Talk:Perfection of Wisdom

I noticed that here, and in a couple other places, "Perfection of Wisdom" has been changed to "Perfection of Insight". I'm not sure this is a good idea. Generally, I'm sympathetic to being bold in improving translations (see Talk:The Gateless Gate). However, in this case, the accepted usage seems to be overwhelmingly in favor of "Perfection of Wisdom" (14,000 hits on google versus about 200 for "insight"). I've seen several other translations for Prajnaparamita, there doesn't seem to be what accepted alternative to "wisdom", anyway. - Nat Krause 04:31, 9 Dec 2004 (UTC)

---

The sample pages pointed to on the Amazon link are all prominently marked as copyrighted material, so I'm assuming the extracts here are also, and I've removed them. Revert them if this isn't so... --Malcolm Farmer

Reverted. Malcolm is now happy that the author is happy to have this extract here. If in doubt ask the author: rcj10@cam.ac.uk

---

Diacritics on anglicised forms. Kukkurovaca has installed diacritics on all instances of the word prajnaparamita in this text. I would propose the following distinction. Where a Sanskrit form is being given (and so normally in italics), all diacritics should be used. Where a term is used un-italicised in running English text, it is assumed to have been anglicised, and so the diacritics are dropped. Otherwise, to be consistent, we should be adding length marks every time the words sutra or mahayana show up, and a retroflexion dot for tripitaka. In general, I think that the anglicised forms should just be the Sanskrit form minus the diacritics except: ś and ṣ become sh (thus Shariputra not Sariputra), and ṃ becomes n except before p, b or m (and so sangha not samgha). What do others think? --Parshva

I think that "c" should become "ch", too (and I've considered whether "ñ" should become "ny"), but I don't want Kukkurovaca to start pulling his hair out. Really, though, your proposal sounds right to me. - Nat Krause 13:15, 30 Jun 2004 (UTC)

 ::laughs:: I appreciate the concern; I'm rather fond of my hair. My objections are (1) diacritics in Sanskrit are essentially the same as good spelling in English; there's an accuracy issue. Indeed, it's a stronger issue than spelling, since spelling variations are negativenative (wow am I dyslexic) to English, while Indian languages are transcribed phonetically. (2) Generally speaking, a modern Encyclopedia (such as Britannica, not that we should imitate them in all things) will use the best transliteration available for a foreign word, following scholarship. (3) Even if we were to adopt a non-diacritic-based transliteration scheme, it's unwise to use a lossy one by which certain characters merge. Palatal "s" and retroflex "s" are not the same letter; similarly, if we use "ch" for "c", how will we represent "ch" (an aspirated "c")? And let's not talk about the distinction between long and short vowels going down the drain. A lossless system like ITRANS would be acceptable, I suppose, but it's ugly as sin. (4) This is important because someone may find a term here and want to look it up elsewhere, and could be impeded by not knowing what "s" is what. I've had this problem myself while doing fact-checking on Wikipedia articles. (5) We certainly shouldn't go around making up romanizations. As far as I'm concerned, if a word is used widely in the English language already (like "nirvana"), it's not a deadly sin not to use proper diacritics (nirvāṇa), because the average user knows the word and isn't going to be confused by it, and isn't going to be impeded in looking for it elsewhere, probably. But if the average person doesn't know what a term is, we should introduce it to them in all seriousness. (6) -- कुक्कुरोवाच|Talk‽ 17:10, 30 Jun 2004 (UTC)

As I understand the proposal, we would continue using the Sanskrit etceterata for words that have not been absorbed into the English lexicon, and use the revised Anglicoid transliteration only for terms (such as nirvana) that have been lexicalized. However, I have no idea what objective criteria to use to distinguish between lexicalized and unlexicalized words. Examples of words in the English lexicon: nirvana, sutra, Buddha, dharma, sangha, karma. But what about prajna or prajnaparamita? Hard to say. Also, what about the names of people? I'm a lot more comfortable with "Shakyamuni" over śakyamuni, but I don't know if I could really suggest using that style in every case. Aśvaghośa or Ashvagosha? - Nat Krause 08:26, 1 Jul 2004 (UTC)

My standard would be something like: walk up to five people on the street, make sure they're not all monks or orientalists, and ask them if they know what the word means and how to spell it. If they all say yes and offer the same spelling, it's officially English. I'm not at all sure "sangha" meets this criterion, though the others you mention at first would, I think. But certainly not prajñā or its derivates. Of course, I don't think I've once heard someone pronounce prajñā, and that's including my religious studies profs and a number of Buddhists, so perhaps asking them to work the diacritics on it is unrealistic. Grrr. If IPA didn't suck so hard, I'd say we should transition to phonetic transcription as a language. Perhaps a modified Shavian... -- कुक्कुरोवाच|Talk‽ 08:45, 1 Jul 2004 (UTC)

You really think five out of five people on the street know what nirvana means and can spell it? A lot of people can't even spell Buddha right. I'm not sure that I know what karma means. So your standard is pretty high. By the way, I've heard people say "prajñā" more than a few times; in fact, I recall hearing several pronunciations. - Nat Krause 09:52, 1 Jul 2004 (UTC)

Ah, my standards aren't quite as high as it seems. I meant provide *a* definition of the word, not the right one. Lot's of people could spout nonsense about karma, nirvana, or Buddha. But if people really can't spell "Buddha" (I mean lots of people who are actually thinking about it), then it clearly hasn't succeeded in truly penetrating the English language. (<-- is full of crap, since spelling is purely an artifact of written text, and language is verbal, but since we're talking about standards of writing things down, I'm not being too bad, though as far as this very sentence goes, if there's one thing worse than referring to yourself in the third person, it's going back and forth) (Can you tell I haven't slept much?). As for prajñā, I meant "pronounce it correctly." -- कुक्कुरोवाच|Talk‽ 11:10, 1 Jul 2004 (UTC)

I'm willing to follow your lead on this. It occurs to me that, given that I think of "prajna" as a word in the English language (and yes, I do pronounce it correctly, I think. Hard j, right?), perhaps I am not quite the modal man-on-the-street that I think of myself as. What do you propose to do for names? - Nat Krause 20:02, 1 Jul 2004 (UTC) - PS: I still think "sangha" is an English word. A lot of your English-speaking Buddhist nowadays groups use "sangha" basically in the sense of a congregation, a very non Sanskrit-specific usage.

What's a hard j? Anyway, I would diferentiate between words current in specialized circles (Buddhists and those who study them), where terms are technical jargon, and the population at large, where we can really talk about the English language. As for names, if it's someone who's alive, follow whatever conventions they follow, and if it's someone who's not, use diacritics. Or at least, I would. Of course, the pain in the ass is naming pages on Wikipedia, since we haven't been switched over to UTF-8. ::growls:: -- कुक्कुरोवाच|Talk‽ 20:16, 1 Jul 2004 (UTC)

(You know, hard j, like in "jargon" as opposed to soft j, like in "je suis Michael Jackson")

Well, jargon or no, sangha has an accepted English-language spelling, which is sangha. About names: you really want us to use śakyamuni? Ugh. The thing about a name is that a person's name isn't necessarily in a particular language. What about the Buddha's tribe, the śakyas? They didn't even speak Sanskrit (not that anybody did), so what's the point of using Sanskrit spelling conventions? - Nat Krause 20:25, 1 Jul 2004 (UTC)

Ah, it's not a question of Sanskrit at all. It's almost a question of Devanagari, but not quite. All Indo-Aryan languages that I know about, at least those that haven't adopted foreign scripts (like Urdu) are written with some descendent of Brahmi, such as Devanagari; these scripts all represent with almost perfect phoneticity the sounds of words. This is why we have the Pali word "Dhamma" rather than "Dharma"; it's just a dialectical pronunciation shift (we see the same thing in the Southern United States, or New England) that results in the elision of the "r" and the strengthening (lengthening, in this case) of the consonant following it. But because of Indian transcription techniques, not to mention the phonetic analysis of grammarians (both Hindu and Buddhist, working in Pali, Sanskrit, and some Prakrits), these changes have to be taken seriously. This holds for virtually any Indo-Aryan language. Also, and amusingly, one of the few exceptions to this phoneticity is jñ, which, at least as I was taught by my Sanksrit professor, in the "j" is more or less completely elided. (This probably wasn't true for some earlier stage of the langauge) -- कुक्कुरोवाच|Talk‽ 20:34, 1 Jul 2004 (UTC)

Thanks for this discussion. First of, I'm troubled by 'ch' for Sanskrit, but don't have a solution. My trouble with it is that Sanskrit ch then becomes the ugly 'chh' unless we also spell it as 'ch'. But it's probably the best of a bad set of alternatives. Here's my responses to your 5 objections, Kukkurovaca. (1) Diacritics are good spelling in Sanskrit, and I'm not suggesting we drop them from words we are giving in Sanskrit. However, I would suggest that in discussions of Buddhist texts, like Prajnaparamita, we borrow the word (even if only temporarily) into the English of the discussion, which is why we don't continue to write it in italics. As an English word, it is no longer Sanskrit, and we should strive for clear communication with English speakers (who may be more Buddhist than Orientalist, in eg the current case). I think losing the diacritics makes things easier for English speakers to read and (if they feel like it in wikipedia) edit the text. (2) The 'best transliteration' is a troubled notion, though I see the point you are making. What I am proposing is transliterating upto the phonemic distinctions available in English plus English spelling (which allows us silent extra 'h's in words like chiropractor and buddha), but sticking with the alphabet of English which excludes diacritics. This means that a non-Sanskritist seeing the transliteration should be able to make a stab at pronouncing it and not come off too badly. (3) I don't see a problem with lossy transcriptions, provided that at the first usage of the term in every wikipedia article, we give the full transcription as well. I completely agree that the non-lossy ASCII forms are ugly as sin, and would prefer to see diacritics in most cases. But giving lossless Sanskrit once is enough, I think, after which I'd like to use a best-fit English friendly form - even if it represents English's failing to distinguish between palatal (ś) and retroflex (ṣ) s's, amongst other distinctions. (4) The full Sanskrit transliteration can be used to cut and paste for searching (and other purposes), however the latter use of the anglicised form may be even more useful for that. There are 265 google finds from Prajñāpāramitā, but 40000 for Prajnaparamita. Both are needed. (5) I don't see what I propose as a new Romanisation, but rather an attempt to state the patterns that people seem to be using, preserving those aspects of Sanskrit that fit readily into English spelling. Using this model, you get: prajnaparamita, buddha, nirvana, yoga, tathagata, arhat, bodhisattva. It does choose between the almost equal (according to google) split between sunyata and shunyata, erring on being more phonologically distinctive. (BTW, I agree about using people defining the usage of their own names, but only in their own language: many Poles in English write Cracow as the name for their historical capital, which seems Victorian to me. Krakow (the Polish spelling minus the diacritic) is a much better modern English form - and lets you find it on a Polish map!).

(1) The italic or nonitalic setting is a purely typographical convention that isn't really and never has been thoroughly and systematically used. Personally, I think it's silly and we should only italicize titles of works or for emphasis. In any case, there's no such thing as "temporarily" borrowing a word into a language, and in any case individuals can't do it. I don't just decide that "Buddha" is a word in English; it has evolved into such over many years. (2) "Best translation following scholarship" is not a very troubled notion; open up (almost) any recent scholarly text in these fields, and you'll see that a certain transliteration scheme is used. However, even if we didn't use the scholarly standard, we must in any case use an adequate, meaning lossless transliteration, for the sake of accuracy. (3) I would invert your suggestions. Every word should be accurately transliterated every time, but after the first occurrence, some approximation of the word's correct pronunciation should be given in parentheses. This way people won't be in the dark about how to pronounce a word, but at the same time we'll be maintaining accuracy and completeness. (4) As it happens, "Prajnaparamita" is the name of the article, and a search for that string will turn it up. When the English wikipedia finally goes UTF-8, we can redirect from "Prajnaparamita" and also reference it as a common variant spelling/transliteration. We have a responsbility to make Wikipedia represent the best knowledge we can, and while we can certainly work to bridge the gap between it and the current level of knowledge, we shouldn't lower the bar. (5) You are proposing a new transliteration, as it isn't an existing standard, but a new attempt to "state the patterns that people seem to be using". And this comes rather close to violating the prohibition against original research. -- कुक्कुरोवाच|Talk‽ 06:48, 4 Jul 2004 (UTC)

I guess, to sum up, I'd say that diacritics get in the way in flowing English text, that a single full transcription should be enough for clarity, and giving only full transcriptions doesn't give the non-specialist reader a spelling and pronunciation for the word as an item in their language.

To sum up my reply, these aren't, by and large, words in the English language. Some are, but others, like prajñāparamitā, are certainly not. Wikipedia has a responsibility to provide the most accurate information it can, and that means a lossless transliteration. It should also be a good first step for people who want to get access to more scholarly materials on a subject, and that means a scholarly transliteration. You're absolutely right that it should tell people how to pronounce words, but it can do so without violating these other responsibilities, by specifying he pronunciation up front, as reference works often do. We can even include sound files, can't we, featuring the correct pronunciations?

BTW, I suspect, Wikipedia has as a nontrivial segment of its userbase people who don't speak English with perfect fluency. For them, the many aspects of English standard spelling that are nonphonetic may be extremely puzzling, and it's possible that we could better reach them by converting to a phonetic English spelling, but that doesn't mean we should do so. And this isn't simply because they're a minority in our community... -- कुक्कुरोवाच|Talk‽ 06:48, 4 Jul 2004 (UTC)