Why short "a" sounds like "er"

Let us just replace all unstressed short a in the Namo Tasso Bhagavato recitation with the schwa character:

Namo Tassə Bhəgəvato Ərəhato Səmmāsambuddhəssə

My personal impression was that the auslaut -ə sounds quite different to the anlaut ə- in Arahato and inlaut -ə- in Bhəgəvato, in fact very different. Moreover, I don’t really think you pronounce the word ‘later’ as the Venerables pronounce ‘tassa’ in the video: schwa has more of the German ‘a’ quality than ‘ä’ quality.

I guess, purely theoretically you should classify the auslaut -ə (sounding actually like a mixture of /æ/ and /ɨ/) as an allophone (positional variant) of the phoneme <ə> used at the end of a word. However, leaving the theory aside, there is a considerable variation between allophones. The American intervocal t as in water actually is pretty much identical to the Slavic or Southern German , so that the famous ‘She’s got it’ from Venus by Shocking Blue used to be pronounced as ‘shizgara’ by Soviet fans who didn’t speak English. In other words, for all intents and purposes except theoretical phonetics the -ə in the Sinhala Pali tradition can be treated as an approximation of /æ/ and /ɨ/.

Thanks for the interesting clarifications! Just to add, while in general the Sinhala pronunciation of Pali is good, and certainly better than in any other traditional culture, it’s not perfect.

I’m no expert in Sinhala, but in addition to the issues cited, my understanding is that they omit the retroflex consonants (ḍ, ṭ, ṇ, ḷ), replacing them with standard unvoiced consonants (t, d, n, l). Meanwhile, the unvoiced consonants are spelled the same as the aspirated consonants, so my name becomes “Sujatho”.

This is merely a spelling convention, as there aren’t any true aspirated consonants in Sinhala.

In any case, this explains why we frequently see what looks to our eyes like eccentric spelling of Pali, such as sathipatthana, etc. I’ve found that some Sri Lankans—and I’m sure this applies to Thais and Burmese as well—don’t realize that there is a universally accepted global spelling convention for Pali. They think the way we spell it is the “English” way!

As a final detail. When I first heard Pali I was amazed, and could not believe that such an old language could be remembered at all. I imagined that the way it was pronounced would be just arbitrary. But I didn’t understand the detail and complexity of the ancient Indian science of linguistics.

Starting, so far as we have access to, with the work of Panini, this is one of the worlds great scientific achievements. From Wikipedia:

Pāṇini’s grammar is the world’s first formal system, developed well before the 19th century innovations of Gottlob Frege and the subsequent development of mathematical logic. In designing his grammar, Pāṇini used the method of “auxiliary symbols”, in which new affixes are designated to mark syntactic categories and the control of grammatical derivations. This technique, rediscovered by the logician Emil Post, became a standard method in the design of computer programming languages.

For more on Panini, see this article. encyclopaedia001.pdf (3.9 MB)

We know from our studies of the suttas that such linguistic sophistication predated the Buddha. In fact it developed for exactly the same reason as we are discussing now. The brahmins had ancient sacred scriptures, the Vedas, but they were so old that the language was archaic and incomprehensible. So they developed incredibly advanced linguistic analysis to maintain these ancient texts.

One aspect of this was pronunciation, which was considered to be especially important since the texts were performed orally. (We don’t know when brahmins adopted writing for the Vedas, but it was certainly long after the Buddhists.) Every syllable was considered sacred, a literal emanation from the cosmic divinity, and the slightest mistake would betray the brahmins’ sacred duty and risk upsetting the cosmic order.

Thus not only the pronunciation was defined and described, but the method of education in correct pronunciation, known simply as shiksha (Pali: sikkhā, training, education), was developed in exquisite detail.

The level of detail was carried over to the writing systems as well, which, throughout the Indic sphere, are perfectly phonetic.

These methods were adopted by Pali linguists, so we know with certainty exactly how Pali is meant to be pronounced. This is not in dispute, and Pali experts and linguists in the different Theravada countries all agree on this. The issue is that due to local influences, incorrect spellings and pronunciations become widespread in the culture.

Incidentally, the driving forces behind the Mahasangiti Pali edition we use on SuttaCentral, the so-called Dhamma Society, were well aware of this issue and wanted to reform Thai use of Pali. This is why they published their text in Roman script only, to try to get the Thais to use correct pronunciation. Oh, well.

6 Likes

I thought I’ve asked a stupid question, it’s not so stupid after all :stuck_out_tongue: In fact, I’m overwhelmed by the theories.

So in that case dhamma doesn’t sound like dhammer? I don’t think dhamma is pronounced as dhammer if I understand Bhikkhu Bodhi’s pronunciation correctly.

1 Like

This is the Pali pronunciation guide in our daily chanting book compiled by Bhante G here at Bhavana. I’ve never heard of an “a” sounding like an er… perhaps this is a Thai thing(as I’ve only ever heard thai western monks say stuff like “dhammer” and particularly only english/australian)? I figured it was just their way of saying the word, not an actual Pali pronunciation they may of learned, interesting.

I learned dhamma as “dhummuh” basically, as a = u in hut, and of course the aspirated h after d. I know this is also how Bhante Sujato says it, because I’ve heard him use it. He is one of the few monastics who really aspirates. I was lucky enough to have a monastic living here for a year who was extremely “anal” about Pali pronunciation and learning the language properly in the first place, so It’s been drilled into me.

##Alphabet

As Pāli does not have its own script, existing alphabets have been adapted to represent Pāli and pronounce it phonetically. We use a script called “Romanized Pāli,” which consists of 41 letters.

The vowels are pronounced as follows:

  • a is like u in hut or us
  • ā is like a in father or barn
  • i is like i in bit or pin
  • ī is like ee in beet or tree
  • u is like u in put or foot
  • ū is like oo in pool or boot
  • e is like a in bake or ache
  • o is like o in hole or bone

When followed by two consonants, the pronunciation of e and o changes as follows:

  • e is like e in met or rest
  • o is like ou in ought

Among the consonants note the following special rules:

  • g is hard, as in gone
  • ṃ is like ng in sing
  • c is like ch in church
  • j is like j in joy
  • ñ is like ny in canyon

The “dental” letters t, th, d, dh, n (pronounced with the tip of the tongue against the teeth), and the “retroflex” letters ṭ, ṭh, ḍ, ḍh, ṇ (pronounced with the tip of the tongue curled back against the palate, giving a hollow sound).

Consonants followed by an h are “aspirates” (they are: kh, gh, ch, jh, th, dh, ph, bh, ṭh, and ḍh.)

Unlike English consonants followed by an h they are pronounced like the first consonant alone followed by a forceful out-breath. English does not offer these sounds. For example, the English p and h in “top-hat” come close to the Pāli ph if the h of “hat” is spoken forcefully; however, to pronounce these aspirates correctly you must first hear them from someone who knows Pāli. The non-aspirate consonants are pronounced with a much softer breath-pulse than any English consonant.

5 Likes

that’s basically sums up the endeavor of learning pronunciation as far as i’m concerned

i don’t even try to learn pronunciation through transcription and explanations, that’s a lost cause, it’s almost the same as with music, because both deal with sound, until you hear how a note rings it’s impossible to interpret its sign on the staff and reproduce it vocally (pitch perfection provided)

1 Like

@LXNDR
I’ve certainly found this as well in studying Pali on my own. Unfortunately I don’t live near a monastic community or anyone who knows Pali. So even though I try to follow the pronunciation guide as shared by @jayantha, it’s really difficult to know what it sounds like. let alone learn to pronounce it correctly, without actually hearing it.

For anyone interested, some sources I’ve found helpful in terms of getting a feeling for how it sounds are:
Ven Jiv’s chants
Ven Brahmali’s Pali class recordings
Ven Nananada’s Nibbana lecture recordings (when quoting the suttas, which he does extensively, he always recites the Pali followed by his own transaltions)

2 Likes

I recently recorded myself chanting our morning Puja chant, this may be helpful as well - https://archive.org/details/BhavanaSocietyDailyChantingByJayantha ( you can download the free pdf of the chanting book from the link in the description.

also I run a little sangha on a voice and text program called Discord where we go over the suttas weekly, I’d be happy to get on voice with anyone here sometime to help with pronunciations

2 Likes

Dear Ven,

Thank you so much for this. I just downloaded the chant. I don’t know what Discord is, can you explain more?

1 Like

In my personal experience, the most efficient way is to combine both. Learning the English or German pronunciation can be really tricky unless someone explains you all the rare positional variants. Just to give you a small example: in the English ‘in the’ the n sound before the interdental th is pronounced with the tongue on the back of your teeth, not at the alveoli, something that you could hardly figure out on your own.

1 Like

That’s interesting. To my ear, my local Thai monks chant:
Tas-sa (short a at the end)
Sri Lankans tend to sound like:
Tas-ser
Here’s a Sri Lankan example:

This is a Thai pronunciation:

I assume that this is a Burmese monk:

1 Like

@mikenz66 actually in all tree of those examples I don’t hear an ER (like tasser), the first two seem to have pretty good pronunciation, the last one he pronounced the tassa as tassā, which for years is how I thought it was.

in all honestly when I say I’ve heard people say like " Buddher", it hasn’t been in chanting, but in reading and speaking.

Until it was pointed out to me, I had no idea people were saying A as U in their chanting, It all just sounded like A, now that I can hear the difference and applied it to my own chanting I can hear it in others.

I kind of view it the same as a photographer. I had a business in lay life and when you become a photographer you develop an eye for a photograph, and you look at the photo and it looks horrid, you see all the things you don’t like about it, but the average non photographer loves it, two different perceptions. You develop an ear for what really is fairly minute differences, and you can tell.

Another example and this is one even the most knowledgeable pali chanters do, simply because it’s just so hard to do the i at the end of a word right,especially in chanting.:

Dutiampi - dutehumpeh not duteeampee (dutīampī)

ī = ee sound
i - like in pit, hit, knit, etc.

@Linda

Thank you so much for this. I just downloaded the chant. I don’t know what Discord is, can you explain more?

Discord is a program for online communities using voice and text. It’s wonderful for getting people together as an online community, especially when it’s time to talk together via voice. Follow this link and it will go directly to the channel " students of the path" - Discord

Isn’t it actually ‘dutiyampi’ with a ‘y’?

Here’s an American speaking :slight_smile: In the vast majorities of the languages using the Latin script A is pronounced as U. Anyway, your own experience is invaluable in helping other Americans chant correctly.

It is because they pronounce it as /ˈbʊdə/, whereas the Sinhala chanters use a sound I would describe - for lack of a better symbol - with /æ/ (sounding more like something between /æ/ and /ɨ/ - and I checkedbooks on the Sinhala phonetics, it is a legit sound), so ‘Buddha’ sounds kinda like /ˈbʊdæ/ and ‘Namo Tassa Bhagavato Arahato Sammasambuddhasa’ as /namo tasæ bhəɡəvato arəhato samːasambudhːasæ/

1 Like

@Vstakan > Dutiampi - dutehumpeh not duteeampee (dutīampī)

Isn’t it actually ‘dutiyampi’ with a ‘y’?

yes you got me! lol, thats what I get for going off memory ;). Still doesn’t change the example though, but thanks for the correction.

1 Like

My best friend was more than 20 years old, when he found out that his big sister couldn’t pronounce the letter R correctly. It just came up one night during a discussion among friends and he was like “Wait what? No she doesn’t.” And he had been living with her his whole life.

So sometimes our minds ignore even pretty big things, if we’re used to them.

Thanks Ven for the guide. I have edited the formatting for clarity, I hope this is okay.

May I suggest a couple of changes?

  1. “o is like ou in ought”: Change this to “o in got”. “Ought” is, for me anyway, a different sound. It’s IPA phoneme /ɔː/, whereas the correct sound is /ɒ/. This, I have just learned, is known in linguistics as the “cot/caught merger”.
  2. I’d also change the example “church” for “c”. If I hold my hand in front of my mouth when saying “church”, the initial ch is more heavily aspirated than the final. This is only inviting unnecessary confusion.
  3. Also, since “ch” is almost always aspirated in English, it should be used as an example of “ch” not “c”. A better example of “c” in English is the “ch” in “mischievous”.
  4. While we’re at it, “barn” is another misleading example. The “r” is pronounced in many varieties of American English, so the pronunciation of father and barn are quite distinct.

Also, the description of aspirates and non-aspirates is not quite correct. English normally aspirates hard consonants (stops). To be sure, we don’t usually aspirate them as heavily as Pali, because they are not so differentiated. However this distinction is so subtle that it is ignored in the IPA phonetics system. A chart might clear it up:

English |       p   |
Pali    |  p  |  ph |

So the English “p” is the same phoneme as Pali “ph”, but it falls somewhat towards “p” in Pali.

For those who don’t know what we’re talking about, here’s how to learn the aspirations. The first step is to hold your hand in front of your mouth and feel the breath. You can clearly feel a puff of air after the p in “pill” and not after the b in in “bill”. The puff of air is called aspiration. To get the Pali sound, just emphasize this a little more.

To get the unaspirated Pali “p”, pronounce the same word with an “s” in front. Compare “pill” and “spill”. In “spill”, the aspiration is almost entirely absent (There are dialectical differences here, but this should mostly work.) Try the same with “kill” vs “skill”, “till” vs “still”, and so on. (Incidentally, this is exactly the same phonetic phenomena we see in Sanskrit skandha vs. Pali khandha.)

Once you can not only feel the difference but hear it as well, try pronouncing “spill”, then dropping the “s” while keeping the rest of the word the same. The “pill” that remains sounds similar to the English “bill”, but is not exactly the same (“b” is voiced). When you can pronounce a “p” that is neither aspirated nor voiced, this is the Pali “p”.

Incidentally, “ph” and “th” are never pronounced as in the English “f” and “th” in “thin” or “there”. These sounds don’t occur in Pali. This is a basic mistake that you hear quite commonly, even among, ahem, Buddhist studies academics.

2 Likes

@Jayantha

Thanks so much!

Interesting. I’m not sure if “er” is a sensible way to describe it, but Sri Lankan’s definitely don’t use the simple short “a” that is use by Thai people.

Surely you agree that the Thai and Sri Lankan pronunciation of tassa is different?

1 Like

If I may, I would also like to present a couple of modifications :slight_smile:

If you listen to the American English, cot is actually pronounced very open, so the American ‘cot’ is frequently perceived as /kat/ by foreigners. The British/Australian /ɒ/ is indeed closer to the Pali /ɔ/ but still very different - when I was doing my bachelor studies in English I had to learn how to pronounce the /ɒ/ sound for about a month. The truth is the English phonetical system, whether British, Australian or American, doesn’t have any correspondence to Pali /ɔ/ at all. I would say the English /ɔː/ as in broad or sawed is the closest counterpart:

  1. The English vowels have three different realizations depending on the coda of the syllable: it can be open, voiced closed or unvoiced closed. The first variant, as in law, is not very suitable as this vowel is just too long. The third is not suitable as well because unvoiced codas are very often realized in non-American English with glottal stops (think of the British pronunciation of brought as /brɔːʔt/ or even /brɔːʔ/ in some positions), so it doesn’t work as well.

  2. The so-called long vowels aren’t actually that long. The phonetical difference between the short and long vowels is less relevant and obvious in the American pronunciation, so when reading the following examples keep in mind Ajahn Brahm’s way of speaking. Terribly sorry for not knowing much about Australian English, but Australia is for many students of English what Laos is for most Western Theravada Buddhists: we know that it exists and that is pretty much it. So, phonologically /ɔː/ in sought is of the same duration or even shorter than /ɒ/ in sod. The phonologically relevant feature of /ɔː/ is its tensity. If the Venerable Ajahn allows you to put your hand on his Adam’s apple when he’s speaking you will feel how it becomes somewhat tense when pronouncing sought and remains not tense when pronouncing sod. Just try to omit the tense part in /ɔː/ as in broad, and you will have a pretty decent Pali /ɔ/.

The English ch is kinda aspirated but not really the same way as in the Indo-Arian languages. In English, the aspirated element of the /tʃ/ sound is /t/ as /tʰ/ is almost always aspirated in English, hence /tʰʃ/. In the Indo-Arian languages the aspirations come not in the middle of the sound but rather towards its end, hence /tʃʰ/. So yeah, I would use ‘mischievous’ as an example for c but wouldn’t use church as an example for ch. A better approximation would be ‘catch him’.

It is not really misleading. The British /bɑːn/ and American /bɑːrn/ differ in the use of /r/, which doesn’t necessarily affect the quality of the vowel. So, this one is okay.

The advice I would give to practically all Pali reciters speaking Germanic languages would be to practice the dental pronunciations of /t/ and /d/ (nevermind the /r/, I abandoned all hope on that one). I personally can’t really listen to English chanting because the alveolar pronunciation of stops tends to detract me too much. The best way to figure out the correct place of articulation is to analyze their pronunciation before interdental consonants as in read this or put that. And thanks for the expert presentation of the aspirated consonants, it was really impressive and well-done. Sadhu!:anjal:

2 Likes

Oh, by the way, if you happen to come across the audiobook version of Ven. Bodhi’s In the Buddha’s Words, you won’t believe your ears. It is so abundant with /bʊdəˈvʌkənə/, /ˈʌrɪjəʊ ətˈθɑːŋɡɪkəʊ mʌɡəʊ/ and other phonetic delicacies that Ven. Buddhaghosa would take the vow of solitary forest living if he heard it.

:anjal:

Dear All,

Not sure if this was brought up already. But here is the Dhammacakkapavattana Sutta chanted by Dhammaruwan. Maybe this would help? I tend to pronounce Pali the way it is chanted here, ever since I started listening to his chants. Good enough for me. I think as long as the enunciation of the words are clear and convey what is to be conveyed, it is good enough! It’s all preference really!

Hope this helps a bit so as not to rack your poor minds :relieved:

Enjoy your vassa :joy:!

in mettā,
russ

:anjal:

2 Likes