Bias in translation

[Lotomístico]

Good day,

Hi, newbie here, finally getting started on a new thread.

Recently read an artícle, about western Taoists and translation of the Tao Te Ching....got me thinking about something regarding translation of Buddhist sutras.

How much bias based on the cultural background of the author is involved? 19th century orientalist scholars perhaps trying to impose their own understanding of what Buddhism is...some even using Christian terminology (one specifically I can think of)

More recent translations possibly leaning towards a new agey or western rationalist overlay to Buddhism.

And of course, sectarian bias in sponsored translations cannot be ignored. Or when it is primarily intended to be mass-marketed with profit as the primary objective.

Do modern translations usually consult existing texts? Are the translators scholars in Sanskrit or classical Chinese? Or do many resort to paraphrase of existing translations?

One could even ask if a similar process didn't occur when Sutras were translated from Sanskrit to Chinese.

I realize that translation is an imperfect science and I'm not accusing all translators of bias, certainly not always intentionally.

Any thoughts on this? I will defer to our more knowledgeable posters for some insights on this theme.

Best and looking forward to thoughtful and considered replies.

[Zhen Li]

The best way to get an idea of whether there's bias involved is to study the primary languages ourselves and translate directly.

But to answer your questions, there are hundreds of people translating Buddhist texts, and hundreds of translations. It is easier to focus on a single example. I have a lot of experience now with Pali, Sanskrit, and Chinese, and a few other languages, so may be able to help with some generalisations in response to your questions.

How much bias based on the cultural background of the author is involved? 19th century orientalist scholars perhaps trying to impose their own understanding of what Buddhism is...some even using Christian terminology (one specifically I can think of)

If someone is directly inserting ideas not found in the source text, or excluding materials, that's going to be seen pretty quickly. Maybe you had in mind someone like I.B. Horner, who chose not to translate passages dealing with monastic sexual misconduct because it appeared offensive. A modern equivalent might be pronoun choice. The Buddha often refers to the assembly as "kulaputrāḥ" sons of good family and the masculine pronoun is standard, but some may prefer to make it inclusive (by the way, many sūtras are gender inclusive and mention "daughters" of good family and use the feminine pronoun). It is probably standard to translate deva as god, but obviously, this could be confused with Abrahamic ideas of monotheistic gods if someone isn't familiar with Buddhism. I think, in general, a lot of the potential confusion is coming on the side of the reader—if you have read a large quantity of sūtras, you will begin to identify what terms the translator is using for what ideas. Others might be a bit harder to parse, for instance "own-being" and "inherent existence" are both common translations of sva-bhāva. So, at the end of the day, some familiarity with Sanskrit is the best solution.

More recent translations possibly leaning towards a new agey or western rationalist overlay to Buddhism.

Stephen Batchelor's translations of the four noble truths as the "four tasks" are obvious examples where an author's idiosyncratic philosophy can distort what the source text is saying.

Do modern translations usually consult existing texts? Are the translators scholars in Sanskrit or classical Chinese? Or do many resort to paraphrase of existing translations?

People tend to use the same terms that have become common place because English-speaking Buddhist audiences are familiar with them. The first lines of sutras are an example: Evaṃ mayā śrūtaṃ is literally "Thus it was heard by me," but the convention is to translate it as "Thus have I heard" removing the instrumental meaning in mayā. The full line Evaṃ mayā śrūtaṃ ekasmin samaye could be read "It was heard by me at one time" but there has been a convention, which may come from manuscripts inserting punctuation markers between śrūtaṃ and ekasmin, of reading it as "It was heard by me. At one time ..." In fact, it is not possible to know the originally intended form of this line, but convention in India eventually went with splitting it. The conventional full translation in English, which most translations go with, is "Thus have I heard. At one time ..."

In fact, the Sanskrit texts are biased inherently. The middle-Indic originals of many terms, such as sutta, are ambiguous. Sutta can be read in Sanskrit as sūkta, well spoken, or as sūtra, thread/discourse. There are good arguments that it should be read as sūkta; sūtra has lots of other uses, such as to indicate sections in grammatical texts, and doesn't necessarily have any reason to be read as sūtra. However, Pāli commentaries accept that sutta can be read as both sūkta and sūtra. That Sanskrit sticks with sūtra, and many other such translation choices, makes it inherently biased in much the same way that our English translations are biased if we are concerned with particular word choice. Even bodhisattva, from bodhisatta, is a choice between a variety of options, such as bodhiśakta. Anyway, the Sanskrit tradition decided that it is bodhisattva, and thus it has been so for two thousand years.

So, issues to keep in mind are really exclusion, interpolation, source manuscript/edition choice, and word choice.

But if you read enough, you will begin to get an idea for what the intended meaning of the Mahāyāna sūtras is. In fact, these distinctions are unlikely to make a huge impact unless one is focusing one's practice only on one translation. Read the sūtras widely and continue reading, it will pay off.

[Malcolm]

The best way to get an idea of whether there's bias involved is to study the primary languages ourselves and translate directly.

But to answer your questions, there are hundreds of people translating Buddhist texts, and hundreds of translations. It is easier to focus on a single example. I have a lot of experience now with Pali, Sanskrit, and Chinese, and a few other languages, so may be able to help with some generalisations in response to your questions.

How much bias based on the cultural background of the author is involved? 19th century orientalist scholars perhaps trying to impose their own understanding of what Buddhism is...some even using Christian terminology (one specifically I can think of)

If someone is directly inserting ideas not found in the source text, or excluding materials, that's going to be seen pretty quickly. Maybe you had in mind someone like I.B. Horner, who chose not to translate passages dealing with monastic sexual misconduct because it appeared offensive. A modern equivalent might be pronoun choice. The Buddha often refers to the assembly as "kulaputrāḥ" sons of good family and the masculine pronoun is standard, but some may prefer to make it inclusive (by the way, many sūtras are gender inclusive and mention "daughters" of good family and use the feminine pronoun). It is probably standard to translate deva as god, but obviously, this could be confused with Abrahamic ideas of monotheistic gods if someone isn't familiar with Buddhism. I think, in general, a lot of the potential confusion is coming on the side of the reader—if you have read a large quantity of sūtras, you will begin to identify what terms the translator is using for what ideas. Others might be a bit harder to parse, for instance "own-being" and "inherent existence" are both common translations of sva-bhāva. So, at the end of the day, some familiarity with Sanskrit is the best solution.

More recent translations possibly leaning towards a new agey or western rationalist overlay to Buddhism.

Stephen Batchelor's translations of the four noble truths as the "four tasks" are obvious examples where an author's idiosyncratic philosophy can distort what the source text is saying.

Do modern translations usually consult existing texts? Are the translators scholars in Sanskrit or classical Chinese? Or do many resort to paraphrase of existing translations?

People tend to use the same terms that have become common place because English-speaking Buddhist audiences are familiar with them. The first lines of sutras are an example: Evaṃ mayā śrūtaṃ is literally "Thus it was heard by me," but the convention is to translate it as "Thus have I heard" removing the instrumental meaning in mayā. The full line Evaṃ mayā śrūtaṃ ekasmin samaye could be read "It was heard by me at one time" but there has been a convention, which may come from manuscripts inserting punctuation markers between śrūtaṃ and ekasmin, of reading it as "It was heard by me. At one time ..." In fact, it is not possible to know the originally intended form of this line, but convention in India eventually went with splitting it. The conventional full translation in English, which most translations go with, is "Thus have I heard. At one time ..."

In fact, the Sanskrit texts are biased inherently. The middle-Indic originals of many terms, such as sutta, are ambiguous. Sutta can be read in Sanskrit as sūkta, well spoken, or as sūtra, thread/discourse. There are good arguments that it should be read as sūkta; sūtra has lots of other uses, such as to indicate sections in grammatical texts, and doesn't necessarily have any reason to be read as sūtra. However, Pāli commentaries accept that sutta can be read as both sūkta and sūtra. That Sanskrit sticks with sūtra, and many other such translation choices, makes it inherently biased in much the same way that our English translations are biased if we are concerned with particular word choice. Even bodhisattva, from bodhisatta, is a choice between a variety of options, such as bodhiśakta. Anyway, the Sanskrit tradition decided that it is bodhisattva, and thus it has been so for two thousand years.

So, issues to keep in mind are really exclusion, interpolation, source manuscript/edition choice, and word choice.

But if you read enough, you will begin to get an idea for what the intended meaning of the Mahāyāna sūtras is. In fact, these distinctions are unlikely to make a huge impact unless one is focusing one's practice only on one translation. Read the sūtras widely and continue reading, it will pay off.

And of course, there is the fact that bodhisatva, hero of awakening, was “corrected” to by bodhisattva by 19th century philologists, leading to a total misunderstanding of this term.

[Lotomístico]

Great replies! Thank you both.

[Zhen Li]

And of course, there is the fact that bodhisatva, hero of awakening, was “corrected” to by bodhisattva by 19th century philologists, leading to a total misunderstanding of this term.

In contemporary Newar, where Vajrācāryas still use (thankfully) unedited manuscripts, bodhisatva is still the standard spelling. A difficulty is that outside of Buddhist usage, declension of bodhisatva as an -an stem nominal is not found in Buddhist texts ("bodhisatvan", declined like ātman), but Buddhist texts decline it as an -a stem. This probably lead to the assumption that it must be sattva. The solution is simply that BHS satva is declined as an -a stem—which is just how the majority of Sanskrit scribes have treated it.

Great replies! Thank you both.

That's no problem, but I just want to suggest that if you want opinions on a translation or translator, if there's not a thread on them on this forum, you can ask about it and you'll probably get plenty of views.

[Lotomístico]

And of course, there is the fact that bodhisatva, hero of awakening, was “corrected” to by bodhisattva by 19th century philologists, leading to a total misunderstanding of this term.

In contemporary Newar, where Vajrācāryas still use (thankfully) unedited manuscripts, bodhisatva is still the standard spelling. A difficulty is that outside of Buddhist usage, declension of bodhisatva as an -an stem nominal is not found in Buddhist texts ("bodhisatvan", declined like ātman), but Buddhist texts decline it as an -a stem. This probably lead to the assumption that it must be sattva. The solution is simply that BHS satva is declined as an -a stem—which is just how the majority of Sanskrit scribes have treated it.

Great replies! Thank you both.

That's no problem, but I just want to suggest that if you want opinions on a translation or translator, if there's not a thread on them on this forum, you can ask about it and you'll probably get plenty of views.

Personally, I like the translations of many of the Sutras and Buddhist texts on the BDK websites, I find them very scholarly and I can download them as free PDF.
I also like the Bunno Kato translation of the Threefold Lotus Sutra (which includes the Lotus Sutra as well as the Sutra of Innumerable Meanings, and the Sutra of Meditation on the Bodhisattva Universal Virtue), which I also downloaded as a free PDF. I read that there's a new revisión of it out or soon to be, I wonder why they revised and how different it is from the previous edition.

[Zhen Li]

Personally, I like the translations of many of the Sutras and Buddhist texts on the BDK websites, I find them very scholarly and I can download them as free PDF.
I also like the Bunno Kato translation of the Threefold Lotus Sutra (which includes the Lotus Sutra as well as the Sutra of Innumerable Meanings, and the Sutra of Meditation on the Bodhisattva Universal Virtue), which I also downloaded as a free PDF. I read that there's a new revisión of it out or soon to be, I wonder why they revised and how different it is from the previous edition.

That is the Rissho Kosei Kai translation. I spoke with a member of theirs who said it was redone to update terminology which is now commonplace in English langauge Buddhism and to improve the flow of the phraseology.

BDK translations are decent in terms of content but editing is inconsistent (especially with their Āgama publications) and I always find so many spelling errors. I think that Blum made a lot of simple mistakes in interpretation with passages in the Nirvāṇa Sūtra and is largely no different from the Yamamoto translation except for layout and word choice. I think they tend to just entrust that the translators know what they're doing but often they are not professional translators but simply scholars in the field—this is something which needs reflection. Some of their translations are less scholarly. Their Awakening of Faith translation is just a reprint of Hakeda's translation from 1967 and is quite poor. They are apparently redoing it, but a 2019 translation recently came out so I am not sure if it is the best priority. In all, they are doing good work, but I think they need to be a bit more careful in some areas.

84,000 has a system of continual revision. If I find an error in the translation or something that looks strange, they usually correct it immediately or put a note. With BDK they cannot do this because they are still working with publication of editions. To make a correction would require an entire second edition. So, definitely the open source internet based model that 84,000 is using is a good way forward. I also find that they seem to have consistent translation and editing quality. Take a look at their Lotus Sutra translation. I personally liked it. It is a different recention from the Chinese ones though.

[ratna]

And of course, there is the fact that bodhisatva, hero of awakening, was “corrected” to by bodhisattva by 19th century philologists, leading to a total misunderstanding of this term.

In contemporary Newar, where Vajrācāryas still use (thankfully) unedited manuscripts, bodhisatva is still the standard spelling. A difficulty is that outside of Buddhist usage, declension of bodhisatva as an -an stem nominal is not found in Buddhist texts ("bodhisatvan", declined like ātman), but Buddhist texts decline it as an -a stem. This probably lead to the assumption that it must be sattva. The solution is simply that BHS satva is declined as an -a stem—which is just how the majority of Sanskrit scribes have treated it.

Great replies! Thank you both.

That's no problem, but I just want to suggest that if you want opinions on a translation or translator, if there's not a thread on them on this forum, you can ask about it and you'll probably get plenty of views.

AFAIK it is standard scribal practice in Sanskrit to degeminate double t before a semivowel (v). Thus, in manuscripts we have not only bodhisatva, but sarvasatva, tatva, etc. Apparently this is kosher according to traditional Sanskrit grammar. In case of bodhisatva, perhaps the Tibetan byang chub sems dpa' recognizes the dual meaning of satva as hero/being and tries to preserve it in translation; hence sems dpa', not byang chub dpa' bo or something like that.

[Malcolm]

In case of bodhisatva, perhaps the Tibetan byang chub sems dpa' recognizes the dual meaning of satva as hero/being and tries to preserve it in translation; hence sems dpa', not byang chub dpa' bo or something like that.

Unlikely, rather, it is calque, "bodhicitta hero."

[Lotomístico]

Personally, I like the translations of many of the Sutras and Buddhist texts on the BDK websites, I find them very scholarly and I can download them as free PDF.
I also like the Bunno Kato translation of the Threefold Lotus Sutra (which includes the Lotus Sutra as well as the Sutra of Innumerable Meanings, and the Sutra of Meditation on the Bodhisattva Universal Virtue), which I also downloaded as a free PDF. I read that there's a new revisión of it out or soon to be, I wonder why they revised and how different it is from the previous edition.

That is the Rissho Kosei Kai translation. I spoke with a member of theirs who said it was redone to update terminology which is now commonplace in English langauge Buddhism and to improve the flow of the phraseology.

BDK translations are decent in terms of content but editing is inconsistent (especially with their Āgama publications) and I always find so many spelling errors. I think that Blum made a lot of simple mistakes in interpretation with passages in the Nirvāṇa Sūtra and is largely no different from the Yamamoto translation except for layout and word choice. I think they tend to just entrust that the translators know what they're doing but often they are not professional translators but simply scholars in the field—this is something which needs reflection. Some of their translations are less scholarly. Their Awakening of Faith translation is just a reprint of Hakeda's translation from 1967 and is quite poor. They are apparently redoing it, but a 2019 translation recently came out so I am not sure if it is the best priority. In all, they are doing good work, but I think they need to be a bit more careful in some areas.

84,000 has a system of continual revision. If I find an error in the translation or something that looks strange, they usually correct it immediately or put a note. With BDK they cannot do this because they are still working with publication of editions. To make a correction would require an entire second edition. So, definitely the open source internet based model that 84,000 is using is a good way forward. I also find that they seem to have consistent translation and editing quality. Take a look at their Lotus Sutra translation. I personally liked it. It is a different recention from the Chinese ones though.

Probably part of the issue with BDK is so many translators of different backgrounds, and reprints of old translations as you mentioned. Though to be fair, translating so many texts is quite an undertaking.
I'm really not familiar with 84.000, they have an extensive collection of most Mahayana texts? I'll have to check them out.

[Zhen Li]

Probably part of the issue with BDK is so many translators of different backgrounds, and reprints of old translations as you mentioned. Though to be fair, translating so many texts is quite an undertaking.
I'm really not familiar with 84.000, they have an extensive collection of most Mahayana texts? I'll have to check them out.

Yes, and of course I am not sure how much profit comes from these publications. BDK as a whole does good work. Just on the matter of whether the translations themselves are the best, it is worth keeping in mind that some of them have issues, but the majority are very good quality.

84,000 are apparently trying to translate as much of the Tibetan canon as possible. I don't think they are anywhere near having a collection of most Mahāyāna texts, but they have a lot and have a very good pace of publication.

[Lotomístico]

Probably part of the issue with BDK is so many translators of different backgrounds, and reprints of old translations as you mentioned. Though to be fair, translating so many texts is quite an undertaking.
I'm really not familiar with 84.000, they have an extensive collection of most Mahayana texts? I'll have to check them out.

Yes, and of course I am not sure how much profit comes from these publications. BDK as a whole does good work. Just on the matter of whether the translations themselves are the best, it is worth keeping in mind that some of them have issues, but the majority are very good quality.

84,000 are apparently trying to translate as much of the Tibetan canon as possible. I don't think they are anywhere near having a collection of most Mahāyāna texts, but they have a lot and have a very good pace of publication.

My impression of BDK is that their products are really not meant for mass marketing or consumption, most editions are costing $35-$50, not in many people's price range (though worth it imho) your average Joe probably isn't looking for a treatise by Vasubandhu or whatnot...and of course all of BDKs books are available as free PDF.