|
Post by stardustpilgrim on Aug 19, 2020 15:11:54 GMT -5
I chanced upon a new translation of the Tao Te Ching, 2020 by KT Fann, Reading "Dao De Jing" in English. He taught philosophy, but has studied the Dao De Jing for over 40 years. He was not happy with his own translation, 1981, and subsequent to that there have been at least 3 old manuscripts found which he has studied and used for his new translation, correcting what he believes are many errors in the latest translations. At least two of the new found copies are on bamboo, proved to be some of the earliest copies. These show that much of the later copies had many additions to the original. My favorite chapter 20 is cut in half.
He also has come to believe the original was a dialogue, with only the answers given. The traditional story is that Lao Tzu was leaving to go into retirement and the guard at the gate learned this and the reputation of Lao Tzu led the guard to ask him to write down some of the Tao. From his research Fann came to see that the guard asked a series of questions, Lao Tzu replied, and immediately afterward the dialogue was finished the guard wrote down the answers from his memory. There are about 60 pages of introduction and commentary where Fann gives reasons for his conclusions. He also gives some alternative interpretations, things he thinks Lao Tzu meant, but which he does not have a direct source. But the quotes below are from actual manuscripts. I give these quotes from the beginning of the Dao De Jing (this is closer to the actual pronunciation of the Chinese than the over 100 years old Wade-Giles) because I find it very interesting that they echo some of the language used here on ST's, therefore understanding, which you will recognize.
Chapter One
Dao can be spoken, but it's not what's commonly called Dao.
Its name can be given, but it's not what's commonly called a name.
"Nothing" is the name for the origin of Heaven and Earth.
"Thing" is the name for the Mother of all things.
Thus, always "Nothing," if you wish to observe a thing's essence.
Always "Thing," if you wish to observe a thing's appearance.
These two come from the same source but are differently named.
Both are called mysteries.
The mystery of mysteries is the gateway to all wonders.
|
|
|
Post by Reefs on Jan 8, 2021 9:51:03 GMT -5
Though it might not understood what this passage means, I value its sense of humor. A poet might invest heavily on his name, heavily on what he might call himself. He looks back on his poetry but there is no poetry (at least not where he thought he might find it). It must be that it is somewhere else. What if the tao means everybody is a poet-- and everything is a poem? It would make more sense if poetry had no limits. It would make more sense if every poet were profoundly important towards the way. That would mean the experience of poetry is an inside job and does not belong to the institutions judging it. If anything I said might be true, then maybe this gives me hope. IMO, classical Chinese is the ideal language for poets, because it is both vague and rich in meaning. The Chinese character 道 (Tao/dao) has many meanings. I remember looking it up in a huge Chinese dictionary once, several volumes and several thousand pages, and 道 was listed with, I think, at least 28 different meanings. Now, some of those meanings are just due to functionality, some others long obsolete, but that should still leave about a dozen meanings. So when you read translations of the Daodejing, they are mostly just approximations because there's no exact translation possible into any other language. There are probably several dozen translation into English alone, and some vary greatly, because the original text leaves so much room for interpretation. And there exist slightly different versions of the original text as well, because every time archeologists dig up a grave from some long vanished dynasty, they seem to dig up a different, yet unknown, version of the Daodejing and other classical texts as well. So you'd be well advised to check with several translations simultaneously. The standard translation in academic circles has usually been the Legge translation. Which I think is fairly good. Personally, I prefer the Mitchell translation. So while the language is extremely vague and open to (mis)interpretation, the message conveyed is actually astonishingly clear and presented in a very concise, almost economical way. So IMO, the Daodejing is both a beautiful work of poetry and a profound treatise on existential issues at the same time. Texts of such caliber are extremely rare. Once you get hooked, that book will keep you engaged for many years to come.
|
|
|
Post by inavalan on Jan 8, 2021 15:36:46 GMT -5
|
|
|
Post by inavalan on Jan 9, 2021 0:07:39 GMT -5
Though it might not understood what this passage means, I value its sense of humor. A poet might invest heavily on his name, heavily on what he might call himself. He looks back on his poetry but there is no poetry (at least not where he thought he might find it). It must be that it is somewhere else. What if the tao means everybody is a poet-- and everything is a poem? It would make more sense if poetry had no limits. It would make more sense if every poet were profoundly important towards the way. That would mean the experience of poetry is an inside job and does not belong to the institutions judging it. If anything I said might be true, then maybe this gives me hope. IMO, classical Chinese is the ideal language for poets, because it is both vague and rich in meaning. The Chinese character 道 (Tao/dao) has many meanings. I remember looking it up in a huge Chinese dictionary once, several volumes and several thousand pages, and 道 was listed with, I think, at least 28 different meanings. Now, some of those meanings are just due to functionality, some others long obsolete, but that should still leave about a dozen meanings. So when you read translations of the Daodejing, they are mostly just approximations because there's no exact translation possible into any other language. There are probably several dozen translation into English alone, and some vary greatly, because the original text leaves so much room for interpretation. And there exist slightly different versions of the original text as well, because every time archeologists dig up a grave from some long vanished dynasty, they seem to dig up a different, yet unknown, version of the Daodejing and other classical texts as well. So you'd be well advised to check with several translations simultaneously. The standard translation in academic circles has usually been the Legge translation. Which I think is fairly good. Personally, I prefer the Mitchell translation. So while the language is extremely vague and open to (mis)interpretation, the message conveyed is actually astonishingly clear and presented in a very concise, almost economical way. So IMO, the Daodejing is both a beautiful work of poetry and a profound treatise on existential issues at the same time. Texts of such caliber are extremely rare. Once you get hooked, that book will keep you engaged for many years to come. Reefs , do you speak / read Chinese? I browsed the first lines of several translations of the Daodejing today, and I was surprised by the differences between them. Then, out of curiosity I ran a couple of machine translations Chinese to English, that seemed far from all the authoritative translations. That made me curious to look closer to the kanjis of the first four lines. 道可道, 非常道。 名可名, 非常名。The structure seems pretty simple: " X1 A X2, B1 B2 X3. Y1 A Y2, B1 B2 Y3." where X is Dao (road, path, way), Y is Ming (name, reputation, ranking, also counter for people), A is Ke (can, be able, certainly), B1 is Fei (prefix for negation), B2 is Chang (always, ever, often, constant), "B1B2" is FeiChang (very much) In Japanese (I have a moderate mastering of the language), the verb is always the last in the sentence (not considering particles), and the qualifiers precede the qualified word. Trying to put this information together, I'm quite far of any translations I've seen. Would you please explain how does the translation of this small paragraph goes? Thanks To me the structure would be something like " an X2 that is able to do like an/the X1, is very much an/the X3" (all Xs are the same word, but the # indicates the position in this Chinese sentence); X being the Dao, the way; Y being a thing, a name. Or: " A way that looks like the Way, is very much the Way. As/because: a duck that quacks like a duck, is very much a duck". "Very much" here being used with the sense of "the real", or such. Surely, B2 can be grouped with X3, and Y3, B2X3 and B2Y3 meaning something like "common X/Y)", which with B1 would be translated as " not a common X/Y" (which is closer to the authoritative translations, and to me makes less sense), but that wouldn't be the usual Japanese topic structure; B1 being a prefix, not a word in itself, should qualify only the B2.
|
|
|
Post by Reefs on Jan 9, 2021 12:49:46 GMT -5
IMO, classical Chinese is the ideal language for poets, because it is both vague and rich in meaning. The Chinese character 道 (Tao/dao) has many meanings. I remember looking it up in a huge Chinese dictionary once, several volumes and several thousand pages, and 道 was listed with, I think, at least 28 different meanings. Now, some of those meanings are just due to functionality, some others long obsolete, but that should still leave about a dozen meanings. So when you read translations of the Daodejing, they are mostly just approximations because there's no exact translation possible into any other language. There are probably several dozen translation into English alone, and some vary greatly, because the original text leaves so much room for interpretation. And there exist slightly different versions of the original text as well, because every time archeologists dig up a grave from some long vanished dynasty, they seem to dig up a different, yet unknown, version of the Daodejing and other classical texts as well. So you'd be well advised to check with several translations simultaneously. The standard translation in academic circles has usually been the Legge translation. Which I think is fairly good. Personally, I prefer the Mitchell translation. So while the language is extremely vague and open to (mis)interpretation, the message conveyed is actually astonishingly clear and presented in a very concise, almost economical way. So IMO, the Daodejing is both a beautiful work of poetry and a profound treatise on existential issues at the same time. Texts of such caliber are extremely rare. Once you get hooked, that book will keep you engaged for many years to come. Reefs , do you speak / read Chinese? I browsed the first lines of several translations of the Daodejing today, and I was surprised by the differences between them. Then, out of curiosity I ran a couple of machine translations Chinese to English, that seemed far from all the authoritative translations. That made me curious to look closer to the kanjis of the first four lines. 道可道, 非常道。 名可名, 非常名。The structure seems pretty simple: " X1 A X2, B1 B2 X3. Y1 A Y2, B1 B2 Y3." where X is Dao (road, path, way), Y is Ming (name, reputation, ranking, also counter for people), A is Ke (can, be able, certainly), B1 is Fei (prefix for negation), B2 is Chang (always, ever, often, constant), "B1B2" is FeiChang (very much) In Japanese (I have a moderate mastering of the language), the verb is always the last in the sentence (not considering particles), and the qualifiers precede the qualified word. Trying to put this information together, I'm quite far of any translations I've seen. Would you please explain how does the translation of this small paragraph goes? Thanks To me the structure would be something like " an X2 that is able to do like an/the X1, is very much an/the X3" (all Xs are the same word, but the # indicates the position in this Chinese sentence); X being the Dao, the way; Y being a thing, a name. Or: " A way that looks like the Way, is very much the Way. As/because: a duck that quacks like a duck, is very much a duck". "Very much" here being used with the sense of "the real", or such. Surely, B2 can be grouped with X3, and Y3, B2X3 and B2Y3 meaning something like "common X/Y)", which with B1 would be translated as " not a common X/Y" (which is closer to the authoritative translations, and to me makes less sense), but that wouldn't be the usual Japanese topic structure; B1 being a prefix, not a word in itself, should qualify only the B2. Yes, I do. There's a big difference between classical Chinese and modern Chinese, both in grammar and vocabulary. In classical Chinese, one character equals one word (usually), in modern Chinese, a lot of words consist of two or more characters these days. For example: 非常 (feichang) means 'extremely' in modern Chinese, as in 'extremely good' 非常好 (feichang hao) or 'extremely beautiful' 非常漂亮(feichang piaoliang). Now, here you see that 'extremely beautiful' consists of 4 characters already. Classical Chinese is a lot more economical. Here 非常 (fei chang) are two words, not one word, i.e. instead of 'extremely' it means 'not eternal'. So, if for modern usage words you go look them up character by character and translate them character by character, you'll get the meaning wrong. And if for classical characters, you read them like modern words, i.e. several characters combined equal one word instead of one character equals one word, you'll get the meaning wrong again. So before you read a text, you have to know what kind of text you are dealing with, classical or modern. When you study Chinese, you will come across two basic kinds of dictionaries, the regular dictionaries or 词典(cidian) which contain all the contemporary words, and the character dictionaries or 字典 (zidian) which only list individual characters, not words. Now, the 28+ meanings of the character 道 (dao) you will not find in a cidian, you can only find them in a zidian. And it is a zidian that you need if you want to translate texts written in classical Chinese. Because the cidian may not contain many of the other meanings that are now obsolete. So if you copy and paste a classical text into google translate, what google will give you is the modern reading. Which almost certainly will be either incomplete, incomprehensible, absolutely ridiculous or completely wrong. Machine translations for western languages are pretty good these days, but for some Asian languages like Chinese that rely heavily on context, apart from pretty standard stuff, it's basically worthless. I'll get to your example later. That will require a bit more explaining. Just wanted to give you a general outline.
|
|
|
Post by inavalan on Jan 9, 2021 15:37:11 GMT -5
Reefs , do you speak / read Chinese? I browsed the first lines of several translations of the Daodejing today, and I was surprised by the differences between them. Then, out of curiosity I ran a couple of machine translations Chinese to English, that seemed far from all the authoritative translations. That made me curious to look closer to the kanjis of the first four lines. 道可道, 非常道。 名可名, 非常名。... Yes, I do. There's a big difference between classical Chinese and modern Chinese, both in grammar and vocabulary. In classical Chinese, one character equals one word (usually), in modern Chinese, a lot of words consist of two or more characters these days. For example: 非常 (feichang) means 'extremely' in modern Chinese, as in 'extremely good' 非常好 (feichang hao) or 'extremely beautiful' 非常漂亮(feichang piaoliang). Now, here you see that 'extremely beautiful' consists of 4 characters already. Classical Chinese is a lot more economical. Here 非常 (fei chang) are two words, not one word, i.e. instead of 'extremely' it means 'not eternal'. So, if for modern usage words you go look them up character by character and translate them character by character, you'll get the meaning wrong. And if for classical characters, you read them like modern words, i.e. several characters combined equal one word instead of one character equals one word, you'll get the meaning wrong again. So before you read a text, you have to know what kind of text you are dealing with, classical or modern. When you study Chinese, you will come across two basic kinds of dictionaries, the regular dictionaries or 词典(cidian) which contain all the contemporary words, and the character dictionaries or 字典 (zidian) which only list individual characters, not words. Now, the 28+ meanings of the character 道 (dao) you will not find in a cidian, you can only find them in a zidian. And it is a zidian that you need if you want to translate texts written in classical Chinese. Because the cidian may not contain many of the other meanings that are now obsolete. So if you copy and paste a classical text into google translate, what google will give you is the modern reading. Which almost certainly will be either incomplete, incomprehensible, absolutely ridiculous or completely wrong. Machine translations for western languages are pretty good these days, but for some Asian languages like Chinese that rely heavily on context, apart from pretty standard stuff, it's basically worthless. I'll get to your example later. That will require a bit more explaining. Just wanted to give you a general outline. Thank you. Quite informative!
|
|
|
Post by Reefs on Jan 11, 2021 21:55:46 GMT -5
IMO, classical Chinese is the ideal language for poets, because it is both vague and rich in meaning. The Chinese character 道 (Tao/dao) has many meanings. I remember looking it up in a huge Chinese dictionary once, several volumes and several thousand pages, and 道 was listed with, I think, at least 28 different meanings. Now, some of those meanings are just due to functionality, some others long obsolete, but that should still leave about a dozen meanings. So when you read translations of the Daodejing, they are mostly just approximations because there's no exact translation possible into any other language. There are probably several dozen translation into English alone, and some vary greatly, because the original text leaves so much room for interpretation. And there exist slightly different versions of the original text as well, because every time archeologists dig up a grave from some long vanished dynasty, they seem to dig up a different, yet unknown, version of the Daodejing and other classical texts as well. So you'd be well advised to check with several translations simultaneously. The standard translation in academic circles has usually been the Legge translation. Which I think is fairly good. Personally, I prefer the Mitchell translation. So while the language is extremely vague and open to (mis)interpretation, the message conveyed is actually astonishingly clear and presented in a very concise, almost economical way. So IMO, the Daodejing is both a beautiful work of poetry and a profound treatise on existential issues at the same time. Texts of such caliber are extremely rare. Once you get hooked, that book will keep you engaged for many years to come. Reefs , do you speak / read Chinese? I browsed the first lines of several translations of the Daodejing today, and I was surprised by the differences between them. Then, out of curiosity I ran a couple of machine translations Chinese to English, that seemed far from all the authoritative translations. That made me curious to look closer to the kanjis of the first four lines. 道可道, 非常道。 名可名, 非常名。The structure seems pretty simple: " X1 A X2, B1 B2 X3. Y1 A Y2, B1 B2 Y3." where X is Dao (road, path, way), Y is Ming (name, reputation, ranking, also counter for people), A is Ke (can, be able, certainly), B1 is Fei (prefix for negation), B2 is Chang (always, ever, often, constant), "B1B2" is FeiChang (very much) In Japanese (I have a moderate mastering of the language), the verb is always the last in the sentence (not considering particles), and the qualifiers precede the qualified word. Trying to put this information together, I'm quite far of any translations I've seen. Would you please explain how does the translation of this small paragraph goes? Thanks To me the structure would be something like " an X2 that is able to do like an/the X1, is very much an/the X3" (all Xs are the same word, but the # indicates the position in this Chinese sentence); X being the Dao, the way; Y being a thing, a name. Or: " A way that looks like the Way, is very much the Way. As/because: a duck that quacks like a duck, is very much a duck". "Very much" here being used with the sense of "the real", or such. Surely, B2 can be grouped with X3, and Y3, B2X3 and B2Y3 meaning something like "common X/Y)", which with B1 would be translated as " not a common X/Y" (which is closer to the authoritative translations, and to me makes less sense), but that wouldn't be the usual Japanese topic structure; B1 being a prefix, not a word in itself, should qualify only the B2. Okay, let's get to your example, I'll keep it short. In modern Chinese, word order certainly matters. Although, Chinese, as English, is a very flexible language (classical Chinese even more so). But this also comes at the price of precision and clarity. So this is where context becomes important. Context will determine if the word you are looking at is a verb, a noun or an adjective or a preposition or just a function word like indicating a comma or the end of a sentence, because they usually have no punctuation marks in classical Chinese (modern editions have them though), it's just a wall of characters, and some characters function as punctuation marks. So if they don't put it neatly into verses or lines, you have to first figure out where the sentence starts and where it ends. Now, the context of the Daodejing is obviously Daoism. And another established classical meaning of the character 道 is 'to speak', you'll see the character 道 quite often in other classical texts with that meaning. So this should already significantly narrow down our options for possible translations: 道 - 1) Tao 2) way 3) speak 可 - 2) can, able; possible 非 - 1) negation particle 常 - 1) lasting, eternal 名 - 1) name 2) naming So, IMO, the standard translation is still the best in preserving both the meaning and the poetry at the same time: The Tao that can be spoken of is not the eternal Tao The name that can be named is not the eternal nameAnd this is sometimes a challenge for translators, depending on your target language, you may be forced to decide to either preserve the beauty of the poem at the expense of clarity, or you decide for maximum clarity but then can't preserve the beauty of the poem, because the target language doesn't allow you to do both. So if we would go just for maximum clarity of meaning, then I'd summarize it like this: The Tao (Truth) can't be conceptualized, and whatever is conceptual, is not the Tao (Truth).Another problem with translating mystical works is that the translator should have a good grasp of the subject at hand and an actual reference for what the text is trying to point to. And that's obviously not always the case. Which I think might be the main reason why translations of the Daodejing vary greatly. Another reason might be that some translators didn't go back to the original Chinese text and used the Japanese version and also went with the gist of the Japanese commentaries, which seems to have been very common in the past. And given that one has that reference, the Daodejing isn't that difficult to translate, actually. It's Chinese philosophy with Chinese concepts and Chinese words. What I find a lot more difficult to translate are early Chan (Zen) texts like the Blue Cliff Record because it's not Chinese philosophy, not Chinese words but translations or transliterations into Chinese, using words that have their roots in Daoism. So that's where you actually need a special dictionary of Buddhist terms in Chinese. I only realized that recently. Just for fun, if you're in the mood, try to translate this: 廓然無聖That's what Bodhidharma replied when the Emperor asked him about the highest meaning of the holy truth. And this: 不識That's what Bodhidharma replied when the Emperor asked him, Who are you? I'll post more about that at some later point when I have more time, in a separate thread about Koans. Turns out, some common translations are debatable, some even wrong. And some Koans just can't be understood without understanding the actual background story and/or Buddhist works they are alluding to or referencing. Fascinating stuff.
|
|
|
Post by inavalan on Jan 12, 2021 0:40:18 GMT -5
Reefs , do you speak / read Chinese? I browsed the first lines of several translations of the Daodejing today, and I was surprised by the differences between them. Then, out of curiosity I ran a couple of machine translations Chinese to English, that seemed far from all the authoritative translations. That made me curious to look closer to the kanjis of the first four lines. 道可道, 非常道。 名可名, 非常名。The structure seems pretty simple: " X1 A X2, B1 B2 X3. Y1 A Y2, B1 B2 Y3." where X is Dao (road, path, way), Y is Ming (name, reputation, ranking, also counter for people), A is Ke (can, be able, certainly), B1 is Fei (prefix for negation), B2 is Chang (always, ever, often, constant), "B1B2" is FeiChang (very much) In Japanese (I have a moderate mastering of the language), the verb is always the last in the sentence (not considering particles), and the qualifiers precede the qualified word. Trying to put this information together, I'm quite far of any translations I've seen. Would you please explain how does the translation of this small paragraph goes? Thanks To me the structure would be something like " an X2 that is able to do like an/the X1, is very much an/the X3" (all Xs are the same word, but the # indicates the position in this Chinese sentence); X being the Dao, the way; Y being a thing, a name. Or: " A way that looks like the Way, is very much the Way. As/because: a duck that quacks like a duck, is very much a duck". "Very much" here being used with the sense of "the real", or such. Surely, B2 can be grouped with X3, and Y3, B2X3 and B2Y3 meaning something like "common X/Y)", which with B1 would be translated as " not a common X/Y" (which is closer to the authoritative translations, and to me makes less sense), but that wouldn't be the usual Japanese topic structure; B1 being a prefix, not a word in itself, should qualify only the B2. Okay, let's get to your example, I'll keep it short. In modern Chinese, word order certainly matters. Although, Chinese, as English, is a very flexible language (classical Chinese even more so). But this also comes at the price of precision and clarity. So this is where context becomes important. Context will determine if the word you are looking at is a verb, a noun or an adjective or a preposition or just a function word like indicating a comma or the end of a sentence, because they usually have no punctuation marks in classical Chinese (modern editions have them though), it's just a wall of characters, and some characters function as punctuation marks. So if they don't put it neatly into verses or lines, you have to first figure out where the sentence starts and where it ends. Now, the context of the Daodejing is obviously Daoism. And another established classical meaning of the character 道 is 'to speak', you'll see the character 道 quite often in other classical texts with that meaning. So this should already significantly narrow down our options for possible translations: 道 - 1) Tao 2) way 3) speak 可 - 2) can, able; possible 非 - 1) negation particle 常 - 1) lasting, eternal 名 - 1) name 2) naming So, IMO, the standard translation is still the best in preserving both the meaning and the poetry at the same time: The Tao that can be spoken of is not the eternal Tao The name that can be named is not the eternal nameAnd this is sometimes a challenge for translators, depending on your target language, you may be forced to decide to either preserve the beauty of the poem at the expense of clarity, or you decide for maximum clarity but then can't preserve the beauty of the poem, because the target language doesn't allow you to do both. So if we would go just for maximum clarity of meaning, then I'd summarize it like this: The Tao (Truth) can't be conceptualized, and whatever is conceptual, is not the Tao (Truth).Another problem with translating mystical works is that the translator should have a good grasp of the subject at hand and an actual reference for what the text is trying to point to. And that's obviously not always the case. Which I think might be the main reason why translations of the Daodejing vary greatly. Another reason might be that some translators didn't go back to the original Chinese text and used the Japanese version and also went with the gist of the Japanese commentaries, which seems to have been very common in the past. And given that one has that reference, the Daodejing isn't that difficult to translate, actually. It's Chinese philosophy with Chinese concepts and Chinese words. What I find a lot more difficult to translate are early Chan (Zen) texts like the Blue Cliff Record because it's not Chinese philosophy, not Chinese words but translations or transliterations into Chinese, using words that have their roots in Daoism. So that's where you actually need a special dictionary of Buddhist terms in Chinese. I only realized that recently. Just for fun, if you're in the mood, try to translate this: 廓然無聖That's what Bodhidharma replied when the Emperor asked him about the highest meaning of the holy truth. And this: 不識That's what Bodhidharma replied when the Emperor asked him, Who are you? I'll post more about that at some later point when I have more time, in a separate thread about Koans. Turns out, some common translations are debatable, some even wrong. And some Koans just can't be understood without understanding the actual background story and/or Buddhist works they are alluding to or referencing. Fascinating stuff. Thanks for taking time to explain. This is quite fascinating. I'll probably be way off, so I feel no pressure to be right
That's what Bodhidharma replied when the Emperor asked him about the highest meaning of the holy truth: 廓然無聖 廓 - big, empty, open 然 - correct, like this, -ly 廓然 - awesome 無 - no, not, -less 聖 - holy; sacred; saint; sage My starting point would be: holy is less impressive; meaning: the holy truth is simplicity.
That's what Bodhidharma replied when the Emperor asked him, Who are you? 不識 不 - negative prefix, not, no 識 - knowledge, to recognize My starting point would be: an ignorant; meaning: somebody who knows nothing
|
|
|
Post by Reefs on Jan 12, 2021 10:27:45 GMT -5
Thanks for taking time to explain. This is quite fascinating. I'll probably be way off, so I feel no pressure to be right
That's what Bodhidharma replied when the Emperor asked him about the highest meaning of the holy truth: 廓然無聖 廓 - big, empty, open 然 - correct, like this, -ly 廓然 - awesome 無 - no, not, -less 聖 - holy; sacred; saint; sage My starting point would be: holy is less impressive; meaning: the holy truth is simplicity.
That's what Bodhidharma replied when the Emperor asked him, Who are you? 不識 不 - negative prefix, not, no 識 - knowledge, to recognize My starting point would be: an ignorant; meaning: somebody who knows nothing You're actually pretty close with the first example. It is usually translated as "Open vastness, nothing holy" or some variation of it. Cleary translates it as "Empty, without holiness". Both translations are okay, IMO. But 廓然 seems to be a standard term, meaning 'emptiness' in Buddhist texts. Although I actually like 'open vastness' better, it's a lot more poetic and less abstract than 'emptiness'. There's something I forgot to mention earlier. There are different particles used for negation, depending on what kind of word you've got - a verb, a noun or an adjective. And since the text uses 不, then 識 would be better translated as a verb. So the usual translation is "I don't know" even though the personal pronoun is omitted, which is actually quite common in conversational Chinese when the context is clear. But is the context really clear here? That would be a good question, because if it isn't, then there are other possible translations, which would change the meaning of the dialog of this famous koan significantly. And different translators and commentators did indeed translate it differently. I don't have a clear answer so far, I'm not a philologist, but it did surprise me when I first looked at the original text and noticed that it was actually that vague and open to interpretation. So the standard translation is at least debatable. I'll go into more details in the koan thread.
|
|
|
Post by inavalan on Jan 12, 2021 14:36:06 GMT -5
Thanks for taking time to explain. This is quite fascinating. I'll probably be way off, so I feel no pressure to be right
That's what Bodhidharma replied when the Emperor asked him about the highest meaning of the holy truth: 廓然無聖 廓 - big, empty, open 然 - correct, like this, -ly 廓然 - awesome 無 - no, not, -less 聖 - holy; sacred; saint; sage My starting point would be: holy is less impressive; meaning: the holy truth is simplicity.
That's what Bodhidharma replied when the Emperor asked him, Who are you? 不識 不 - negative prefix, not, no 識 - knowledge, to recognize My starting point would be: an ignorant; meaning: somebody who knows nothing You're actually pretty close with the first example. It is usually translated as "Open vastness, nothing holy" or some variation of it. Cleary translates it as "Empty, without holiness". Both translations are okay, IMO. But 廓然 seems to be a standard term, meaning 'emptiness' in Buddhist texts. Although I actually like 'open vastness' better, it's a lot more poetic and less abstract than 'emptiness'. There's something I forgot to mention earlier. There are different particles used for negation, depending on what kind of word you've got - a verb, a noun or an adjective. And since the text uses 不, then 識 would be better translated as a verb. So the usual translation is "I don't know" even though the personal pronoun is omitted, which is actually quite common in conversational Chinese when the context is clear. But is the context really clear here? That would be a good question, because if it isn't, then there are other possible translations, which would change the meaning of the dialog of this famous koan significantly. And different translators and commentators did indeed translate it differently. I don't have a clear answer so far, I'm not a philologist, but it did surprise me when I first looked at the original text and noticed that it was actually that vague and open to interpretation. So the standard translation is at least debatable. I'll go into more details in the koan thread. Thanks. This is really interesting. The multiple possibilities that need to be filtered through context, show how difficult is to get the original meaning. I see an advantage in using Chinese characters vs. an alphabet, as they may better convey the spirit of a word. On the other hand, it is likely that the spirit of a character changed over the years. It surprised me too to find out out that reading a translation can lead you away in such a degree from the original meaning. It happens the same if you read Nostradsmus, the Bible, or any other translation.
|
|
|
Post by inavalan on Jan 12, 2021 15:07:19 GMT -5
Thanks for taking time to explain. This is quite fascinating. I'll probably be way off, so I feel no pressure to be right
That's what Bodhidharma replied when the Emperor asked him about the highest meaning of the holy truth: 廓然無聖 廓 - big, empty, open 然 - correct, like this, -ly 廓然 - awesome 無 - no, not, -less 聖 - holy; sacred; saint; sage My starting point would be: holy is less impressive; meaning: the holy truth is simplicity.
That's what Bodhidharma replied when the Emperor asked him, Who are you? 不識 不 - negative prefix, not, no 識 - knowledge, to recognize My starting point would be: an ignorant; meaning: somebody who knows nothing You're actually pretty close with the first example. It is usually translated as "Open vastness, nothing holy" or some variation of it. Cleary translates it as "Empty, without holiness". Both translations are okay, IMO. But 廓然 seems to be a standard term, meaning 'emptiness' in Buddhist texts. Although I actually like 'open vastness' better, it's a lot more poetic and less abstract than 'emptiness'. There's something I forgot to mention earlier. There are different particles used for negation, depending on what kind of word you've got - a verb, a noun or an adjective. And since the text uses 不, then 識 would be better translated as a verb. So the usual translation is "I don't know" even though the personal pronoun is omitted, which is actually quite common in conversational Chinese when the context is clear. But is the context really clear here? That would be a good question, because if it isn't, then there are other possible translations, which would change the meaning of the dialog of this famous koan significantly. And different translators and commentators did indeed translate it differently. I don't have a clear answer so far, I'm not a philologist, but it did surprise me when I first looked at the original text and noticed that it was actually that vague and open to interpretation. So the standard translation is at least debatable. I'll go into more details in the koan thread. I found this interesting conversation: "English Translation; Bodhidharma's reply to Emperor Wu "不識"" https://www.reddit.com/r/zen/comments/2s1nzc/english_translation_bodhidharmas_reply_to_emperor/
|
|
|
Post by laughter on Jan 12, 2021 16:43:39 GMT -5
Thanks for taking time to explain. This is quite fascinating. I'll probably be way off, so I feel no pressure to be right
That's what Bodhidharma replied when the Emperor asked him about the highest meaning of the holy truth: 廓然無聖 廓 - big, empty, open 然 - correct, like this, -ly 廓然 - awesome 無 - no, not, -less 聖 - holy; sacred; saint; sage My starting point would be: holy is less impressive; meaning: the holy truth is simplicity.
That's what Bodhidharma replied when the Emperor asked him, Who are you? 不識 不 - negative prefix, not, no 識 - knowledge, to recognize My starting point would be: an ignorant; meaning: somebody who knows nothing You're actually pretty close with the first example. It is usually translated as "Open vastness, nothing holy" or some variation of it. Cleary translates it as "Empty, without holiness". Both translations are okay, IMO. But 廓然 seems to be a standard term, meaning 'emptiness' in Buddhist texts. Although I actually like 'open vastness' better, it's a lot more poetic and less abstract than 'emptiness'. There's something I forgot to mention earlier. There are different particles used for negation, depending on what kind of word you've got - a verb, a noun or an adjective. And since the text uses 不, then 識 would be better translated as a verb. So the usual translation is "I don't know" even though the personal pronoun is omitted, which is actually quite common in conversational Chinese when the context is clear. But is the context really clear here? That would be a good question, because if it isn't, then there are other possible translations, which would change the meaning of the dialog of this famous koan significantly. And different translators and commentators did indeed translate it differently. I don't have a clear answer so far, I'm not a philologist, but it did surprise me when I first looked at the original text and noticed that it was actually that vague and open to interpretation. So the standard translation is at least debatable. I'll go into more details in the koan thread. The punchline to the "Heart Sutra" is called the "Heart Sutra" because it's not the "debate with people and figure stuff out sutra". But, it strikes me as rather sweetly ironic that it sort of succinctly describes the new-age interpretations of the "Quantum Observer" in terms of "Consciousness", with "Consciousness" == "Emptiness". And, the "Copenhagen Interpretation" can be understood as prior-to the new-age metaphysics that followed, as a statement of an absence of knowledge. Neils Bohr included a Yin/yang symbol on his Nobel coat-of-arms.
|
|
|
Post by Reefs on Jan 13, 2021 9:07:06 GMT -5
Thanks. This is really interesting. The multiple possibilities that need to be filtered through context, show how difficult is to get the original meaning. I see an advantage in using Chinese characters vs. an alphabet, as they may better convey the spirit of a word. On the other hand, it is likely that the spirit of a character changed over the years. It surprised me too to find out out that reading a translation can lead you away in such a degree from the original meaning. It happens the same if you read Nostradsmus, the Bible, or any other translation. If you learn a new language, you always also learn a new culture, and with a new culture, you also get a new perspective on life and the world. If you don't get the culture, you don't get the language and its nuances. And if you don't get the language, you don't get the culture, only an abstraction of the culture. So you can't really separate the two. The way I got into Chinese was via the culture, Daoism and Chan/Zen in particular. Because it always seemed to me that westerners who got into Daoism or Zen but didn't know the language were at a disadvantage, they had to rely on translators and commentators and while that may look like the easy way, if the goal is true understanding and mastery, it actually is the hard way. Because translators and commentators all have their own bias and gaps in understanding. And they will pass that on to future generations without anyone noticing. And once an erroneous view or interpretation becomes the standard, it's hard to correct that. In general, I wouldn't say that characters really changed meaning (although there may be some), it's rather that some meanings have become obsolete or priorities changed in the listing of meanings. What has changed significantly over time, however, is the shape of characters. The oldest characters almost look like hieroglyphs, the character for 'heart' really looks like a simple sketch of a heart, or the character for 'bird' or 'fish' really looks like a simple sketch of a bird or a fish. So that, in theory, even if you don't speak the language, you at least could read it. And even when the characters got standardizes at some point, the characters that were pictograms remained highly detailed. But over time, they became less and less detailed. There are also different variants of one and the same character in terms of stroke count (sometimes up to 3 or 5 different versions) and that's where it really gets tricky for researchers who dig up graves from ancient times and find remnants of classical texts there. That's how they found different versions of the Daodejing with slightly different characters and also chapter order reversed or no order whatsoever. So even before the translators get their hands on the 'original' text, there have already been people playing around with that text by either assigning meaning to characters previously unknown or by just putting verses into a specific order. So You probably already know that in mainland China they use simplified characters now, but outside of mainland China, they still use traditional characters. And to me it seems switching to simplified characters was just part of this trend of the characters becoming more and more abstract over time. Using simplified characters for writing is a bit like writing shorthand. It doesn't look as elegant as the traditional characters, but it does the job by saving you a lot of time and ink. ETA: There are some local Chinese dialects that actually use a different set of characters. In Cantonese they use a number of special characters that you can't find in a regular dictionary and you won't be able to type them with the standard Chinese input method on your computer either. So if you don't understand Cantonese, you won't be able to make sense of it. Just some fun examples (first one Cantonese, second one Mandarin) 乜野 (matje) = 什么 (shenme) [which, what] 冇 (mou) = 没有 (mei you) [don't have] 唔 (ng) = 不 (bu) [not, no] Also, Cantonese is interesting because it seems much closer to classical Chinese than standard Mandarin. Where standard Mandarin would use two characters for one word, Cantonese often still uses only one character for one word, for example: very beautiful in Mandarin would be 很漂亮 (hen piaoliang) and in Cantonese just 好靚 (hou loeng).
|
|
|
Post by Reefs on Jan 13, 2021 10:06:36 GMT -5
You're actually pretty close with the first example. It is usually translated as "Open vastness, nothing holy" or some variation of it. Cleary translates it as "Empty, without holiness". Both translations are okay, IMO. But 廓然 seems to be a standard term, meaning 'emptiness' in Buddhist texts. Although I actually like 'open vastness' better, it's a lot more poetic and less abstract than 'emptiness'. There's something I forgot to mention earlier. There are different particles used for negation, depending on what kind of word you've got - a verb, a noun or an adjective. And since the text uses 不, then 識 would be better translated as a verb. So the usual translation is "I don't know" even though the personal pronoun is omitted, which is actually quite common in conversational Chinese when the context is clear. But is the context really clear here? That would be a good question, because if it isn't, then there are other possible translations, which would change the meaning of the dialog of this famous koan significantly. And different translators and commentators did indeed translate it differently. I don't have a clear answer so far, I'm not a philologist, but it did surprise me when I first looked at the original text and noticed that it was actually that vague and open to interpretation. So the standard translation is at least debatable. I'll go into more details in the koan thread. I found this interesting conversation: "English Translation; Bodhidharma's reply to Emperor Wu "不識"" https://www.reddit.com/r/zen/comments/2s1nzc/english_translation_bodhidharmas_reply_to_emperor/Yes, they've raised the same issue. Although they seem to rely on the Japanese text/commentaries. And it seems even they can't really agree. The thing here is that in classical Chinese if the context is clear, words that are not needed will likely be left out. So even in modern Chinese, if you are were just talking to me and no one else there, and you ask me: "Who are you?" then I could reply with 我不知道 "I don't know" or just 不知道 "Don't know" which would be exactly the same in terms of meaning. The problem with that koan example is that there are not just 2 people, but 3 people present: the emperor, Bodhidharma and duke Zhi. So if you would ask me "Who are you?" and instead of looking at you, I would look at the other person in the room and reply "don't know" , theoretically, it could also mean "Ah, s/he doesn't know". I'll go into more details in the koan thread.
|
|
|
Post by Reefs on Jan 13, 2021 10:12:23 GMT -5
The punchline to the "Heart Sutra" is called the "Heart Sutra" because it's not the "debate with people and figure stuff out sutra". But, it strikes me as rather sweetly ironic that it sort of succinctly describes the new-age interpretations of the "Quantum Observer" in terms of "Consciousness", with "Consciousness" == "Emptiness". And, the "Copenhagen Interpretation" can be understood as prior-to the new-age metaphysics that followed, as a statement of an absence of knowledge. Neils Bohr included a Yin/yang symbol on his Nobel coat-of-arms. Yes, it should dovetail nicely with the Laozi interpretation of the Quantum Tao.
|
|