Re: [Pali] Accent in Pali
>Stress/emphasis - where it is placed in Pali words?In the old grammars rules governing stress are only given for verse, not
for prose. The main source is the Vuttodaya by San.gharakkhita, but the
rules are too long for me to summarize in an e-mail. I suggest that you
consult one of the following:
Vuttodaya, A Study of Pali Metre. R. Siddharatha. Calcutta 1929 (there is
also a very cheap reprint by Sri Satguru Publications, New Delhi, 1981).
Pali Metre. A.K. Warder (PTS 1967)
If you read Russian then look out for a copy of "Yazyk Pali" by Tatiana Y.
Elizarenkova & Vladimir N. Toporov (USSR Academy of Sciences, Moscow 1976).
Its chapters on phonemics and versification cover about the same ground as
Warder's book for a fraction of the cost.
As for the accent in prose passages, here are the views of two modern scholars:
"Regarding accent it is difficult to state anything definite about its
phonological significance on the basis of surviving Pali texts because they
are not written in scripts that show accents. We can only confine ourselves
to informed hypotheses established chiefly on diachronic data. On this
basis Pali can reasonably be viewed as having lost the tonal system that
was characteristic of Old Indian, but retaining the original placing of the
accent. This (along with the expiratory character of Pali stress) is often
indicated by changes in the vocalism of the unstressed (from the point of
view of Old Indian) syllables in Pali."
(Yazyk Pali, Elizarenkova & Toporov)
"Nothing has been handed down to us about the nature of the Pali accent. It
is, however, improbable, that the ancient Indian accent was still in force.
It is more likely that, as Jakobi has suggested for Prakrit, the Sanskritic
accent was the rule in Pali. This is suggested by the change in vocalism in
Pali, such as the weakening of a vowel after the accented syllable (in the
Skt form) or its strengthening in the main accented syllable."
(Pali Grammar, Wilhelm Geiger)
If Geiger is correct that "the Sanskritic accent was the rule in Pali",
then stress should be placed on:
1) Long penultimates:
2) Ante-penultimates (whether long or short) followed by a short syllable:
3) The fourth syllable from the end (whether long or short) when two short
Of course there are many exceptions to these rules, so to get the full
picture you would need to consult a Sanskrit grammar or one of the online
As for the modern pronunciation of Pali (i.e. by Thai, Sinhalese and
Burmese scholars when reading a Pali passage) there isn't any
universally accepted convention governing stress. The Sinhalese, so I'm
told, place the stress according to the same rules that govern Sanskrit, as
I have given them above.
Thais, on the other hand, when reading a Pali text will intone the words
according to the same tonal rules that govern the Thai language, but place
the stress wherever they feel like it. But when chanting Pali, both tone
and stress will depend on which style of chanting they are following; there
are several of them in use. As for the Burmese, I've no idea what they do.
- Dear Venerables; Hi Group;
WARNING: Just some "untraditional thinking" from one guy who is *not* an
expert in Pali and *not* and expert in languages!
I had this same question about stress and emphasis when I started my Pali
studies. I am not satisfied by the current thinking about Pali
pronunciation, so I did a mini-analysis. I have not yet taken the time to
formalize my thinking in a systematic way. Here are a few notes and a few
questions I've had for some time.
a) Pali would be very easy for any human being to pronounce. A Buddha would
not teach in a language which was hard for any human being to speak
"out-loud". He knew that the teaching would move to different languages and
cultures. 2,500 years ago, most people were illiterate and they used
speaking for teaching and learning. Pali would have to be able to move
easily with the teaching.
b) Whoever created the romanized Pali transcription system would have
included *all* of the information necessary to pronounce the Pali word
correctly (including stress and emphasis).
a) the underdot m that is so commonly used at the end of words
b) Pali word order
c) the use of the letter "h"
d) the doubling of hard and soft consonants.
e) word inflection redundancy
I'm not really an "expert" in languages. I only speak English (various
centuries), French, German (a couple of centuries), Italian and Spanish. I
know some Pali, and at various times over the years I've studied a tiny tad
of Japanese, Chinese, Korean and Thai. Still, I think it would be fair to
say that I do have a good "overview" of the construction and pronunciation
of human languages.
I also enjoy composing music. Basically, it is my (limited) knowledge of
formal musical notation that is the root of my thinking about Pali
In the context of figuring out Pali pronunciation I would say this:
a) The easiest and most fun language to speak that I know is Italian. The
*sounds* are very easy to pronounce and they fit easily together.
b) German is interesting because of word order. Since the word order is very
formal and verbs (in anything more than a simple sentence) will occur at the
end of a sentence, a person needs to speak slowly and very clearly. The
listener needs time to keep track of all the bits and pieces of the
sentence. Then, at the *end* of a sentence, the listener will also need a
short moment to "put it all together and think about it".
c) Most Asian languages are interesting because they are tonal, making them
"harder" to pronounce.
d) word inflection redundancy is typical of very old languages - where
people spoke many regional dialects; most people were illiterate; books were
expensive luxuries; and teaching was done by speaking (not reading). The
inflectional redundancy would help people figure out what the speaker was
trying to say (Example: the inflection of the adjective should agree with
the inflection of the noun. If the speaker messed this up, there would be
enough information in the sentence for the listener to figure out what the
speaker was trying to say.)
My Current Working Hypothesis:
a) The overdot and underdot marks are *not* used to change the pronunciation
of the character. They are used as stress and emphasis marks. A specific
case: the underdot-m (.m) at the end of a word. I see no need for this to
have the rather annoying and difficult "ng" sound - again and again and
again - in a sentence. I think that this is simply means "pronounce m - and
emphasize it a bit and wait a split-second to make it easy for the listener
to note that this is the end of the word before you roll into the next word.
The listener needs a split-second to become aware of the inflected ending,
figure out what it might imply, and mentally note it for use when the verb
finally shows up later in the sentence."
b) In English, we have long vowels and short vowels. We also have aspirated
vowels (like "happy"). The "h" character is used in Pali to create an
aspirated vowel anywhere in a word. Again, this aspirated vowel is used more
as a mechanical device so that the pronunciation of the word has correct
*timing*, stress and emphasis. I think it is more
c) The doubling of consonants. Obviously, the doubling of a hard consonant
is a "timing and emphasis" mark. You don't "pronounce it twice". (Example:
dukkha. You don't say duK-Kha. You say doo-Kha.) Which means that romanized
Pali is *not* perfectly phonetic, where every character is pronounced. The
doubling of soft consonants (Example: dhamma.) is also a stress and emphasis
To create a fictious example: duka, dukka, dukha, dukkha - according to my
current theory, where's the timing, stress and emphasis? du-ka, du-Ka,
Another fictious example: dama, damma, dhama, dhamma - da-ma, da-Ma, dHa-ma,
Another fictious example: rata, ratta, ratha, rattha / ra.ta, ra.t.t.a,
ra.tha, ra.t.tha - ra-ta, ra-Ta, ra-tHa, ra-Tha / rat-a, RAT-a, rat-Ha,
Putting it all together (another fictious example): dama.m, damma.m,
dhama.m, dhamma.m - da-mam da-Mam, dHa-mam, dHa-Mam.
The overdot n ("n) works to tie the n to the following consonant and avoid
triple consonants. A triple consonant would make it impossible to tell where
the emphasis went. Example: sa"ngha. This is pronounced sang-gHa. If it was
written sanggha we could have san-Gha as our pronunciation. To confirm this,
I used my software to looked all of the words using the "n character in the
Paliwords dictionary. There are 587 basewords that use "n. None of them have
triple consonants. (ie "ngg, "nbb, "nkk, etc.)
Fictious exampels: sangha, sa.ngha, sa"ngha - san-gHa, SAN-gHa, san(g)-gHa
I think the only character that has a different "sound" as a result of an
inflection mark is ~n, where I think the "nyuh" sound commonly used is
I do not think that Pali uses "tonality" at all.
a) What was the nationality of the person who created romanized Pali?
b) What year did that person create it?
c) What form of English (British, American, India) did they know?
WARNING AGAIN: Just some "untraditional thinking" from one guy who is *not*
an expert in Pali and *not* and expert in languages!
thanks and peace from
From: Дмитрий Алексеевич Ивахненко (Dimitry A. Ivakhnenko)
Sent: Sunday, June 09, 2002 9:20 PM
To: Robert Eddison
Subject: Re: [Pali] Accent in Pali
RE> What do you mean by accent? Are you referring to pronunciation or
RE> to stress/emphasis?
Stress/emphasis - where it is placed in Pali words?
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Yahoo! Groups members can set their delivery options to daily digest or web
[Send Message] firstname.lastname@example.org
[Mailing List] http://groups.yahoo.com/group/pali
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
- Dear All,
Here comes another non-expert to share what he learnt from others and thought out himself.
1. The niggahita (overdot/underdot m)
This something that I've given quite some thought about. I was at first quite puzzled as to the way the Sri Lankans render it. There's absolutely no difference in pronunciation between the overdot/underdot m and the overdot n. They both sound like 'ng' as in 'sing'. The Burmese on the other hand pronounces it just as 'm'.
Now this doesn't make sense to me. As in the way languages evolve, the spoken word comes before the written. So, if "m and "n (or "m and m) are pronounced the same, how did one sound split into two characters?
A couple of years ago, I have the good chance of meeting a Burmese monk studying in India. Upon my asking, he told me that it is pronounced as a nasalized m. Like .n, it's nasalized, but the lips are closed at the end.
It seemed very reasonable to me. Furthermore, questions that I have about the peculiarity of sandhis involving the niggahita begin to fall into place. I'll not illustrate them. Rather, I'd like to invite members of the group to take this as a hypothesis and test it in your reading of the Pali texts, bearing in mind always that Pali was more of a spoken language.
A related issue is whether the niggahita is better romanized as an overdot or underdot m. Considering that it is nasalised and not retroflexed (like .t, .d, .n), I personally prefer the dot on top for the sake of consistency.
2. Vocal emphasis
I do not know how the emphasis should be done technically. However, having chanted verses with my teacher, keeping what should be chanted long long, and short short, the idea of how to put the stresses at the right syllables seems to fall into place by itself. It's easier to get a feel of it through chanting verses, whereby the words are specially arranged so that there's a particular pattern as to the long and short syllables, producing a consistent rhythm throughout the verse. (A side-effect of such a special arrangement is that poetic license is liberally used, making Pali verses hard to understand and translate.)
3. Pali Pronunciation
There's obviously such a thing as long and short syllables. E.g.:
bala means strength; power; force
baala means young in years; ignorant; foolish; child; fool.
baalaa means girl.
Another example: vata & vatta, which are two very different words.
For dukkha, you could say that the 'k' sound occurs twice, or that theres a pause between the two 'k's. So, it's pronounced as duk-kha, as not doo-kha as I often here from native English-speakers.
Many native English-speakers also say Boo-dha. Any Sri Lankan can tell you that's wrong, and that the right way to pronounce it is Bud-dha.
While the Buddha must have known that the teaching would move to different cultures, it wouldn't make sense for him to modify current speech suit everybody, everywhere, then and now. The people then would have trouble understanding him.
It's common for people of different cultures have difficulty producing certain sounds of a foreign language. What may be difficult for someone to pronounce may not be so with another. Many Chinese learning English as a foreign language have a tough time trying to pronounce "the" correctly, simply because the "th" sound is not within their vocal scheme.
At 04:13 AM 11-06-02, Andy wrote:
>Questions:I think they were English. They later on established the Pali Texts Society.
>a) What was the nationality of the person who created romanized Pali?
>b) What year did that person create it?Probably in the 19th century.
>c) What form of English (British, American, India) did they know?British.