Victor Mair at the Log posts about what sounds like a very stimulating book, Mareshi Saito’s Kanbunmyaku: The Literary Sinitic Context and the Birth of Modern Japanese Language and Literature. From the author’s Introduction:

The chief aim of this book is to consider the language space of modern Japan from the perspective of what I am calling kanbunmyaku 漢文脈 in Japanese, translated here as “Literary Sinitic Context.” I use the term “Literary Sinitic” to designate what is often referred to as “Classical Chinese” or “Literary Chinese” in English, wenyan 文言 in Mandarin Chinese, kanbun 漢文 in Japanese (sometimes referred to as “Sino-Japanese” in English), and hanmun 漢文 in Korean. The Context in Literary Sinitic Context translates the -myaku of kanbunmyaku, and usually implies a pulse, vein, flow, or path, but is also the second constituent element of the Sino-Japanese term bunmyaku 文脈 meaning “(textual, literary) context.” I use the term Literary Sinitic Context to encompass both Literary Sinitic proper, as well as orthographic and literary styles (buntai 文体) derived from Literary Sinitic, such as glossed reading (kundoku 訓読) or Literary Japanese (bungobun 文語文), which mix sinographs (kanji 漢字, i.e., “Chinese” characters) and katakana. In addition to styles I also consider Literary Sinitic thought and sensibility at the core of which lie Literary Sinitic poetry (kanshi 漢詩) and prose (kanbun 漢文), collectively termed kanshibun 漢詩文.

From the publisher’s blurb:

Saito Mareshi demonstrates the centrality of Literary Sinitic poetry and prose in the creation of modern literary Japanese. Saito’s new understanding of the role of “kanbunmyaku” in the formation of Japanese literary modernity challenges dominant narratives tied to translations from modern Western literatures and problematizes the antagonism between Literary Sinitic and Japanese in the modern academy. Saito shows how kundoku (vernacular reading) and its rhythms were central to the rise of new inscriptional styles, charts the changing relationship of modern poets and novelists to kanbunmyaku, and concludes that the chronotope of modern Japan was based in a language world supported by the Literary Sinitic Context.

(Minor gripe: I don’t see the point of italicizing kanji and katakana, which are perfectly good English words, when the poor reader is already faced with a slew of genuinely foreign italicized terms.) I’m fascinated by this stuff, and I hope readers who know East Asian literatures and their history will have things to say.


  1. Cool stuff! One thing I do almost every morning is listen to a recording of the Heart Sutra in Sino-Japanese, sometimes trying to follow along with the characters. I often think I could happily spend the next several lifetimes learning enough Sanskrit, Chinese, and Japanese (among many other languages) to follow the course of Buddhist concepts across the Silk Road, and to study all the translations and mutations certain terms and characters went through.

    One thjng that’s struck me in my very early investigations in this area is that the Roman transliterations of Japanese and Korean versions of many Buddhist terms seem way closer to the sounds of the Sanskrit or Pali than the intermediary Chinese. I guess Chinese has gone through more drastic sound changes than the other languages – or the Roman transliteration system is just not doing it?

    example picked by leafing through the Princeton Dictionary of Buddhism:

    Rādha. (C. Luotuo; J. Rada; K. Rada 羅陀).

  2. I guess Chinese has gone through more drastic sound changes than the other languages

    You betcha. It’s the Irish of East Asian languages.

  3. January First-of-May says:

    I guess Chinese has gone through more drastic sound changes than the other languages

    IIRC, that’s indeed pretty much it. For that matter, Chinese has gone through different sound changes; there are probably other cases where the Chinese is closer because the Japanese and/or Korean form was significantly changed.

    There are comparable examples of, e.g., Germanic borrowings in Slavic and/or Finnic (or, less commonly, vice versa) where the borrowed form preserves the original sound a lot better than the inherited form of the original word does. IIRC there’s a bunch of similar cases between English and French as well.

  4. the borrowed form preserves the original sound a lot better than the inherited form of the original word does

    And then there are cases where the original form has transmogrified past recognition:

    Estonian õis1 : õie : õit ‘katteseemnetaime erksavärviline sugulise paljunemise organ’ = flower; bloom
    ◊ õile, õilis, õitsema ‘to bloom”
    ← balti = from Baltic
    Lithuanian (leedu) žiedas ‘õis’; žiedėti ‘õitsema; hallitama’ to bloom; to grow mouldy
    Latvian (läti) zieds ‘õis’; ziedēt ‘õitsema; hallitama’
    ● liivi ēdrikšõ ‘õitseda’; ēdrõm ‘õis’
    vadja eďďittsää ‘õitseda (peamiselt rukki kohta)’; eďďõlmo ‘õietolm’
    Finnish (soome) van heitiä ‘õitseda (teravilja kohta)’(obsolete) to bloom (of cereals); hedelmä ‘vili; puuvili’ fruit

    I’ve looked it up to see where heisi, as in koiranheisi/koirainhesipuu ‘Guelder rose/калина’ has come from.

  5. You left out the best part:

    õitsema on tuletis, < *heiδ-itse- ja samuti õile, tõenäoliselt < *heiδ-elmä. õilis (tuletis sõnast õile) võeti kirjakeeles keeleuuenduse ajal kasutusele uues tähenduses, sest selle vastena oli eesti rahvalaulu saksakeelses tõlkes ekslikult kasutatud sõna edel ‘üllas, õilis’.

    The second sentence means (via GT) “õilis (a derivative of the word õile) was introduced in the written language in a new sense during the language renewal, because the word edel ‘noble, noble’ was erroneously used in the German translation of Estonian folk songs.”

  6. I don’t see the point of italicizing kanji and katakana, which are perfectly good English words,

    Indeed, more English speakers can probably explain what “kanji” means than can explain what a “chronotope” is.

  7. Bathrobe says:

    I was going to comment at the LL post but decided against it. I don’t read kanbun and I don’t read pre-modern kanbun-kundoku, which was the style used by officials up till the Meiji restoration, but I had to wonder exactly what’s new about this particular book. It’s true that this style was ditched in favour of genbun-itchi (the new amalgam of written and spoken language that the Japanese decided to adopt on the model of standardised languages of Western nation-states), and therefore fell out of favour (although schoolchildren still learn to read kanbun), but I don’t think anyone ever doubted the importance of literary Sinitic in the creation of modern Japanese. It’s just that it wasn’t fashionable…. Too bookish for those educated in the new language, too Chinese for people who wanted to focus on native culture… But there have always been people who can read that kind of language. One Japanese I knew in China told me had no problem at all reading old Chinese texts because he’d studied kanbun at school. Perhaps he was exceptional for his interest in China, but I think it’s a mark of the traditional well-educated intellectual to be familiar with these older styles of Japanese and Sino-Japanese.

    Definitely look at the Wikipedia article on Kanbun.

  8. David Eddyshaw says:

    Sansom’s Historical Grammar of Japanese (published 1928) clearly traces “Ordinary Meiji Prose” back to “Sinico-Japanese”, as the peculiar sort of Classical Japanese that results from reading Kanbun became a style that Japanese literati actually composed in directly.

    McIlroy’s revision of Chamberlain’s grammar of the prewar formal written style (published 1924) says “the Chinese style, or Sinico-Japanese … is founded on the literal translations of the Chinese classics, which were formerly the textbooks in every school. This style is used in the higher type of contemporary literature. We find it in technical books, in some magazines, and in editorials of newspapers.”

    On the face of it, “new understanding” seems to be pushing it rather. Or to put it another way, false.

  9. @AG: For 羅陀 < Rādha, the regular Sino-Korean reading would be 라타 Rata instead of 라다 Rada. By the way, at the beginning of a word, ㄹ r in Sino-Korean automatically turns into ㄴ n in South Korea, so it would be 나타 Nata on its own.

    Some Buddhist terms, especially many of the chants in the Heart Sutra, do have irregular Sino-Korean readings. But I can’t find any source indicating Rada for 羅陀, just the expected Rata.

    Sino-Korean readings do approximate the original Sanskrit or Pali pronunciation better for many characters that appear in such transcriptions, such as 인 in for 因 < "in", 라 ra for 羅 < "ra", or 하 ha for 訶 < "ha". In Mandarin Chinese they would be and yīn, luó, and respectively. Sino-Korean also does well in those cases where the coda consonants of Middle Chinese were employed in the transcription, such as 那落/奈落 < naraka which becomes 나락 narak. Some cases such as 발마 balma for 跋摩 < varma(n) must be lucky coincidences, the original t-coda of the first character representing a liquid in both the intended Middle Chinese transcription and contemporary Sino-Korean.

    Sino-Korean of course went through its own unique developments. But many of the most salient differences such as the palatalizations only affected characters that wouldn't have been used much in transcribing Indic terms. So it's not surprising that Sino-Korean seems to approximate the original Indic pronunciations better than Mandarin. By the way, some of the translations of Buddhist texts may have been done by Korean monks writing in Classical Chinese, so maybe a Korean-style pronunciation of Chinese characters was already an influence on the characters chosen in the transcriptions, even though is just wild speculation on my part (it's not even clear when the Sino-Korean readings became standardized).

    Note that I'm only talking about Mandarin here, as other Sinitic languages developed their own readings. Cantonese, the one I'm most familiar with, had lots of vowel shifts that made it quite different from Middle Chinese. The Min varieties seem to have readings that are closest to Sino-Korean, especially in the literary readings (the vernacular readings tend to reflect the more divergent characteristics of Min Chinese).

  10. J.W. Brewer says:

    To Vanya’s point, google translate renders “chronotope” into Chinese (traditional characters) as 計時表, which however back-translates into “chronograph” in the sense “fancy/accurate wristwatch.” You get the same result if you input Bakhtin’s original xронотоп. Japanese quite sensibly avoids kanji and just goes with the phonetic クロノトープ.

  11. You get the same result if you input Bakhtin’s original xронотоп.
    … presumably because GT is using English as an intermediate step.

    If you try inputting “chronotope” as part of a sentence, GT will sometimes offer other renderings, e.g. 年代記 (“annalistic chronicle”). But I couldn’t find any input sentences that would induce GT to come up with the standard Chinese translation for “chronotope,” which is 时空体, lit. “space-time form.”

    Published Japanese translations of Bakhtin use 時空間, “space-time.”

  12. As for Saitō Mareshi, the commenters above who think the publisher’s blurb oversells the novelty of his claims certainly have a point. The editor’s preface by Ross King and Christina Laffin (available free on the Brill website) gives a better summary of the current state of the field in academic discussions of these issues in English.

    I do think it unfortunate that the translators rendered Saitō’s kanbunmyaku using the bland and slightly awkward phrase “Literary Sinitic Context.” As a term of art, “kanbun vein” would be more evocative and memorable. But the translation overall seems to have been done quite well, with generous annotations on aspects of the Japanese cultural background that Saitō takes for granted that his readers will understand.

  13. David Marjanović says:

    I don’t think I’ve encountered chronotope before. Does it just mean “a particular place at a particular time”?

    I guess Chinese has gone through more drastic sound changes than the other languages –

    Especially Mandarin, “the French of Sinitic”.

  14. I don’t think I’ve encountered chronotope before. Does it just mean “a particular place at a particular time”?

    No, it’s more a configuration of place and time characteristic of a particular literary form. There’s a (not very helpful) description at the Wikipedia article; the Literary Encyclopedia article looks pretty good, but unfortunately you can only read the first bit without a login:

    A term taken over by Mikhail Bakhtin from 1920s science to describe the manner in which literature represents time and space. In different kinds of writing there are differing chronotopes, by which changing historical conceptions of time and space are realised. Thus the ancient Greek novel is dominated by “adventure time”, in which the adventures of hero and heroine occur but which has no developmental impact upon their characters; like the space in which their adventures happen, it is effectively empty. By contrast, the time and space of the chivalric romance, though it retains elements of this adventure time, is dominated by the irruptions of the miraculous, which manifest themselves in narrative terms by the presence of “…

  15. Here’s a decent discussion, with a couple of examples.

  16. Bathrobe says:

    the standard Chinese translation for “chronotope,” which is 时空体

    Smart aleck! I don’t even know what chronotope means in English!


    While I was dismissive of the hype over Saitō Mareshi’s book, I certainly do think an English-language work, newer than the current nearly-century-old offerings, that presents the state of research in the field is welcome. I would probably be interested in reading it myself. I was just a little put off by Victor Mair’s enthusiasm.

    I should make a personal confession here. I was never interested in kanbun. I belonged to the school of “If I want to read Chinese I’ll learn Chinese”, not realising the extraordinary linguistic interest of such a translinguistic phenomenon as kanbun.

    My own epiphany came one day when I realised that that strange beast known as a 漢和辞典 kanwa jiten, literally ‘Han-Japanese dictionary’, which I’d always assumed was the Japanese equivalent to Nelson’s character dictionary, was actually a dictionary of the Chinese language (usually Classical, but some dictionaries throw in modern pinyin) almost bodily translated into Japanese. It’s as though someone had translated Websters into Japanese, leaving only the English headwords and the English example sentences, with the rest translated or annotated in Japanese.

    It was then that I realised how deeply Chinese had penetrated into the Japanese language, and not simply “Japanese has borrowed a lot of words from Chinese, so what?”

  17. Bathrobe says:

    Having read the preface by King and Laffin, I find myself both in agreement with and slightly annoyed by what they say. I’ve previously discussed at LH the way in which Japanese literature “declared independence” from Chinese culture in the late 19th century (the rise of 国文学 ‘national literature’, which pointedly excludes from the scope of Japanese literature most things written in Chinese) and switched over to the West as a model, both in developing a standard national language and in adopting the Western-style printed book. The preface underlines that change, one which I was always aware of.

    The annoying part was the very modern concern with adopting and justifying terminological innovations. It may be impossible to restore the single cultural and literary zone that once united East Asia, and it might be “Sinocentric” to use the word “Chinese” in the manner of early Western interpreters of Japan, but the decision to adopt and propagate terms like “Sinitic” (which seems somehow precious to me) grated a little. It seems just a little “gassy”, by which I mean somewhat vacuous and ever so slightly “puffed up”.

  18. John Cowan says:

    a dictionary of the Chinese language […] almost bodily translated into Japanese

    This is not uncommon. Lewis Short’s Latin-English dictionary is in fact a revision of a translation by Andrews of a Latin-German dictionary. This was riginally edited by Scheller in 1783 and then heavily re-edited by Freund and published in 1834-45. Freund also helped to correct the English translation before it was handed over to L & S. Short worked on the letters A-C and Lewis all the rest, but the publisher lost Short’s text for B and C, so Lewis had to re-edit them.

    It’s still a whole lot easier to do all this than to write a bilingual dictionary from scratch, which is what the Oxford Latin Dictionary is. L & S has the advantage that though it doesn’t systematically cover Latin after 200 CE (the cutoff date for the OLD), it is still a handy one-volume reference to the most common words and senses, and much easier to use than Du Cange (10 volumes of medieval Latin) even now.


    That’s just Chinese in the decent obscurity of a learned language. In any case it has long been used for the family of Chinese languages, mostly called “X Chinese” in English with the exception of Dungan, which is historically a particular variety of Mandarin but sociolinguistically quite distinct from the other Sinitic languages.

  19. Bathrobe says:

    Yes, I have an English-Vietnamese dictionary that is essentially a translation of an English dictionary of English. This approach has its advantages (lots of examples, explanations from the point of view of English) but doesn’t quite scratch the itch, i.e., by providing straightforward translations from one language to another.

    Perhaps I should have explained more clearly about 漢和辞典 kanwa jiten. A 漢和辞典 is a dictionary of Chinese characters. It is organised by character, often by pronunciation (reading), but with a lookup index arranged by radicals and number of strokes, and possibly some alternative method. It will list the old Chinese meanings of each character way back to the time of Confucius, with uniquely Japanese meanings merely tacked on at the end. It will indicate if a character is a 国字 kokuji, character ‘made in Japan’. It will give two types of pronunciation for each character, on-yomi and kun-yomi, and maybe pinyin as an extra. Typically there will NOT be a list of words, with definitions, containing the character in question (for 国 you might expect it to list 国家 ‘nation’, 国営 ‘state-run’, 国立 ‘national (state-established), etc., etc., but these are usually not so much in evidence.)

    These features will, of course, depend on the dictionary, and I don’t have any with me at the moment to use as an example.

    What makes the 漢和辞典 different from any English-Latin dictionary is that it is regarded in Japan as one of the two essential types of dictionary needed for the Japanese language. The other is the 国語辞典 ‘national language dictionary’, which is a dictionary of Japanese words listed in kana order. Such a dictionary is similar to what we are familiar with in English.

    Nelson’s Japanese-English character dictionary is quite different from any 漢和辞典. Of course it lists characters in the same way as a 漢和辞典, with lookup methods to help locate the character you want. But at each entry, in addition to listing the meaning of each character in English (which, if I remember rightly, is oriented to modern Japanese and doesn’t place a great deal of emphasis on the original meaning in the Chinese classics), it gives a long list of character compounds in modern Japanese — such as 国家, 国営, 国立, as I mentioned above — with English translations.

    For a long time I was mystified why you would want such an animal as a 漢和辞典, which doesn’t help a lot in explaining the actual meaning of words. But when you understand it as a dictionary of characters, giving their historical meaning from ancient times till the present day, it starts to make sense. And the way it makes most sense is as a dictionary compiled from the point of view of kanbun, i.e., designed not so much to explain the modern meanings of words, but to understand characters as they are historically used in Chinese, and particularly in the Chinese classics.

    Which was my point: the existence of this kind of dictionary, and the way it approaches the explanation of Chinese characters, demonstrates the way in which kanbun and the Chinese classics still occupy an important place at the heart of the Japanese language.

  20. Bathrobe says:

    With regard to “Sinitic”, yes, of course it is just an erudite word. What struck me was the way they were attempting to come up with what they saw as more ‘politically-correct’ or more appropriate vocabulary in order to present the topic from a less historically-entrenched angle. This is not necessarily a bad thing. I suspect (just suspect, mind you) that it resonates with a message that might be found in the book: Literary Chinese has many different manifestations across the Kanji-sphere, which are not necessarily some pale reflection of the real thing (i.e., China). Or perhaps they themselves were very conscious of how they wanted to present the topic to English readers.

    This might be considered a refreshing overhaul of vocabulary for our postmodern era, but I’m still not totally convinced that this new vocabulary is necessary or helpful, or that their choices are always happy ones (as seen in comments on how kanbunmyaku itself should be translated). And I found the preface rather stodgy reading because of their concentration on their vocabulary choices. Needless to say, one of the reasons for Victor Mair’s enthusiasm for the book is that it has adopted some of his suggested usages.

  21. Bathrobe says:

    * It is organised by character, often by pronunciation (reading)

    I’ve had too much exposure to different dictionaries. I’ve seen dictionaries arranged by pronunciation and by radical/number of strokes. I can’t remember which is normally used in 漢和辞典.

  22. David Eddyshaw says:

    The dying art of Shigin


    consists of singing Classical Chinese poetry in Classical Japanese, via the magic of Kanbun.
    Basically, it doesn’t rhyme or scan in Japanese: it’s poetry because the Chinese Vorlage rhymes and scans.

    I don’t know of any parallel in any other culture, though if WP is to be believed it exists (or existed) in Vietnamese. I wonder if there is, or was, a Korean equivalent?

  23. I myself am not aware of any sung tradition of Classical Chinese poetry in Korea. In the rare cases that they are recited, they are simply read in the metre without any attempt at singing.

    But this doesn’t mean that such a tradition doesn’t exist. Shigin would be 시음 詩吟 si’eum in Sino-Korean and it is listed in the dictionary though I didn’t know the term. It is defined as “reciting poetry with a melody”, and a citation is given from a Korean novel, so something like this must have existed at one point at least.

    I tried in vain however to find any examples on YouTube. All examples of Classical Chinese poetry recited in Korean I could find lacked any melody.

  24. Bathrobe says:

    It’s a pity Elessorn hasn’t turned up to comment on this post. They know more about this topic than all of the regular commenters put together.

Speak Your Mind