Korean oralization of Literary Sinitic

« previous post | next post »

Si Nae Park came to Penn last Thursday (4/18/24) to talk about kugyŏl / gugyeol / kwukyel 구결 口訣 ("oral glossing").

Gugyeol, or kwukyel, is a system for rendering texts written in Classical Chinese into understandable Korean. It was used chiefly during the Joseon dynasty, when readings of the Chinese classics were of paramount social importance. Thus, in gugyeol, the original text in Classical Chinese was not modified, and the additional markers were simply inserted between phrases.

The parts of the Chinese sentence would then be read in Korean out of sequence to approximate Korean (SOV) rather than Chinese (SVO) word order. A similar system for reading Classical Chinese is still used in Japan and is known as kanbun kundoku.


Park's analyses and explanations were like a revelation to me for a number of reasons.  First of all, I was already familiar with the analogous Japanese method for reading Literary Sinitic, called kundoku, which involves a lot of rearrangement, modification, and annotation of the text to make it more like Japanese, whereas it seems that kugyŏl tries to stay closer to the Literary Sinitic.

I was also long aware of the Sinitic expression kǒujué 口訣, but in Chinese it means something quite different than it does in Korean:


This is not to say that premodern Chinese did not see a need for making the content of Literary Sinitic available for those who were unable to read it.  For this purpose, socially sensitive individuals resorted to a variety of devices, including oral and written translations into the vernacular, as I demonstrated in "Language and Ideology in the Written Popularizations of the Sacred Edict", in David Johnson, Andrew J. Nathan, and Evelyn S. Rawski, eds., Popular Culture in Late Imperial China (Berkeley and Los Angeles:  University of California Press, 1985), pp. 325-359.

Chinese referred to these devices as zhíjiě 直解 ("direct explanation"), zhíshuō yàolüè 直說要略 ("directly expounded essentials"), yǎnyì 演義 ("elaboration"), tújiě 圖解 ("illustrated explanation"), and many others, which shows that there was a need for making literary texts available to the broader, uneducated populace, and that it was being met by various means.

Nowadays, almost all the major literary and classical Chinese texts have been rendered into Mandarin, and these are called 白話翻譯 ("vernacular translations").

The Koreans during the middle of the second millennium AD also had textbooks for learning vernacular Sinitic.

Bak Tongsa (Chinese: 朴通事; lit. 'Pak the interpreter') was a textbook of colloquial northern Chinese published by the Bureau of Interpreters in Korea in various editions between the 14th and 18th centuries. Like the contemporaneous Nogeoldae ('Old Cathayan'), it is an important source on both Late Middle Korean and the history of Mandarin Chinese. Whereas the Nogeoldae consists of dialogues and focusses on travelling merchants, Bak Tongsa is a narrative text covering society and culture.


Lest I overlook another significant Korean means for annotating Chinese-language texts, I should mention eonhae 언해 諺解, which the Japanese also had, genkai げんかい 諺解 (lit., "aphoristic explanation").

In sum, I will make two main points:  1.there's a sharp difference between oralization and vernacularization, 2. kugyŏl belongs to the former, beon-yeog 번역 / hon'yaku 翻訳 / fānyì 翻譯 to the latter.


Selected readings


  1. AntC said,

    April 23, 2024 @ 11:43 pm

    … into understandable Korean.

    1.there's a sharp difference between oralization and vernacularization, 2. kugyŏl belongs to the former, …

    Thank you Prof Mair, but I'm bewildered. (And the rest of that wikip article makes me more bewildered. For example: "… gugyeol sought to render Chinese texts into Korean with a minimum of distortion." — undistorted Chinese would be un-"understandable Korean", I would have thought.)

    An oralization technique would support reading out the Chinese characters so they sound like Korean(?)/use Korean phonemes. But they wouldn't be _understandable_ Korean unless the vocabularies of the two languages are the same, mutatis mutandis the pronunciation and the word/constituent order.

    Or is it that the whole Chinese dictionary as of that time got imported into Korean, so they ended up with twice as many words for many concepts. (I appreciate 'word' might not be a happy term here.) Comparable to the English word-hoard being enormous; because Norman invasion?

    Does it end up as something worse than Franglais: word order and grammatical particles in one language; 'content' words from another?

  2. Thomas said,

    April 24, 2024 @ 12:08 am

    While the wikipedia article on kanbun kundoku provides an example, none for gugyeol is to be found. Can anyone provide such an example? Also, was one expected to read such jumbled sentences fluently back in the day?

  3. Jonathan Smith said,

    April 24, 2024 @ 7:38 am


    A contemporary example via Google Books is here if p. 86 of the book (Bunkyo Kin, Literary Sinitic and East Asia: A Cultural Sphere of Vernacular Reading happens to be visible for you…

    "the whole Chinese dictionary as of that time got imported into Korean" — ya pretty much, or at least the lexicons of the relevant texts — in the example above, all lexemes have Sino-Korean "readings," very much unlike the contemporary Japanese example on the previous page (p. 85).

  4. Weh said,

    April 24, 2024 @ 11:16 am

    Interestingly, methods similar to Korean kwukyel and Japanese kundoku were used in some Sinitic languages as well. The Hokkien dictionary published in 1832 by W.H. Medhurst has a lot of such examples:

    Page 260:
    求也,爲之聚斂,而附益之。 Kiû iá, ûi chi chī-liám, jî hū-ek chi.
    求也,替伊拾稅,而添伊較濟。 Kiû iá, thè i khioh-sòe, jî thiam i khah chè.
    "Kiû gathered the taxes for him, and thus increased his wealth."

    Page 586:
    設爲庠序學校以教之。 Siat-ûi siông-sī ha̍k-hāu í kàu chi.
    設做庠序學校以教伊。 Siat-chò siông-sī ha̍k-hāu í kà i.
    "And the appointed schools and seminaries of instruction to teach the people."

    Page 656:
    小人之道,的然而日亡。 Siáu-jîn chi tō, tek-jiân jî ji̍t bông.
    小人兮道,光然而一日無去。 Siáu-jîn ê tō, kuiⁿ-jiân jî chi̍t ji̍t bô-khì.
    "The way of a worthless man is glittering, but in a single day he is forgotten."

    The first line is the original text in Literary Chinese, and the second line is the Hokkien paraphrase from the Medhurst's dictionary.

  5. Victor Mair said,

    April 24, 2024 @ 1:31 pm

    @Jonathan Smith and @Weh

    Thank you for the invaluable citations to Kin Bunkyō and W. H. Medhurst.

  6. Jongseong Park said,

    April 26, 2024 @ 2:08 am

    The way gugyeol functioned changed over the period that it was in use, which I think is causing a bit of confusion. In the more recent type that became prominent during the Joseon dynasty, Korean particles and connectors were inserted to aid in parsing the original text which was read in Sino-Korean pronunciation.

    To illustrate, here's the opening passage from the Analects where I have underlined the inserted elements in gugyeol:

    Original Literary Sinitic: 學而時習之不亦悅乎
    Sino-Korean reading: 학이시습지불역열호 hak i si seup ji bul yeok yeol ho
    Gugyeol text: 學而時習之面不亦悅乎牙
    Sino-Korean reading: 학이시습지면불역열호아 hak i si seup ji myeon bul yeok yeol ho a

    The inserted elements correspond to the -(으)면 -(eu)myeon, a suffix meaning 'if', and -아 -a, an interrogative suffix (Modern Korean would prefer -가 -ga here).

    So the reader can tell from the inserted gugyeol text that this is a question asking 'If A, then B?' even if they cannot figure out what the original text means. As AntC guessed, simply reading out the original text with the Sino-Korean sound values results in 'un-understandable Korean' . It is basically gibberish to Korean speakers, apart from being recognizable as a famous passage – much like 'Veni, vidi, vici' might be for English speakers today.

    Koreans learning Classical Chinese would have learnt what individual characters meant, and the inserted gugyeol gave hints about how these fit together to form meaningful sentences. The added particles and connectors became a standard part of how many of these lines were recited, so many Koreans today still quote the lines as '학이시습지면 불역열호아? Hagisiseupji-myeon buryeogyeolho-a?'.

    An imperfect comparison might be applying modern punctuation and capitalization to texts in Latin. Writing 'Et tu, Brute?' indicates to the reader where there is a natural break, that 'Brute' is a form of a proper name, and that the utterance is a question even if they don't know what the words mean.

    This type of gugyeol is described as eumdok gugyeol (음독구결, 音讀口訣) because it uses the Sino-Korean readings of the characters, and as sundok gugyeol (순독구결, 順讀口訣) because it follows the reading order of the original text without modification.

    But in earlier versions, often called seokdok gugyeol (석독구결, 釋讀口訣), there could be signs to indicate how to replace the characters with native equivalents, as well as how to change the reading order to suit the Korean word order. The type that changes the word order is also called yeokdok gugyeol (역독구결, 逆讀口訣). The existence of these complex systems was rediscovered in the 1970s and it is a topic I'm not very familiar with. Suffice it to say that there was huge variation in the signs (written in ink or pressed in the paper without ink, in simplified dots and strokes or in full characters, etc.) that fall under the umbrella term of gugyeol.

    My impression is that earlier, a more complex system was used to indicate how to semi-translate the original text into Korean by substituting frequently with native equivalents and changing the word order. Later, especially as the invention of the Korean alphabet made it possible to explicitly translate the original text into Korean in what is termed eonhae (언해, 諺解), the use of gugyeol simplified to basically serve as reading aids for Koreans studying text in Classical Chinese.

    The National Hangeul Museum in Seoul displays a replica of a 13th or 14th century woodblock print of the Buddhist text Yuga saji ron (유가사지론, 瑜伽師地論, 'Yogācārabhūmi-Śāstra') which employs gugyeol signs for indicating how to change the reading order. I think that would be around the time the transition would have been happening to the later types of gugyeol.

  7. Victor Mair said,

    April 26, 2024 @ 7:00 am

    @Jongseong Park:

    As always, glad to have you back.

RSS feed for comments on this post