The hardest part, I find, about the philosophical vocabulary of Sanskrit is not its enormity or its technicality, but the fact that it includes a couple of “basic” or even “common-sense” words whose import, in any given context, can be difficult to determine and impossible to translate. I’ll focus on prāp- right now, since I think that is the hardest, but many of these remarks could equally apply to gam-, bhū-, jñā-, sādh-, etc.
The first problem is the impossibility of word-for-word translation, which results from the impossibility of translating Sanskrit words, with all of their amazing suffixes, into corresponding English words. We don’t have productive causative suffixes, for example. Big deal, you might say. But let’s say we decide to translate prāp- with “obtain,” or “apply,” or “attain.” These translations might work in some cases, but what about derivatives like prāpti? “Obtainment” isn’t English. “Application” is different. “Attainment” works in some contexts but not in others. Then we have the choice of rendering suffixed formations with paraphrases: “causes to attain” and so on. But when these pile up, the English paraphrase becomes unintelligible: something is to be caused to be attained, or the instrument of causing to attain, or a thing that causes something else to be attained.
For these reasons, I suppose, most scholars have tried to translate the meaning rather than the words. But this involves a judgment about what the meaning in fact is. I am thinking in this context of the translations of prāpaṇīya- in Dharmottara’s exposition of Dharmakīrti’s Nyāyabindu. Stcherbatsky translated it as “definitively realized” (importing the notion of adhyavasāna into the translation). Arnold translated it as “intended,” because he saw the problem that Dharmottara was addressing in terms of the modern debate about intentionality. I have to say I prefer McCrea and Patil’s simple and literal translation of “attainable.”
In the case of Dharmottara, we can see that “attaining” an object and its “attainability” is related to the criterion of practical efficacy on the one hand. On the other hand, it is the cognition itself that “causes us to attain” the object and therefore makes it “attainable” by presenting it to our awareness in a certain form—which leads me to think that “presentation” is not a bad translation for the causative prāp-I- in this epistemological sense.
In the context of the discussion of rules within Mīmāṃsā, prāpta- refers to something that is “given.” Although “given,” in English, is a funny word when you think about it, and prāpta- is no less complex. Frequently we are asked to imagine what would be the case if a certain rule did not operate; what is prāpta- is “given” but only hypothetically, or indeed counterfactually. And we can see that this evolves out of a longstanding idiom in Mīmāṃsā (and other learned discourse) where prāp- and its derivatives refer to the horizon of our expectations: kiṃ prāptaṃ, what would we expect? But these naive and even colloquial uses are, in some texts anyway, enlisted to do some heavy philosophical lifting. Again, I am thinking of the discourse on rules, where the question is precisely how a rule can apply (a condition that is itself called prāpti-) or refer to something that is “not given” (aprāpte). Don’t certain things (what things?) have to be “given” in order for us to sensibly understand rules, even rules that generate a new obligation, in the first place?
I may just be confusing myself, but these are some of the issues in producing accessible translations of Sanskrit texts for English-speaking readers: you want to be readable, but you also want to make clear the conceptual connections that are inherent in the choice of Sanskrit words; you want to relate to contemporary discussions, or at least be relatable to those discussions, but you don’t want to insert your own judgments into the text. Or do you? What other words fall into this category? What solutions have you come up with?
Andrew, first a quick comment – ‘obtainment’ is a working English word, and does the job there (as you implicitly concede it would if it were an English word).
Very interesting to see you work through ‘prāp-‘, as I have spent half a lifetime at least, grappling with ‘jñā-‘, and latterly, with ‘bhū-‘ (Reid Locklin in a review kindly calls my translations ‘mildly idiosyncratic’…).
On readability and conceptual connections, mentioned in your concluding paragraph, it may matter what kind of text you are working on and what you want to bring out. In the past, when focussing on technical questions in pramāṇavāda, I admit that I have often flattened the literary contours of texts, striving much more to show the consistency with which arguments are presented and concepts function around some particular debate. But in my current project, which deliberately engages with a range of genres, including highly literary productions like the Naiṣadhacarita, I am driving myself spare trying to both create something literary /and/ find a way of reading it philosophically. And then again, readability brings its own cultural baggage – look at the recent work by Frank Clooney, where he is consciously trying to move his treatment of Tamil bhakti works away from the once pioneering style of A.K. Ramanujan.
On your concluding questions, I am not sure how exactly to distinguish between something called ‘relatable to contemporary discussions’ and ‘inserting your own judgements’. I think it matters a great deal what your larger purpose is, and whether your readers follow you in acknowledging that purpose. Look at the issue going the other way too – e.g., the way David Shulman in More Than Real argues for a wide range of words to be doing the work of ‘imagination’.
Andrew, your last question is really hard. Personally, I am slightly suspicious of too idiomatic translations, the ones which tend to annihilate the feeling of extraneousness of a given text. Because, after all, Maṇḍana is *not* Wittgenstein and Madhva is not Russell, so that making them appear very similar risks to be misleading. At the same time, I also dislike translations which are not at all accessible without the Sanskrit text. The latter are in fact glosses to the Sanskrit text, not translations, and in this sense I think one should directly write a gloss (like John Taber did in his Kumārila on Perception) and not disguise a gloss as a translation. Thus, I tend to adopt an (I hope) intermediate approach, in the sense that I write extensive commentaries of each text I translate and I try to translate in a way which is understandable but by using a terminology (explained in a glossary or in the introduction), which echoes the complex interplay of terms in the Sanskrit text. A last point is that I tend to use English technical terms, if they happen to be available, and that I definitely avoid using terms which are used in a certain way by Western philosophers in order to convey a different meaning (say, I would not use “referentialism” for something different than its Western referent).
Now, this all clashes with the fact that English is not my mother tongue and that I thus risk to think that “obtainment” is, indeed, an English word…:-)
How would you describe what do you do?
As for prāp-, I would add that kiṃ prāptam is often just “What is the prima facie view?” (if I remember correctly, I translated it once as “What is obtained [as the prima facie view]?”). prāp- , especially prāpta, also indicates what is already established and can thus serve as a fix point.
It seems to me that your last point is that one must understand the context to understand (at least) some parts of a translation. Which is a universal for reading philosophical texts. Wittgenstein wrote in English (just down the road from where I live) and yet I have no hope of just diving in and reading his texts without the appropriate background knowledge (which I know from experience). I need to know who he was in dialogue with and some history of Western Philosophy.
No one is ever going to translate a philosophical text that does not require specialist knowledge of jargon. And this seems to me to be a partially separate issue to your first point about the difficulties of translating technical jargon where word-for-word translations are not possible. Of course knowing the technical jargon makes reading the text possible, and not knowing it makes it difficult at best. But this is always the case.
When I was reading Sanskrit texts with the Cambridge graduate students I often found just this problem. I could look a word up in a dictionary, but it didn’t help me comprehend it in context. The only worldview I’m thoroughly familiar with is Buddhist and so reading a vyākaraṇa or Mīmāṃsā text often baffled me because it was laden with jargon. Each new word required a huge amount of effort to contextualise.
This goes to a point that Alex Studholme makes in the book of his PhD thesis: The Origins of Oṃ Maṇipadme Hūṃ. Sometimes a translation is not the best medium for communicating the content of a text. Alex did a study of a later Mahāyāna Buddhist text, in which discussed the technical jargon at length and contextualised it, without offering a full translation. It’s a very useful book for understanding the transition from non-Tantric to Tantric Buddhism. A plain translation would have been much less helpful since the context requires the kind of study Alex did in order to understand it.
At the very least translation must always be accompanied by exegesis. I would guess that for every line of text we usually have about 10 of traditional exegesis; and sometimes many more contemporary lines. A well studied text might suffer several translations and dozens of commentaries and still be quite opaque to the uninitiated.
Jayarava, I very much agree with you.
Let me just add, as additional evidence, that 1) although Wittgenstein might sound difficult, his texts are nonetheless much more accessible to us, since we share so much of his background (even just insofar as we come from the same part of the world and even more so in the case of people having studied philosophy, at least at school). 2) In the case of Sanskrit śāstra, there is often the problem that one believes that knowing “Sanskrit” (as if it were a monolithic entity) should be enough to read each Sanskrit text, as if there were no specific expertise required (this is one of my pet topics, thus I apologise if I have mentioned it already). I have just finished writing an article with a linguist, Artemij Keidan, on the theory of meaning in Jayanta’s Nyāyamañjarī, where we argue that one needs a specific expertise in order to understand the text, although Jayanta at first seems a jargon-free author.
Pingback: Keywords: jnā– and vid– | elisa freschi