SuttaCentral

Principles of translation

I have recently written out the principles of translation that I have tried to employ while translating the Vinaya Piṭaka. I would like to share them here for feedback and discussion. Also, it would be great to hear of further principles that I have not mentioned.

Here are the principles in summary, followed by a more detailed discussion below.

  1. Translating (nearly) everything
  2. Translating transparently and meaningfully
  3. Principle of the least meaning
  4. (Partially) transforming the oral nature of the text into literary form
  5. Using simple language and vocabulary
  6. Avoiding Buddhist hybrid English
  7. Using American spelling and vocabulary
  8. Sometimes using multiple renderings for each Pali term or expression
  9. Principle of lectio difficilior potior

(1) Translating (nearly) everything

My guiding principle has been to make the sacred Buddhist literature as accessible as possible, regardless of the reader’s acquaintance with Buddhism or Pali. In practice this has meant that I have translated all Pali terms and expressions, with the following exceptions: Buddha, Sangha, and paṇḍaka.

There is a tendency in the translation of Pali texts to leave a significant number of terms untranslated. Typical examples of words that are left untranslated include core ideas such as Dhamma, jhāna, arahant, uposatha, kamma, Tathāgata, and nibbāna. The problem with this approach is that it may leave readers, especially newcomers to Buddhism, in the dark. In my opinion, it is better to translate a Pali term even in cases where a fully satisfactory English rendering cannot be found. At least the reader will get an approximate idea of the meaning of the text, and thereby avoiding an abrupt break in the reading experience.

(2) Translating transparently and meaningfully

My guiding ideal has been to render all the Pali text in clear and unambiguous English.

The Buddha’s purpose was to communicate his discovery to his audience. As such, it seems reasonable to assume that his words were generally intended to convey a clear and unambiguous meaning, with occasional exceptions, in particular in verse. In my experience, translations from Pali are occasionally hard to understand, especially when the underlying Pali text is itself difficult, and perhaps even corrupt. In such cases it sometimes appears as if the translator has opted for a strictly literal translation to avoid taking a stand on the meaning. The result can be a rendering that is either meaningless in English or at the very least hard to interpret. If, after trying one’s best to understand the Pali, the text remained ambiguous, it seems right to me to decide on the most likely meaning and translate accordingly. Occasionally this may result in an erroneous translation, but at the very least it gives a proper foundation for further discussion, a potential that is largely lost if the translation is meaningless or hard to grasp.

(3) Principle of the least meaning

My guiding ideal has been to render the Pali as it stands without adding unnecessary interpretation.

The Pali Canon is a spiritual corpus, sometimes with a deep meaning. Given this, it is easy to read depth and hidden meanings into the Pali texts, thereby adding one’s own interpretation. To my mind, the straightforward meaning of the text is always the most likely one. The Buddha was communicating with ordinary people. In trying to communicate his insights, he would have had to use language that was accessible to ordinary members of society and that was a common denominator for the population at large.

Assuming that the most straightforward meaning is the right one, reduces the problem of the translator adding their own interpretation of the text. A practical example of this might be rendering a term such Nibbāna as “extinguishment”, rather than giving it some deeper mystical sense, or even leaving it untranslated, which arguably leaves the text too open for the reader to project their own prejudices onto the word.

(4) (Partially) transforming the oral nature of the text into literary form

My guiding ideal has been to reduce the repetitiveness of the Pali, so as to give a more natural reading experience.

The Pali Canon has abundant signs of being an originally oral “text”, especially in its abundant use of repetition. Repetition would have aided memory and helped to stabilise the text for later generations. In written form, however, such repetition does not serve the same purpose. Moreover, it often comes across as tiresome and boring, with the result that modern readers are often put off. To remedy this and to make the texts more reader friendly, I believe it is reasonable to reduce such repetitiveness. Here are some examples:

  • The Pali text often employs series of synonyms to convey the meaning of an idea. In English it is often sufficient to reduce this to a single word.
  • When a number of items are listed, as is especially common in the Aṅguttara Nikāya, the text surrounding each item is often identical for each item. In my opinion, in English translation it is often sufficient to simply list the various items and not repeat the identical phrasing that surrounds each one.
  • The Pali texts, especially in shorter suttas of the Aṅguttara and the Samyutta, often give the same summary line at the beginning and the end of the sutta. In general it seems sufficient to keep one of these.

(5) Using simple language and vocabulary

My guiding principle has been to use ordinary words and a straightforward language.

English translations will be read by all sorts of people, including, crucially, those with English as their second language. To maximise the benefit of these texts for all people, it is important that the language is easily accessible.

It has been common, at least until fairly recently, to give translations of Pali text an artificially archaic and elevated flavour, presumably out of respect for the texts. However, it is my understanding that the main purpose of the Buddha is to communicate meaning. An artificially elevated language often hinders good communication. There is, in fact, Canonical evidence that Buddha specifically wanted to avoid this.

(6) Avoiding Buddhist hybrid English

My guiding principle has been to render everything into idiomatic English.

Too often the English of texts translated from Pali is stilted and artificial in nature. This may be reflected in the choice of vocabulary or sentence structure, which often stay too close to the Pali original. As I see it, the purpose of a translation should be to render meaning. The actual language should fade into the background and barely be noticeable. In other words, one’s attention should not be unnecessarily detained by strange expressions and words that are not immediately clear. To ensure the English is as natural as possible, it may be useful to reflect on the meaning of larger units of texts, e.g. entire paragraphs.

(7) Using American spelling and vocabulary

My guiding principle has been to use the English form that is likely to be most easily understood by the majority of people.

It is a fact, whether one likes it or not, that American English is globally better understood than other forms of English. If these texts are to reach a global audience, it is therefore pragmatic to use American spelling, vocabulary, and expressions.

(8) Sometimes using multiple renderings for each Pali term or expression

My guiding principle has been to translate words according to context.

There is usually no perfect semantic overlap between a given Pali word and any chosen translation into English. Translation projects from Pali have sometimes been hampered by an insistence on a one-to-one relationship between the source word in Pali and the chosen English rendering. Yet to do full justice to the meaning of the Pali, any given Pali word or phrase will often require multiple renderings. To avoid a sprawl of different renderings for the same Pali term, it is important that each rendering stays the same for the same or similar contexts.

(9) Principle of lectio difficilior potior

Unusual readings of the Pali should be taken seriously. Such readings may sometimes be the result of scribal faults, but often they may reflect interesting and even important archaic features of the text. Unnecessarily homogenisation of the text will diminish the richness of the original.

15 Likes

Dear Ajahn Brahmali,

I wonder if you could say a few words on the subject of use of jargon/technical terms and how to treat this as opposed to prose and verse. You have touched on it in points 1 and 6, but I wonder if you could draw it out a bit more :slight_smile:

It’s something that I feel has an inherent tension.

Given that jargon is about establishing a meaning beyond the common usage of a word, and the Buddha himself did this, carefully defining a term (transforming it into jargon) and then using it all the time in that specific way - I think of this aspect as ‘Doctrine/principles’. But as you say he would also use very accessible common language to explain, describe and illustrate these principles and how they apply in daily life by using prose and verse.

The Buddha did not have a ‘glossary’ (oral transmission), but most big translation projects use one in order to make the meaning of these terms/jargon clear as they apply throughout the translation. The definitions are in prose, but the terms themselves have become jargon or ‘symbols/labels’ with a specific meaning, that is no longer common usage.

I’d be interested in hearing your thoughts on how to approach this aspect of translating, and what factors are the most important to consider.

Many Thanks :slight_smile:

5 Likes

I am guessing that by “jargon/technical terms” you mean words such as jhāna, arahant, and nibbāna. In many ways we are in a better position to understand these than the contemporaries of the Buddha. After 2,500 years of Buddhism, all these terms have been thoroughly discussed. For the Buddha, however, it was much more difficult. He was working with an existing vocabulary with culturally given meanings, and then had to redefine them for Buddhist purposes.

But I suppose you mean how do we correctly convey the meaning of a specialized spiritual vocabulary in English. Here are a few principles:

(1) Don’t rely too much on literal meaning or etymology, but instead on context.
(2) If necessary, render a given Pali term into more than one word in English, e.g. “willed activity” or “conditioned phenomenon” for saṅkhāra, or “placing the mind” for vitakka.
(3) Sometimes it is hard to find literal words that are suitable and it may be best to use words that are largely metaphorical, such as “absorption” for jhāna or “stillness” for samādhi.

But really, it is impossible to get the full meaning across merely by translating “correctly”. This is one of the reasons for Ven. Bhikkhu Bodhi’s extensive endnotes. Even then it is hard to fully convey the meaning of the Pali. In the end, the reader will have to consult extensively in the Pali Canon, gradually building up an understanding.

Do you have any suggestions for how to render such specialized vocabulary? I could do with some alternative perspectives! :slightly_smiling_face:

6 Likes

Dear Bhante Brahmali,

When using those words for the first time in english, bracket the pali words so that the not-new-to-pali folks know which pali word is being translated into which english word.

I find this a bit more readable, compared to only having a listing of pali-english terminology mapping at the end or beginning.

2 Likes

Hi Xin Zhao, how are you? I have noticed your new appearance as a monk. Where are you these days?

Yes, sometimes it is a good idea to put Pali words in brackets. But in the case of SuttaCentral the display options are so good - that is, you can display the English with the Pali, line-by-line - that adding Pali words in brackets largely seems redundant. If at all possible, I try to avoid burdening the texts with bracketed words and other extraneous things, so as to make the reading experience as comfortable and straightforward as possible.

Good luck with your monastic life!

5 Likes

I love this approach of translating meaningfully, but as a reader who knows almost no Pāli, I have no clue as to when your reading is high or low confidence, and that would be good information to know!

Having read on SuttaCentral for a few years now, I have grown a bit frustrated by the lack of such notes. I know this isn’t the intent, but it comes across as overconfident: “this is what it means and that is that.” While translating meaningfully opens up the opportunity for disagreement and discussion, the lack of notes makes it feel like you’re closing off that discussion, so I do hope you and Bhante Sujato will reconsider the “no notes” presentation on SuttaCentral :pray:

Especially as these English translations are starting to be used as resources for translating the texts into other languages, such “translator’s notes” would be extremely helpful.

6 Likes

Thank You Ajahn for your reply :pray:

I was thinking more along the lines of things like lists and categories eg. 5 Hindrances, 5 Aggregates, 37 Awakening Factors, 6 sense doors etc

I’ll use the term Aggregates as an example as I know you don’t really like it - (but I do) :smiley:
In your sutta classes you often say you don’t like it because it doesn’t give a good idea of what is being discussed, and I believe that you have on occasion suggested an alternative of ‘personality factors’. Now I’m only using this as an example to work through the thinking - I have no fixed view or vested interest.

5 Aggregates is obviously a summarising technical term (jargon). From my perspective no word can possible adequately describe it - it requires much explanation and is mostly focused on categorising each of the 5 aggregates into a whole, by ‘function’. As such one could call it “group 13b”, or ‘xcghr’ or even :aquarius: The intended meaning is not inherent in the word itself but has to come from the explanation.

I’m a fan of the term aggregate, because it captures something of the tone of ‘component parts’ of a Human - not a single indivisible ‘thing’ but a group of things or attributes etc.

I find the fact that it is relatively meaning-neutral is actually useful, it has only enough of inherent meaning in the word to give a clue what it relates to, to help jog memory - As opposed to having to learn a different word/designation or symbol for it.

This relative neutrality of meaning is why I prefer it to ‘personality factors’. That term is much more loaded with meaning and can mislead people as to its meaning. It is a standard term in western psychology, if you google it you get a long list (I attach a link at the end), and most people will have this meaning conditioned into their thinking to some degree - so there is a very real danger of misleading (by the unconscious assumptions of) the reader.

This is the aspect that I find most troublesome - that Jargon should clearly be recognisable as jargon, so that people don’t make an assumption that they know what it is. I know that in my old field (psychology) this is especially challenging since most of the words are common usage words, but used with specific and different meanings. One of the ways of designating that a term is jargon is by using Capital Letters as a signal.

I have resisted learning Pali, but even in Pali one has to learn the jargon of the Buddha - the terms he used for teaching doctine. Whether I had to learn them in Pali or in English really didn’t seem to make any difference to practice.

What I find does have a significant impact on practice though, is when these terms have multiple and changing translations. It gets very confusing, one has to learn the jargon of each translator. I don’t think there is any easy solution here, and I do love the fresh perspectives that come from new works. But you can imagine - say one is coming to terms with the teachings, and one has learned 20 lists… next author… having to learn 20 new sets of lists (or even worse just fragments being different here and there) and then transposing them for comparisons becomes quite challenging. In this case it then makes sense just to learn the pali so that one has one standard set of terms to deal with.

I don’t imagine for one moment that there is an easy or ‘right’ way to go about it. But there is perhaps some scope to consider not varying the translations of some terms, that have come to be widely used, just to capture a tiny nuance, in this class of terms. To go back to the example of Aggregates… I really don’t care what exact word is used, because it is clearly already code/jargon. As long as I recognise it easily. To really understand the meaning of it (or any replacement word) requires a lot more than just reading it on it’s own.

I hope this somewhat brings out what I’m trying to get at :smiley:

Addendum: I suppose I’m opening the can of worms called standardisation :scream::astonished: but only as it applies to a narrow set of terms :innocent:

Actually the more i think about it the more problems i see with what I’ve said… please consider it just a :bug: (worms) eye view :grinning_face_with_smiling_eyes::pray:t2::sunflower:

4 Likes

Dear Bhante,

Thanks! I am fine. Training with the forest monks in Malaysia, Sasanarakkha Buddhist Santuary for a while. Extremely limited internet here. Still apprehensive about only 1 set of 3 robes. (how often to wash, smelly, what to wear when washing, I was wondering if can wear the double robe when upper and lower robes are being washed.)

On the translation, sorry I thought it’s for a book (printed or ebook), since it’s for the suttacentral website, it’s definitely good enough as it is for the options to view pali available.

Having read on SuttaCentral for a few years now, I have grown a bit frustrated by the lack of such notes. I know this isn’t the intent, but it comes across as overconfident: “this is what it means and that is that.” While translating meaningfully opens up the opportunity for disagreement and discussion, the lack of notes makes it feel like you’re closing off that discussion, so I do hope you and Bhante Sujato will reconsider the “no notes” presentation on SuttaCentral :pray:

I thought this forum is the discussion place for the suttas? Haha.

3 Likes

Both the venerables are correct! The forum is indeed a place for discussion. But our new Bilara system does in fact support notes. We don’t have them ready for publication yet. Ven Brahmali has written notes for the Vinaya, and I have some scrappy comments on my translations, and am hoping to pull together a team to write a full set of notes in the future.

One of the reasons for bringing notes in is that our new system allows for a clean separation between notes and text. In HTML-ish systems you have to write something like

here is some text<note>here is a note</note> and here is more text.

And this gets real ugly real fast. Now we can do something like this:

translation file:
mn123:123: "here is some text"
mn123:124: "and here is some more text"

Completely separate notes file:
mn123:123: "here is a note"

Then we mush them together for display, in any way we like.

8 Likes

Indeed that’s what I end up doing, but the forum search is only so good and is a bit cumbersome (it requires people knowing about this forum and how to use it, etc).

EDIT: Great to hear a future version of SC may support notes! :grinning_face_with_smiling_eyes: Thanks so much Venerable for all your hard work :pray:

3 Likes

One thing I’ve been mulling over is that with electronic texts we possess the ability to allow them to be fairly extensively configured based on a reader’s preference. So, if I prefer “aggregate,” or perhaps Ajahn Brahmali’s translation (which I’m quite fond of), over the Pali word khandha I can tick a box that will set which translation is used in the sutta I’m reading. Or I could chose, as a group, which Pali words that are already widely known to English speakers to leave untranslated, like aggregate, arahant, and nibbana. In an electronic text we aren’t limited to having to use only one translation.

We could have some something similar for all the repetition found in Buddhist texts. We could hide the repetition by default, but allow a user to view all the repetitions by choosing that in a settings dialogue. Of course, all of this requires a lot of work. So time and person-power are major factors in deciding whether or not something like this ever comes to pass. Before even thinking about attempting to add these kinds of features, we should collect some data from users to gauge interest.

1 Like

Harder than it sounds …

2 Likes

Haha, I’m sure it is.

1 Like

As a rule, if there is a note that discusses a term, a phrase, or any larger piece of text, you can take it that the interpretation is less than certain. And notes are coming, so say the powers that be.

Then there are the cases where I was sure of the interpretation, but didn’t even realize my own ignorance. I will have to rely on feedback to sort those out. Would you like to contribute?

Sure, but some renderings will describe it better than others. A good translation will at least point the reader in the right direction. That’s sort of been my guiding principle.

I do see what you mean, and you certainly have a point. Khandha is actually quite close in meaning to “aggregate” or “group” or “collection”, all of which would be satisfactory renderings. And yes, “aspects of personality” is itself far from unambiguous and definitely weighted down by considerable baggage. Now that I think of it, maybe “aspects of experience” or “aspects of personal existence” would be better. (“Personality” usually means “character”, which is not the intended meaning here.)

Still, I am not very enamoured of “aggregate”. The word itself is not just technical, but fairly rare. It doesn’t sit well with ordinary speech. Would a spiritual teacher today use such a word to communicate their unique insight? It seems unlikely to me. They would use a a more natural language. My sense is that khandha is much more of an everyday word than aggregate. In part this follows from its fairly common use in the suttas.

But more generally, I think it is good to make the text more intuitively understandable to the reader. If you hear the phrase “five aggregates” out of context, you will be completely stumped. You might even lose you interest in the suttas. I think it is better to give a meaning that is imperfect—even potentially misleading, depending on one’s background—but that at least allows the reader to carry on, perhaps with an underlying suspicion that there is more to be learned.

I think the best advice is to find a good translator and then stay with them until you have a good grasp of the teaching. When you have reached a certain level of confidence, I believe it can be enriching to read alternative renderings.

We can certainly agree on this. :grinning: And thanks so much for your thoughtful contribution!

Great! Please pass on my respects and best wishes to Bhante Aggacitta.

5 Likes

Though this is getting away from the Principles of Translation… I just thought I’d share… here is a picture of my favorite ‘souvenir’. I picked it up off the side of the road going to Jhana Grove, when I was last there. It is my representation of Not Self… and as beautiful as any Buddha statue

The rock (all over Bodhinyana and JG), that is nothing but a group of component parts bound together - No Self - just an aggregate :slightly_smiling_face::relieved:

6 Likes

He’s not here at the moment, Bhante AriyaDhammika is here.

https://sasanarakkha.org/monastery/

This is the biggest thing that I run into as a reader. There becomes a tipping point beyond which it makes sense to use at least some Pali or Sanskrit terms. One example that comes to mind is when a term like jhāna is translated as simply meditation, which does not help to deepen the reader’s understanding of the text. It also prevents them from being able to use the translation in their own personal studies.

As a reader, I like it when some terms like buddha, dharma, saṃgha, arhat, pratyekabuddha, dhyāna, samādhi, etc., are kept just like that. There are some other terms that I could go either way on, though, like ṛṣi, bhagavān, sūtra, or gāthā. They add flavor, and sometimes it’s nice to know the original word, but they could be translated without losing much meaning.

I agree wholeheartedly with regards to Pali (but I’m not so sure about mixing in Sanskrit). Some Pali words just don’t translate neatly into an english word and it’s better in the long run to investigate those words, learn what the Pali is conveying so that the suttas make more sense. For instance, I just can’t resonate with translating dukkha as stress.

I’ve always thought it would be useful to have an online translation that retains all the difficult-to-translate Pāli terms, but when you hover your mouse cursor over the term a little box pops up that gives a concise definition. This would allow the experienced reader, who already understands the terms, to avoid the dissonance of idiosyncratic translation choices and would also allow the new reader to gradually learn the terms organically, in the context of a sutta, by activating the pop-up whenever they need a reminder.

5 Likes

I’m just giving examples of the types of terms, but I’m not referring to Pali specifically.

When translating from, e.g. Chinese or Sanskrit, it’s typical to use Sanskrit terms.

1 Like