Thanks for the feedback, you’ve raised some important points, so allow me to respond at some length.
Just to clarify, Ven Bodhi doesn’t translate terms the same way all the time. They are varied both when the meaning changes, but also when the parts of speech in English don’t match those in Pali.
Eg. viññāṇa is “consciousness”, while vijānāti is “cognizes”. Paññā is “wisdom”, while pajānāti is “understands”. On the other hand, saññā is “perception” while sañjānāti is “perceives”. And sometimes vijānāti means “understands” and must be translated as such. Sometimes pajānāti means “knows”, as does sañjānāti and vijānāti. And so on.
To further complicate things, “wisdom” and its derivatives are used for a variety of other Pali terms, which are not connected with paññā at all: paṇḍita, dhīra (which is sometimes rendered “steadfast”), medhāvī, viññū, and so on.
There are countless issues like this, faced in virtually every sentence. It’s quite simply impossible to translate one language to another and keep a simple one to one relation between terms. Language just doesn’t work like that. The translator who came closest was the late-period Ñāṇamoli, but no translator has followed his path, which is understandable if you’re ever read those texts.
I’m using a glossary of terms, originally based on Ven Bodhi’s usage, which I’m expanding and adapting as I go. Currently there’s 1675 technical terms and phrases. Every time I encounter a new context, I look at how the word works in the context, and try to work out what is the best possible rendering of the phrase in English. If it uses any technical tem, I see if the term has the same meaning I’ve used before, and if so, can it be rendered consistently. If not, how best express it? And on it goes, every time. Translation is slow and uncertain work, and no rule is absolute.
On the whole, however, the balance of consistency vs. context in my translation will be not very different from Ven Bodhi’s recent work. The main difference is that I strive to use the simplest expression possible, using more common and colloquial terms, and simplifying the syntax as far as possible.
As for secondary translations, be clear: this is what is happening, and will continue to happen, whatever you or I think about it.
If you want to do translations from the Pali into the world’s languages, great! Here’s how.
We’d need around 10 or so full-time Pali students each year. (Just to be clear, I don’t think there’s a single University level course in the non-Buddhist world that consistently attracts ten students per year to study Pali.)
After five years or so of full time study, and then some postgrad work, some might be ready to do mature translations. Of course, most will have dropped out, or not be very good, or have good Pali but not write well in their native language, or not be interested, and so on. And they’d have to be not just language students, but serious Dhamma practitioners, who not only had some depth and maturity of practice themselves, but had been part of a Dhamma community for a long time.
Say, then, we spend $50,000 per student per year. That’s maybe $5,000,000 per language over a ten year period, to even have a reasonable chance of getting decent translators working in the field. We’ve got texts from 30 languages on SC, so multiply that by thirty, and we’ve got $150 million over the ten years. Given that ten years is not very long, and that there’s a lot more than thirty languages in the world, let’s say, ballpark figure, half a billion dollars to equip and prepare competent translators.
I’m not joking about this, this is an entirely reasonable scenario. These kinds of numbers are in fact what’s involved in the Christian world of Bible translations. There are multiple institutions, hundreds of scholars, Universities, courses, and so on, translating the Bible into over a thousand languages. (The whole Bible is, incidentally, a little shorter than the four nikayas).
And you know what? They still rely on amateur, untrained translators doing secondary translations from “plain language”, non-literal primary translations. Typically the first Bible translations are done by missionaries working in obscure locations, and only later, if at all, are “professional” translations produced.
Google, incidentally, does the same. When you use Google Translate to go from, say, French to Hindi, it actually translates French to English, then English to Hindi. Obviously the people at Google know this is not ideal, and obviously the translations are not great. But that’s missing the point. Google is driven by data, and their decision is based on: what is the best that’s actually achievable? And it clearly makes more sense for them to work on improving translations to and from English, rather than attempt direct translations of the world’s language pairs.
So we also have to consider, what’s actually achievable? Years ago we had a guy staying at the monastery called Peter. We were making a shoe-rack for visitors, and trying to figure out the best place for it. Peter said, “put the shoe rack where people put their shoes”. Brilliant! So that’s what I’ve been doing. Look at how people are in fact producing translations, and ask, what can we practically achieve that can help make better ones?
There are good people all around the world, trying to translate the Dhamma into their own language. Of course they are well aware of their own limitations, and the difficulties and compromises involved. But they are making a start, and I am happy that we can do a little to support them. Hopefully once our translation software is widely available, people will start to use it and it will help improve the quality and consistency.
Don’t get fixated on a question like “primary or secondary” translation. It’s just one of many factors. There are so many factors involved in translations, it’s quite possible that a secondary translation can be better than a primary one. Take the case of German, for example. The classic German translations by Neumann were, I believe, made from the Pali. But they are notoriously unreliable, and serve more as a vehicle for the translator’s romantic ideals than anything else. More recently an updated translation of the Majjhima was made by Ven Mettiko, based on Ven Bodhi’s English translation, and it is much better. It’s made by someone who actually understands the Dhamma.
Another example is the several translations made in traditional Buddhist countries. Most of these were made by very good Pali experts, and are very accurate. They’re also, in several cases, terrible translations. They are archaic, difficult, overly literal, and often virtually unreadable by native speakers. The same could be said for some academic translations into English. If you’re not going to do something that native speakers can read, why bother?
Don’t be afraid of diversity. We’ll keep our own version here, and would be happy to host other versions, if they are good. But the world is what it is, and poor work will fall by the wayside.
For anyone who’s interested in these issues, there are some good, in-depth articles on Bible translation here. Obviously some of the issues are different, but many of them are similar. I found this site very useful, mainly because I disagree with the author’s position on many things, but I still found him persuasive and well-argued.
And check this out for a project somewhat similar to what we’re doing: http://openenglishbible.org/
