of Modern
English pronounce ci.g:Ar as cEE.gar making pronunciation easier, although at the expense of
clarity and distinctness of the speech sounds following the stress.

This tendency attracts stress especially to the CV syllables, turning them in effect into CVV, with a
phonetically super-long vowel acting as two vowels. Apparently, such pronunciation is easy and
smooth.

When stress falls on a CVC syllable, it tends to become CCVC which, it may be surmised, is similar
in weight to CVV because one vowel can be approximated in length to two consonants.
Consider a cvccvc word. It can be pronounced thus:
- cVV.ccvc with stress on the first long vowel leading to its phonetic super-elongation; the second
  vowel turns almost into a shwa. This tendency is sometimes obscured by non-aspirated plosives
  (mOt-ley, vs. mO-thley), vowels of highly different strength, and other factors.
- cvc.cvc with no explicit stress, such as in liturgical singing
- c[v].ccVc with stress on the second vowel; the first vowel phonetically shortens.

Thus, syllable is the function of stress, not vice versa.

The tendency of the stress accent to attract more weight into the stressed syllable may be likened
to a piece of dough: the harder one pulls in the middle, the more dough is lifted.

There is also what can be termed 'consonantal stress,' as in the Hebrew mappiq: an especially
clear pronunciation of the last consonant, making it louder. This was apparently important in short
vowelless suffixes.

There is also syntactical stress generally placed on the top of the 'normal' stress, and elongating
the vowel into the super-long variety to impart the questioning intonation: students arrived, vs. had
the students arrIIved?


I would like to thank the participants for offering their opinions on the exact nature of the stress
accent in language (in the order of the letters received):
Steven McCartney
Jorge Gurlekian
Bart van Bezooijen
Martin Weikmann
Justin Barker
Susannah V. Levi
Bruce Moren
Roger Lass
George Aubin
Neil Salmond
Stanley Whitley
Peter T. Daniels

Linguistic Field(s): Historical Linguistics; Phonetics; Phonology



-----------------------------------------------------------
LINGUIST List: Vol-15-2897