"Stress" (Was: kalabaw)

[Olli Salmi]

At 16:50 +0200 3.12.2000, Waruno Mahdi wrote:
>Incidentally, nuclear Malay typically has stress on the penultimate,
>except when the penultimate vowel is schwa (as in _kerbau_ /k at rbáw/), in
>which case the stress is on the ultimate (I won't go into technicalities
>on what kind of "stress" it is in Malay).

I wish you had gone into technicalities. I have tried to find out how Malay
is stressed but I lost hope because books do not agree with what I hear.
I'll give here a rough sketch of how I hear phrasal stress in Malaysian
Malay. I'd be very happy for any technicalities, and also for
non-technicalities or references. I'm sorry this is a long posting.

Last year I listened to the Linguaphone Kursus Bahasa Malaysia. To my
surprise I quite often heard the stress on the last syllable. After
listening to the tapes with the help of Praat, I came to the conclusion
that it's basically stress on the last syllable sentence medially, and
penultimate stress sentence finally (stress=f0 peak).

A few months ago I got hold of Halim's "Intonation in relation to syntax in
Indonesian" and he seems to have come to the same conclusion (South
Sumatran Indonesian). I also bought "Colloquial Malay" and I've listened to
the tapes, again with the help of Praat. It seems that Halim's system is
not the whole truth and some additions have to be made.

1) Halim's stress is on the pitch peak. On the tapes that I have been
listening, it is often difficult to tell where the peak is, because it is
somewhere just before or just after a consonant. It is better to say that
in such cases the stress is on the vowel with the rise. This way you get
more regular patterns than if you use the pitch peak.

2) According to Halim, the pause between pause groups (intermediate or
intonational phrases) need not be realized. Pauses are easy to notice with
Praat and I think we should make a difference between actual and realized
pauses.

For Halim the intonation of a sentence is like this:

__/, __/\  or __/, __/, __/\

In my experience, topics are united into groups:

_/_____/, _____/\

The low syllables (_) start from high, where the previous syllable ends,
and descend gradually, which I cannot reproduce with the underscore
character.

It seems that in longer sentences the phrase (limited by actual pauses)
usually has two stresses, one near the beginning of the phrase and one at
the end. Sometimes the rise at the end is noticeably lower than the first
peak.

Final syllables can have /\ crammed into one syllable. This is emphasis,
according to Halim, or a pepet in the penultimate (although it seems that
the pepet is often high and short, as if the normal rise didn't fit on it).
Some speakers seem to stress final syllables rather routinely, often
lengthening the preceding vowel and sometimes even the consonant (Hendon's
clarity accent?).

Strangely enough, questions have the pattern

/\________/\

In other words, question words at the beginning of the sentence have
penultimate stress.

3) The most interesting feature is that final stress often occurs on the
first part of a compound:

Kamí juga ingin lihat Istana ágong.  (acute accent means high (rising)
pitch="stress", / means rising)
Istaná Agong/ agak jaúh dari síni.

This is probably not true for every non-final compound. This is a case of
two stresses in a pause group which I mentioned above.

Similar stress can occur on prepositions:
Darí sini/
Dalám darjah dúa| ada tiga pulúh orang murid. (| level or falling, not sure)
untúk mengetahui [menge-tahui downstep] berita| pada pági itu\ (\ falling end)

>>From the Linguaphone course I've noted down the following cases:
anák saya  (but also _ának sáya_)
kebanyakán kami
di satú sudut
marí kita bercakap...
ramái juga
kamí juga
hanya sedikít sahaja (very common with _sahaja_)
In compounds:
petínyanyi
pirínghitam
tingkát bawah
anák tangga
waktú siang
waktú malam
oráng gaji kamí/
sátu piring lagí/
Reduplicated nouns:
lagú-lagu

kitá nampak (subject verb)
berítahu (verb verb)
biár saya kira dahúlu (verb object)
Saya suká kuih suji.
Boléh tuan...
Kámi/ jaráng pergi ke majlis tari-menári. (Kámi/ sounds like a fall rise)

4) A further qualification to Halim is that his contour 211 f#, low
falling, can occur after words with final stress.

dan saudara akan nampák Tasik Perdana\
Anak-anak perempuan Aní nakál tak macam Ani dulu\

Sometimes a whole sentence can have this contour:

Máaf\ saya tak boleh guna kad kredit saya lagi\
H Sungguh ba:nyak duit anda L\ (H high, L low, probably redundant, this
speaker likes to stress the final syllables and to geminate intervocalic
consonants)

5) Another thing is that with Halim all syllables before the stressed one
have pitch level 2, by definition. However, there seem to be cases of a
high beginning:

Túgú Négará di tengáh kawasán Tasik Perdaná\

or upsteps to high
untuk bersámá káwán-káwánnya\

It is possible that the pitch of unstressed syllables ascends or descends
word by word or foot by foot, not syllable by syllable. The same may be the
case with the 211 f# contour. This may give an impression of penultimate
stress.

If Cohn's feet have any reality, they are small ripples that can sometimes
be seen on the baseline.

I reached Lesson 9 in Colloquial Malay before giving up. It seems that most
cases of non-final penultimate stress are some verbs, especially verbs that
end in -kan.

Earlier I believed that the default stresses are a manifestation of
intonation and the non-final stress on the penultimate syllable are
emphatic stresses (sort of like in French), but that is probably not
correct.

Indonesian accents seem to vary a lot, but, although they are rare,
examples of most of the above features can be found on the North Illinois
University site at http://www.seasite.niu.edu/Indonesian. Unfortunately
they are in RealAudio format, which cannot be read by Praat. Narrative
texts seem to differ from Malaysian more than dialogues. There are passages
(like the poems) which have only penultimate stresses with falling boundary
intonation, if my ear can be trusted.

Recently I bought Ewald Ebing's "Form and Function of pitch movements in
Indonesian. It seems that Indonesian speakers accept almost any kind of
stress. I find this very strange. Van Zanten & van Heuven's paper
"Position, Shape and Acceptability of Accent-Lending Pitch Movements in
Indonesian" shows the same thing. Does it mean that I can use my Finnish
initial stress and nobody will notice anything? Or is it a question of
phrase stress vs. word stress? I don't think word stress has much meaning
in Malay.

All the best,
Olli Salmi