Abstract
-
In the early 20th century, a number of Korean textbooks on Hanmun (Korean Literary Sinitic) were published. These textbooks introduced new ways of understanding and teaching Hanmun grammar. Some of these innovations are highly original and explore experimental ideas.
The hŏsa/silsa paradigm (usually translated as “empty” and “full words” in English) forms the basis of many pre-modern and early modern Literary Sinitic grammars, but these concepts are understood in different ways in each of them. The conventional understanding that hŏsa are ‘function words’ does not sufficiently explain their descriptive usage in early modern Korean grammars. This article investigates how the concept of hŏsa/silsa is understood in Korean Literary Sinitic grammars and textbooks from the Late Chosŏn and Colonial periods, and how it is applied in teaching.
Although these old grammars are clearly outdated in many ways, the grammatical concepts they present still hold some didactic value. Some of these concepts, particularly the hŏsa/silsa concept, are still useful for teaching basic Literary Sinitic courses.
-
Keywords: Hanmun, Classical Chinese, Literary Chinese, textbook, history of grammar
Introduction1
Grammar in early 20th-century Korean textbooks of Literary Sinitic is a complex issue, especially because of the wealth of concepts used and the experimental terminology employed. Among these grammatical ideas, the duality of hŏsa and silsa (meaning “empty words” and “full words,” respectively) is a key concept of East Asian grammatical thought and it forms the basis of many pre-modern and early modern Literary Sinitic grammars. These concepts are interpreted differently in early 20th-century grammars of Literary Sinitic in Korea, which result in grammatical models that have little in common with each other.
The conventional understanding that hŏsa are “function words” (possibly inspired by Ma Jianzhong’s grammar, as laid out later in this article) is insufficient to explain their descriptive usage in early modern Korean grammars.
Although the concepts set out in these textbooks are no longer considered state of the art in Hanmun pedagogy, they remain of interest for two reasons. First: They shed light on how early 20th-century grammar teachers approached Hanmun. They demonstrate the range of concepts that teachers could draw upon, as well as their tendency to experiment with new ideas. Secondly, although most of these ideas have long since been forgotten, they remain relevant to Hanmun teaching today.
2 Many of the problems that Hanmun teachers faced in the early 20th century persist in Korea to this day. These problems have not yet been fully solved. There is also room for new methods to complement contemporary Hanmun teaching. In other words, these old textbooks offer practical lessons.
In my paper, I will examine how early modern (early 20th century) Korean Hanmun grammars describe grammar, the terminology they use, and how they present Hanmun grammar to readers. I will examine how grammatical concepts and terminology are understood in Korean grammars and Literary Sinitic textbooks from the Late Choson and Colonial periods, and how they are applied in teaching. The grammars I will discuss are all intended for educational use. This will provide insight into how the future of Hanmun teaching was envisioned in the early 20th century Korea. From this, I hope to extract some ideas that are still useful for Hanmun education today.
Methodologically, I follow what I believe to be the time-honoured principles of philology: Elements of linguistical and historical analysis are freely combined. Since this text also aims at giving the reader some insights into early 20th century textbooks of Hanmun in Korea, some room is given to their exposition. This may fall short of the stricter standards of historical linguistics, but philology has traditionally been a chameleon amongst disciplines and I believe that it is worth-while to keep this spirit alive in Western language Korean Studies. While there is need for highly specialized literature, it is also useful to include some elements of general surveys into articles if the subject has not already been treated elsewhere (in English). The result may seem less focused than is desirable by today’s standards for academic writing. My excuse is that these topics have not yet been presented in English and that it seems worthwhile to provide the reader with some examples of what can be found in these textbooks.
For these reasons, this article is partly a descriptive study of early 20th century Korean Hanmun grammars, partly a history of the hŏsa/silsa paradigm, and partly an inquiry into potential uses for old didactic ideas in contemporary Hanmun teaching.
Most of these grammars use the terminology of Western grammar, which was introduced to Korea in the late 19th century through Japanese and Chinese adaptations, as well as by reformers and academics who had studied Western languages.
3 Meanwhile, the concepts of
hŏsa and
silsa remained the foundation of some Literary Sinitic grammars.
The issues of writing versus language and grammar versus etymology are key questions in the fields of philology and linguistics in pre-modern East Asia. The nature of Chinese writing 漢字 (Ch. hanzi, Kor. hancha) — either in itself or in contrast to other writing systems — was a key focus of Hanmun scholarship in early modern Korea. This is also reflected in the textbooks of that period, which strike a balance between traditional assumptions and modern (Western?) linguistics.
Although many of the concepts used in these grammars are outdated, I will argue that they still have some educational value, and that some of these concepts, such as the hŏsa/silsa dichotomy, can and should still be used in introductory Literary Sinitic courses.
One more not is necessary on the concept of the “word” in Hanmun. Terminology is inconsistent in premodern grammars and the question of where to assume word boundaries remains a problem in Sinitic linguistics up to today. The case is relatively clear when sa 詞 is used for “word”. But often we do instead find cha 字, meaning “character”. But “character” in Literary Sinitic in most cases is identical with “word” and while we can safely assume that competent users of Literary Sinitic were aware of the existence of compound words (in which one word consists of more than one character), this differences are not always explicitly noted. This leads to some terminological ambiguities which may seem problematic from the perspective of present-day linguistics. They were, apparently, not considered problematic by the authors covered here.
Background: Traditions of Hanmun
Learners of Literary Sinitic (Hanmun) in pre-modern Korea, like in pre-modern China, did not approach the language by systematic grammatical analysis, but rather by rote learning and repetition of texts. Passive reading skills as well as active writing skills were formed by re-arranging patterns and sentences into new constructions. These didactic methods were highly effective and their results are, without question, very good. By these methods, highly competent users of the language were educated, who excelled in all sorts of conventionalized genres as well as in creative language use, displaying a skillful and comprehensive mastery of Literary Sinitic.
The downside of this traditional approach is that it takes a decade or two of intensive training to reach competency. For a society, like that of the Chosŏn state, for which fostering scholars, who would build their livelihood and that of their clans on this knowledge, was an integral part of its culture and ideology, this was not only unproblematic, but rather a desired side-effect, allowing for gate-keeping by the elites and also serving the propagation of learning as an esoteric activity. But after the abolition of the Confucian classics state examinations during the 1894 (kabo 甲午) reforms of, which destroyed the socio-economic foundations that made the investment of time and resources into Literary Sinitic education a sensible decision, interest in learning by this method faded rapidly.
Instead, historians and archive specialists interested in Literary Sinitic either as a working language or as a symbolic resource became increasingly aware of the need for alternative methods of learning it. As few learners were willing to invest the time required by the traditional system, new systems with new didactic approaches were invented and tested. The textbooks introduced in this article played a part in an attempt to modernize Hanmun education and test new ways of describing the language to make it more accessible to students. In doing so, authors of textbooks faced difficulties due to Japanese colonial censorship, which suspected Hanmun textbooks to potentially carry subversive political ideas.
4
The idea of Hanmun as part of a generalist, liberal arts education for schoolchildren appears to be a later development, emerging only after the discourse on the modernization of the humanities had waned, making way for approaches that prioritized the promotion of Hanmun education as part of Korean identity.
Sources: Five Early 20th Century Hanmun Textbooks
This article describes and analyzes some features found in a total of five grammar books from early 20th cent. Korea. They are (in chronological order):
(1) Ch’odŭng changmunbŏp初等作文法. By Wŏn Yŏngŭi 元泳義 (Yunghŭi 隆熙 2, 1908): Kyŏngsŏng 京城: Kwangdong sŏguk 光東書局.
(2) Hanmun ŭidok chahae 漢文義讀自解. By Sin Chaeyŏng 辛在英 (Taishō 大正 5, 1916): Kyŏngsŏng 京城: Sin Chaeyŏng ka 辛在英家.
(3) Hanmunbŏp cheyo 漢文法提要. By Kang Mae 姜邁 (Taishō 大正 6, 1917): Kyŏngsŏng 京城: Pangmun sŏgwan博文書館.
(4) Silji ŭngyong changmun taebang 實地應用作文大方. By the Yŏngch’ang sŏgwan p’yŏnjip pu 永昌書館編輯部 (Taishō 大正 10, 1921): Kyŏngsŏng 京城: Yŏngch’ang sŏgwan永昌書館
(5) Hanmun kyosu ch’ŏpkyŏng 漢文教授捷徑. And the accompanying Hanmun kyosu ch’ŏpkyŏng chusŏk 漢文教授捷徑註釋. Both by Chŏng Ik 鄭瀷 (Shōwa 昭 和 4, 1929): Kyŏngsŏng 京城: Unhyang sŏru 芸香書樓.
There are some Korean language studies that have explored some aspects of these textbooks. Chŏng Sŏng’im has written an article on adverbs in all of these textbooks.
5 There are at least six detailed studies on aspects of the
Hanmunbŏp cheyo by Kim Yonghan
6 and a study on the grammatical system of the
Hanmun ŭidok chahae by the same author.
7 Also, the question of the use of Korean script in those textbooks has attracted some attention.
8
This selection of textbooks covers most of the pre-war colonial period in Korea. While it is unclear to what extent these textbooks are representative of those from the colonial era, this is not problematic given that the aim of this study is to discover, identify and describe unique and unusual features. However, the conclusions drawn from this selection cannot be considered representative of textbooks from that period in general.
Most of the textbooks clearly state their motivation. Interestingly, all of the textbooks mentioned here have their introduction written in Hanmun – apparently the tradition of the preface genre, featuring elaborate language and displaying the authors ability to write in the higher registers of the language, was very much alive in early 20th cent. Korean Hanmun education circles.
Wŏn Yŏngŭi 元泳義 (1852-1928) in his 1908
Ch’odŭng changmunbŏp> 初等作文 法, which is famed as the first Hanmun grammar in Korea,
9 does give no explanation of his motivation to write that textbook.
10 It is worth to note that Wŏn Yŏngŭi also published a Korean language textbook in the same year (1908), titled
Kungmun kwabon.
11 He also wrote other, mostly introductory, textbooks in the fields of language and history.
12
Sin Chaeyŏng 辛在英 in his 1916
Hanmun ŭidok chahae 漢文義讀自解 states that he wrote the book to spread his “new” method of reading Hanmun by word inversion, resulting in something approximate to Korean order.
13
Kang Mae 姜邁 in his 1917
Hanmunbŏp cheyo 漢文法提要 gives a stylistically elaborate introduction to his work in Hanmun with remarkably developed interpunctuation, distinguishing systematically between commata and full stop dots (which mark the end of paragraphs, not sentences).
14 He explicitly quotes Ma Jianzhong’s 馬建忠
Mashi wentong 馬氏文通 and others as inspirations for a systematic treatment of Chinese grammar. Korea, on the other hand, is still relying on old methods, according to Kang Mae. These old methods are ineffective and use materials that are badly suited to the needs of teaching, especially because they use texts that are too difficult and too hard to understand. His textbook is the result of his experiences with this ineffective system and is an attempt to reform this on the basis of teaching notes that he prepared during his classes.
15
The 1921
Silji ŭngyong changmun taebang, authored by a collective of authors, begins the preface – entirely written in Hanmun without interpunctuation – with a lament on how style has deteriorated over time and how stylistic expression has become ever more convoluted and that contemporary readers wrongly value overly long and complicated style. The result, in their opinion, is a preference for lofty and meandering style which lacks substance.
16 Somewhat ironically, the style of the preface is itself rather long-winding and far from clear. Nevertheless the preface claims that the book is aimed at providing the foundation for a better understanding of grammar and style, thus leading to more substantial writing.
17
In comparison, the prefaces to the 1929
Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑 (by Chŏng Ik 鄭瀷) are written in clear and simple Hanmun, which is made even more accessible by interpunctuation. The style differs with the different authors of the prefaces, but the effort to write in a lucid and coherent is clearly visible.
18 The main motivation of the author is stated to be aimed at a speedier and more effective way of Hanmun teaching. The old methods simply take too long and the author suggests that by the methods laid out in his book, proficiency in Hanmun can be reached quicker, without sacrificing quality.
19 This faster acquisition of the language is also considered an important feature since trends at the time call for the abolition of Hanmun studies. Better Hanmun education is thus a means of saving Hanmun from being purged from the curriculum.
20 It must be remarked that this seems rather similar to the problems that present-day Hanmun education is facing.
The respective target audiences of the textbooks are also display a wide range of variation:
|
Title |
Content |
|
Wŏn Yŏngŭi 元泳義 (1908): Ch‘odŭng changmunbŏp 初等作文法. |
Condensed explanation of syntax, word classes and function words, mostly through example sentences without further annotation or explanation. Meant to be a textbook, providing materials to be used in the classroom with explanations being added by the teacher for every sentence. |
|
Sin Chaeyŏng 辛在英 (1916): Hanmun ŭidok chahae 漢文義讀自解. |
No table of contents. Short list of grammatical items at the start, then mostly just (partially annotated) practice sentences. Meant to be a textbook with additional explanations being provided by the teacher. |
|
Kang Mae 姜邁 (1917): Hanmunbŏp cheyo 漢文法提要. |
Systematic explanation of word classes, syntax, and of function words. Includes example sentences for the various function words. |
|
Yŏngch’ang sŏgwan p’yŏnjip pu 永昌書館編輯部 (1921): Silji ŭngyong changmun taebang 實地應用作文大方. |
Systematic explanations for word classes, syntax, function words, and stylistics. Also includes explanations on genre. Extensive example texts and reading section. All Hanmun texts are annotated with t’o. |
|
Chŏng Ik 鄭 瀷 (1929): Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑 and Hanmun kyosu ch‘ŏpkyŏng chusŏk 漢文教授捷徑註釋. |
Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑: Collection of sample sentences for the explanation of style. Also includes explanations on function words. Hanmun kyosu ch‘ŏpkyŏng chusŏk 漢文教 授捷徑註釋:Vocabulary and additional materials to supplement the main volume. The vocabulary includes function words and their explanation. |
It is interesting to note that the focus of grammatical explanations is on function words, while sentences (in which these function words appear) mostly only feature as examples. This is despite the fact that, as Victor Mair has noted, “syntax is to an isolating language what morphology is to an inflected language.”
21 Apparently, the traditional Chinese focus on words (or rather: written characters) is still an important influence for these books.
As can be see from the above list, the content of the textbooks varies significantly. This is due to the different approaches of explaining grammar in the textbooks and of how they present the subject of Literary Sinitic to their audience. Broadly speaking, we can differentiate between two different approaches: Those textbooks that include systematic explanations of grammar and those that leave this to the teacher.
|
Textbook |
systematic explanation of grammar or not |
|
Kang Mae 姜邁 (1917): Hanmunbŏp cheyo 漢文法提要. |
yes |
|
Yŏngch’ang sŏgwan p’yŏnjip pu 永昌書館編輯部 (1921): Silji ŭngyong changmun taebang 實地應用作文大方. |
yes |
|
Wŏn Yŏngŭi 元泳義 (1908): Ch‘odŭng changmunbŏp 初等作文法. |
no |
|
Sin Chaeyŏng 辛在英 (1916): Hanmun ŭidok chahae 漢文義讀自解. |
no |
|
Chŏng Ik 鄭瀷 (1929): Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑(including the appendix Hanmun kyosu ch‘ŏpkyŏng chusŏk 漢文教授 捷徑註釋). |
no |
As we can see, both types of textbook seem to have coexisted, presumably catering for different needs and teaching situations.
Little is known about how these textbooks were actually used. We also know nothing about the print runs or distribution of these books. As they were commercially printed despite the considerable typographical difficulties involved in producing such books, we can assume that they were intended to be commercially viable.
Hŏsa and silsa
The concept of hŏsa (empty words) and silsa (full words) is one of the oldest concepts of Chinese grammar that is still used today.
In short, hŏsa are equivalent to what modern grammars would call ‘particles’, ‘function words’, ‘markers’ or ‘grammatical morphemes’. These include conjunctions and case markers. Silsa, on the other hand, encompass all other words, including nouns, verbs, adjectives and most adverbs.
Sinology does not provide a unified vision of what hŏsa are exactly W.A.C.H. Dobson wrote in the introduction to his 1974 Dictionary of the Chinese Particles:
“While sinologists are familiar in a general way with the class of words known as ‘particles’ (shiu-tzyh [xuzi in Hanyu pinyin] 虛字 the ‘empty words’ of traditional sinology), the class is, in fact, only loosely and broadly defined. Compilers of the traditional lexicons of the ‘empty words’ differ in important detail as to what words are ‘particles.’”
22
This is still true today: grammars of Literary Sinitic (including Hanmun) still disagree on exactly what these ‘empty words’ are. The concept is generally clear enough ('function words'), but the details, as always, are complex.
Hŏsa and silsa provide a basic, binary classification of words, which can be useful for introducing students to the concept of word classes. Silsa can then be explained further to include nouns, verbs, and adverbs, while hŏsa can be differentiated further to include postpositions, conjunctions, and so on.
Taking this approach to Literary Sinitic word classes from the perspective of the basicpair of hŏsa and silsa also provides a starting point for understanding the flexibility of word classes for most Literary Sinitic words (or rather, ‘characters’). Thinking of a word as a silsa that can be used as a verb, noun, adverb, etc. depending on the context is helpful for understanding this flexibility in Literary Sinitic syntax.
The classification of hŏsa is different in all our five early modern Korean Grammars. Not only are the scope of the the grammars and the examples that they choose different from each other, but the taxonomy of grammar is different. While the basic idea of what hŏsa are is similar in all grammars, the details of their delineation and the terminology used for the resulting categories differs.
The text that established the concept of
hŏsa/
silsa for the analysis of Chinese in the context of modern linguistics is Ma Jianzhong’s
Mashi wentong. The modern grammatical tradition of Literary Sinitic in East Asia is often said to have begun with the Chinese scholar Ma Jianzhong 馬建忠 (1845-1900). Although he was educated in France as a lawyer and was renowned during his lifetime as a specialist in Western law, his most significant contribution was in the field of linguistics, specifically descriptive grammar. And he was extraordinarily well-prepared for that task, having learned Latin, Greek, French, and German.
23 His grammar book, originally titled simply
Wentong 文 通, which is now generally known as
Mashi Wentong 馬氏文通 (1898), is often lauded – and sometimes critizized – as the first work to apply a systematic framework of analysis to Chinese grammar by implementing modern linguistic methods.
24 The influence of Ma Jianzhong’s grammar can hardly be overstated – there are few contemporary works on Chinese grammar that cannot be in some way linked to the
Mashi wentong.
In the introductory explanations to his 1908
Ch‘odŭng changmunbŏp 初等作文法, the author Wŏn Yŏngŭi 元泳義 explains that the book is aimed at children and beginners.
25
The target audience of Sin Chaeyŏng’s 1916
Hanmun ŭidok chahae 漢文義讀自解 are beginners, especially children. He also mentions that his method is useful for the education of first-graders, even though they should be expected to need about two years of instruction, instead of the two months that he sees as normal for other students.
26
Kang Mae 姜邁 (1917):
Hanmunbŏp cheyo 漢文法提要 is intended for learners of Hanmun and is assumed to be suitable for beginners and for children. But it also claims to be a treatment of Hanmun grammar that is useful for anyone interested in that language.
27
The 1921
Silji ŭngyong changmun taebang 實地應用作文大方 is aimed at Hanmun writers wishing to rectify their style and expression.
28 This presumably includes advanced writers of Hanmun and is not limited to learners at an early stage or to children. The explanatory texts are written in a Hanmun style that makes heavy use of
han’gŭl annotations. Im Sangsŏk explains that this style should be understood as “Sinicized Korean” (
Kukhanmun).
29
Chŏng Ik’s 1929
Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑 is aimed at the teachers of children and beginners in Hanmun learning.
30 He even includes a list of topics to teach for children beginning at six years of age until eight years of age – thus planning for his course to be taught over a duration of three years in total.
31
According to Chŏng Ik’s textbook, instruction is also supposed to include
han’gŭl (called “
Chosŏnmun” 朝鮮文 by Chŏng Ik) in the first year of instruction, teaching one letter per day. The process is laid out inconsiderable detail, as are the teaching methods for
han’gŭl.
32 Remarkably, this contradicts the widely-cited idea that
han’gŭl is so easy that it can be picked up in a morning. It is also very possible that Chŏng Ik realistically assumed that teachers were not used to teaching
han’gŭl and that they would need guidance on how to do so effectively. But the curriculum laid out by Chŏng Ik generally seems to be relatively slow in its progression. As the students progress, they would be gradually introduced to longer, more complex phrases and also to the systematical explanation of various
hŏsa and their use.
33 He allots a total of 300 days of instruction for his course, which seems realistic and would leave time for some additional explanations.
34 Also, Chŏng Ik stresses the importance of repetition and allots time to this.
35 In addition, students are supposed to study the
Kyemongp’yŏn 啟 蒙編, the
Xiaojing 孝經, the
Lunyu 論語, and the
Mengzi 孟子.
36 It is not clear, when and how exactly these texts should be studied and no guidance is supplied by Chŏng Ik. Very likely he assumed that teachers would know these texts (by heart even) and would have their own ideas on how to teach them.
As expected for language textbooks, the focus of most is on teaching elementary skills to schoolchildren. However, there is some variation in how this goal is reflected in different textbooks. The content of the textbooks is also influenced by their different motivations. The content of the textbooks differs according to their different motivations and target groups:
|
Title |
Content |
|
Wŏn Yŏngŭi 元泳義 (1908): Ch‘odŭng changmunbŏp 初等作文法. |
Condensed explanation of syntax, word classes and function words, mostly through example sentences without further annotation or explanation. Meant to be a textbook, providing materials to be used in the classroom with explanations being added by the teacher for every sentence. |
|
Sin Chaeyŏng 辛在英 (1916): Hanmun ŭidok chahae 漢文義讀自解. |
No table of contents. Short list of grammatical items at the start, then mostly just (partially annotated) practice sentences. Meant to be a textbook with additional explanations being provided by the teacher. |
|
Kang Mae 姜邁 (1917): Hanmunbŏp cheyo 漢文法提要. |
Systematic explanation of word classes, syntax, and of function words. Includes example sentences for the various function words. |
|
Yŏngch’ang sŏgwan p’yŏnjip pu 永昌書館編輯部 (1921): Silji ŭngyong changmun taebang 實地應用作文大方. |
Systematic explanations for word classes, syntax, function words, and stylistics. Also includes explanations on genre. Extensive example texts and reading section. All Hanmun texts are annotated with t’o. |
|
Chŏng Ik 鄭 瀷 (1929): Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑 and Hanmun kyosu ch‘ŏpkyŏng chusŏk 漢文教授捷徑註 釋. |
Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑: Collection of sample sentences for the explanation of style. Also includes explanations on function words. |
|
Hanmun kyosu ch‘ŏpkyŏng chusŏk 漢文教授捷徑註 釋:Vocabulary and additional materials to supplement the main volume. The vocabulary includes function words and their explanation. |
The work was criticized in China for incorrectly applying Western methods to the Chinese language, thereby imitating Western grammar. This criticism is interesting because it touches on something. There may be a certain pedagogical value in treating Literary Sinitic as one would Latin, which could be useful for those with a background in Western classical philology. However, relying closely on Latin and Greek grammar to explain the structure of Literary Sinitic is not without problems, since it easily leads to the structures of the target language being confused with what is actually found in Literary Sinitic texts. The fact that a certain Literary Sinitic structure should be translated in a certain way into Greek, Latin or English does not mean that the underlying grammatical features are identical.
In any case, Ma Jianzhong should most likely be exonerated from these charges, since his grammar does not mention any of the numerous Chinese grammars in Western languages that existed during his time. He clearly wanted to create something new based on Latin and Greek grammars due to the lack of a descriptive model.
37 However, the observation that forcing Western grammatical models onto the Chinese language could be problematic is worth bearing in mind.
The concepts that the early 20th cent. authors use are often difficult to understand and sometimes seem to not make much sense at all. Victor Mair has noted that he finds Ma Jianzhong‘s
Mashi wentong very hard to understand – so perhaps it is no surprise that the same is true for early Korean Hanmun grammars.
38 Maybe it is not unusual that the beginnings of grammatical traditions are a bit opaque, since they still have to agree on terminology and methods and are still in an explorative phase.
Defining hŏsa
The problem of how to define
hŏsa is by no means solved. For example, W.A.C.H. Dobson in his
Dictionary of the Chinese Particles includes a large number of words and usages that most grammars would treat as adverbs.
39 Adverbs such as
yŏk 亦,
u 又 and
i 已 would be considered
hŏsa by most grammarians.
40 But Dobson includes words that nearly everyone would obviously class as
silsa, such as
chil 疾,
41 si 時,
42 and
kyŏng 輕
43 and even verbs such as
kŏ 居
44 and nouns such as
in 人.
45 Of course, Dobson classes these words as what he calls “auxiliaries”, that is words that can modify verbs. And, of course, his system of word classification makes no reference to the
silsa/
hŏsa paradigm at all. The reason this is worth mentioning is because it illustrates the incompatibility of the traditional and arguably more advanced systems of word classification. It also points towards an intrinsic problem of the
hŏsa/
silsa categorization, which is that this system does not normally allow for the variation of usages and would surely class
chil 疾,
si 時,
kyŏng 輕,
kŏ 居 and most definitely
in 人 as
silsa.
The marker 所 is a good example of an elusive element of Chinese grammar that is difficult to explain in the terms of the Latin-based grammar. It is also notoriously difficult to gloss for this reason.
46 Most scholars today agree that 所 is most likely a pronoun, referring to the object of a verb and replacing it, such as Gassmann
47, Bruneton
48 and Pulleyblank.
49 Michael Fuller in his
Introduction to Literary Chinese has suggested that it should be analyzed as an adjective.
50 Perhaps because he wants to avoid a word class of function words. Functionally, however, it seems much more appropriate to treat 所 as a nominalizing function word, which turns a verbal phrase into a nominal phrase and thus forms groups of words that can fulfill syntactical functions such as subjects, topics, or objects.
Edwin Pulleyblank too only gives the explanation as a relative pronoun in the index to his book, whereas the explanation of the function of 所 is under the heading of “nominalization”.
51
W.A.C.H. Dobson explains 所 as an “indefinite substitute”
52, which is an elegant solution. On the one hand, this avoids framing 所 as a pronoun – even though one could argue that his term implies a pronoun without actually calling it that. On the other hand, “indefinite substitute” is an accurate description of what 所 does functionally.
For all these reasons, and because it is relatively difficult to explain, 所 seems a good example. So let us look at how 所 is treated in early modern grammars from Korea:
In the 1921
Silji ŭngyong changmun taebang 實地應用作文大方, the character 所 is explained as an
ŏjŏsa 語助詞 (“function word”) and examples are given for its usage.
53 However, there is no systematic explanation of the function of this character. Chŏng Ik lists 所 as part of his treatment of
ŏjŏsa54 He also only gives examples, but there is no explanation of the function of the word.
Kang Mae avoids a systematic explanation of 所, but he explains the construction “爲 …… 所 ……” as a marker of the passive voice.
55 He also mentions that the passive can be expressed by 所 alone.
56
All these explanations seem useful, but also incomplete. It appears that the early modern Korean Hanmun grammars exhibit some reluctance to address the more challenging aspects of Hanmun grammar, such as the wealth of usages of the marker 所. However, it should be noted that in many cases, the character can simply be omitted from the example sentences without affecting their meaning, thus functioning as an ‘empty word’ in the true sense of the term.
Hŏsa and Word Classes
The concept of hŏsa/silsa is used in many different ways. Different authors in different texts have different ideas about how hŏsa/silsa work and how they should be classified. Let us look at some examples from my sample of early modern Korean Hanmun textbooks.
For the unnamed authors of the 1921 Changmun taebang, hŏsa encompasses the following categories, not all of which are easy to find equivalents for in current terminology:
|
Name of the category |
Content of the category |
|
kiŏ hŏja 起語虛字 |
Modal particles initiating new information 57
|
|
chŏp’ŏ hŏja 接語虛字 |
Conjunctions, modal particles and pronouns referencing old information 58
|
|
chŏn’ŏ hŏja 轉語虛字 |
conjunctions of various grammatical relations 59
|
|
ch’in’ŏ hŏja 櫬語虛字 |
markers of nominalization, case markers of locative, instrumental, tempus markers of the future, markers of intention etc. 60
|
|
sog’ŏ hŏja 束語虛字 |
Only two entries: chong 総 (“in total”) and taeryak 大約(“generally”). Explained as words for summing up old information at the start of a change of topic. 61
|
|
hŏr’ŏ hŏja 歇語虛字 |
Sentence end markers of various sort, marking modality, tempus, aspects. Also includes markers of exclamation, emphasis etc.. Interestingly, it is also stated that these includes both hŏsa and silsa, but no examples for silsa are given. 62
|
As we can see, the focus here is not so much on functional categories as on where these words appear in the text.
63 This partly corresponds with the modern linguistic concept of focus and the shift between left and right focus, or the shift in emphasis from new to old information. While there is some awareness of the differences in the function of different word classes, the focus is clearly on providing a practical description of where these words appear and how they are used in text composition. The aim is didactic rather than strictly analytical, and the goal is to help readers improve their use of these
hŏsa in their own writing. Interestingly,
hŏsa and
silsa are not so much understood as grammatical terms as they are as structuring elements in the sense of rhetorical categories. Aesthetic choices appear to be an important factor here, which is not unusual in Literary Sinitic.
64
Kang Mae in his
Hanmunbŏp cheyo 漢文法提要 (1917) does only briefly mention the
hŏsa/
silsa system:
65
There are in fact up to 50,000 characters in Chinese writing (hancha), but these can generally be grouped into only two categories: Empty (hŏ) and full (sil). If we want to conveniently group these characters in a more detailed way, then we have nouns (名字)， pronouns (代字)， adjectives (形字)， verbs (動字)， adverbs (副字)， prepositions (介字)，conjunctions (連字)， particles (助字)，and interjections (歎 字).
漢字의 數는 實노 (sic!) 五萬以上에 達하나 此를 大分하면 虛實兩種에 不過하고 便히 此를 細別하면 名字，代字，形字，動字，副字，介字，連字，助字，歎字 等이 有할 而已라
The term ‘silja’ describes those [characters] that have something tangible that can be understood, while the term ‘hŏja’ describes those [characters] that do not have something tangible that can be explained but that are used to help with the [explication of the] quality and state of silja.
實字는 凡字의 可解할 事理가 有한 者를 云함이오 虛字는 可解할 事理는 無하나써 實字의 情態를 助하는 者라
名字，代字，形字，動字，副字，and 歎字 belong to silja. 介字，連字，助字 belong to hŏja.
名字，代字，形字，動字，副字，歎字等은 實字에 屬하고 介字，連字，助字等은 虛字에 屬하니라
His grammar is based on a system of nine word-classes (if we accept that 字 refers to “words” here), which are clearly derived from Western grammars. These word classes have many subclasses, creating a highly elaborate system that closely resembles the categories found in present-day grammars.
66
Chŏng Ik in his 1929
Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑 stresses that a good introduction to learning Hanmun is to memorize famous sayings and phrases. Grammar comes after this, with syntax being covered first. He leaves the systematic teaching of
hŏsa until a later stage.
67 Hŏsa are introduced without the
hŏsa –
silsa dichotymy, probably assuming that the reader is familiar with the concept.
68 Since the audience of his book are teachers, this brevity is to be excused.
In Wŏn Yŏngŭi’s 1908
Ch‘odŭng changmunbŏp 初等作文法, the intention behind explaining
hŏsa is stated at the beginning of his introduction: While
silsa are many, they are easy to understand. As for
hŏsa, they are few but difficult to understand and should thus be explained in more detail.
69 In his
Ch‘odŭng changmunbŏp, the term
hŏsa is only used as a general term in the introduction, while the actual treatment of Hanmun grammar later in the book uses the technical terms of the Western linguistic tradition. Wŏn Yŏngŭi uses five word classes that could be associated with
hŏsa:
70 hyŏngyongsa 形容詞，
pusa 副詞，
chŏpsoks 接續詞，
t’ansa 歎詞，
chisa 止詞. One would expect that they should correspond to the following word classes in modern terminology:
|
hyŏngyongsa 形容詞 |
*adjectives |
|
pusa 副詞 |
*adverbs |
|
chŏpsoksa 接續詞 |
*conjunction |
|
t’ansa 歎詞 |
*interjections |
|
chisa 止詞 |
(unclear category, very unusual) |
However, they do not. Instead they are as follows:
|
hyŏngyongsa 形容詞71
|
words describing qualities; covers many adjectives (or qualitative verbs), some adverbs and even some nouns which are used as the object of the verb 如 |
|
pusa 副詞72
|
adverbs expressing different degrees and comparisons, very complex system of describing five different types of relative degree. |
|
chŏpsoksa 接續詞73
|
defined as such hŏsa (function words) that the silsa that are connected by them depend on entirely. Examples given are 諸 (clearly meant to be the prepositional (plus pronoun) usage that we would today explain as standing for 之於)74, 乎 (as locative)75, 以 (as an instrumental, both as a preposition and as a postposition, the later arguably being a marker of the adverbial use)76, etc. But also includes conjunctions such as 而.77
|
|
t’ansa 歎詞78
|
very short category, exclamations: only 嗚乎，噫，吁. |
|
chisa 止詞79
|
Sentence end markers, including exclamative markers. |
This system is difficult to fully make sense of, and I have failed to do so so far. Kim Yonghan has given a detailed treatment of these word categories in his study on early Korean Hanmun grammars.
80 He also states that some categories are quite different from what would be expected and that there is no direct equivalent in modern terminology for
chisa 止詞.
81
However, there are some things we can learn from his categorisation that may still be useful today. For instance, his impressionistic use of terminology and grammatical categories seems to be more an attempt to describe the function of words in a sentence than to attach systematic categories to them. It is also clear from his grammar model that words change function frequently in Literary Sinitic, and that all labels attached to them are temporary.
Furthermore, Wŏn Yŏngŭi’s grammar arguably has the advantage of describing what he actually found in the language, rather than attempting to find things in the grammar of Literary Sinitic that one might expect to find because they are present in Western languages.
Kang Mae, in his
Hanmunbŏp cheyo, gives some explanations of changes in word class, which he assumes happen when words are used in another than their original word class. For example, nouns can be used as verbs. His example is a sentence in which
pyŏng 兵 is used as a verb, not as a noun:
82
左右欲兵之 “(Those to the) right and (those to the) left wanted to attack him.”
Kang Mae assumes that words have an original part of speech; in this case, he assumes that pyŏng is a noun. However, this assumption is far from self-evident for Literary Chinese. He does not explain how he determines their primary word class, presumably assuming that proficient readers of Literary Chinese have a good understanding of how words and characters are used in most texts. So the details are left for the reader to figure out himself.
Another interesting case is that of characters that can be either
hŏsa or
silsa. The 1921
Silji ŭngyong changmun taebang 實地應用作文大方, gives examples for the
silsa-use of words that are otherwise explained as
hŏsa. For example, 所 is explained as an
ŏjŏsa 語助詞 (“function word”), but also as a
silsa meaning place (“方所”).
83 And the character
che 諸 is explained as a
silsa meaning “many, all” (chung 眾).
This second usage illustrates the problem of discerning between silsa and hŏsa: While che as a marker of the plural would be analyzed by most modern grammars as a function word, this is not the case here – apparently the meaning of che is still tangible enough to allow for its treatment as a hŏsa.
Some Grammatical Phenomena and Explanations
Equally diverse are the didactic approaches taken in the textbooks. Here are some examples to illustrate the diversity of approaches and some of the unusual concepts tried in the textbooks of the early 20th cent.
Comparison
There is a very useful collection of forms of comparison and expression of degree in the
Ch’odŭng changmunbŏp.
84 The collection is not systematic in terms of grammatical categories, but gives a comprehensive overview of the different ways to express different qualities in absolute terms and in relation to each other. The aim clearly is didactic and the result is of admirable clarity, even in the somewhat experimental terminology employed in that textbook.
Sin Ch’aeyŏng in his 1916
Hanmun ŭidok chahae 漢文義讀自解 has a very peculiar idea for ordering the pronouns:
85
|
kŭnch’ing 近稱 |
ch’a 此 |
|
chungch’ing 中稱 |
ki 其 |
|
wŏnch’ing 遠稱 |
p’i 彼 |
Although the nature and usage of pronouns in Chinese has been the subject of much research, there is nothing to suggest such a system. It is much more likely that this reflects the Korean or Japanese system of (demonstrative) pronouns, which has a clear structure for close, far and very far references, as set out in the following table:
|
Korean |
Japanese |
Chinese |
|
kŭnch’ing 近稱 |
이 (i)
|
こ (ko)
|
?此 (cǐ)
|
|
chungch’ing 中稱 |
그 (kŭ)
|
そ (so)
|
?其 (qí)
|
|
wŏnch’ing 遠稱 |
저 (chŏ)
|
あ (a)
|
?彼(bǐ)
|
I would argue that neither Literary Sinitic, nor modern Chinese does have the chungch’ing 中稱 category. Also, ki 其 seems a very strange candidate for this category. While sometimes used as a demonstrative pronoun, 其 typically is used as an attributive pronoun, functionally replacing any other pronoun, regardless of near or far, plus the attributive marker chi 之.
It seems that some pronouns in Literary Sinitic are more often used to refer to old information, while others are used to introduce new information. Robert Gassmann analyzes
si 是 and to a lesser extent
ch’a 此 as anaphoric pronouns; that is “backwardspointing pronouns”, which refer back to old information. The pronoun
p’i 彼 on the other hand is analyzed by Gassmann as a cataphoric pronoun; “a forward-pointing pronoun”, which refers to new information that is not yet given in the text and is introduced by the pronoun.
86
The categories of kŭnch’ing 近稱 and wŏnch’ing 遠稱 present no major problem, especially since they can also be understood as referring to old and new information and not only to spatial relations. But chungch’ing 中稱 is difficult to explain from Hanmun grammar. While a systematic comparison of the various possible sources for this is beyond the scope of this paper, it is more likely that this is taken from Korean or Japanese grammar textbooks and then applied to Hanmun; arguably without any intrinsic necessity to do so. The pronoun 其 is most likely chosen because of its somewhat elusive character and because it does not fit other categories.
Interestingly, Sin Chaeyŏng proposes a dual system of second person personal pronouns for Hanmun grammar:
87
|
Form |
pronouns |
|
“normal address” |
i 爾，yŏ 汝 |
|
“respectful address” |
kong 公， kwiha 貴下 |
The idea of a twofold hierarchy of pronouns in Hanmun seems strange, as does the selection of examples. Kwiha 貴下 is not frequent in pre-modern Chinese and is most likely a mistake, since the word is frequent in literary forms of Japanese (and is also still used in formal Korean today). Apart from this strange choice for an example, the concept itself is highly doubtful, as Literary Sinitic has a complicated system of addressing by pronouns and pseudo-pronouns (titles, offices, names etc.), which far exceeds the binary system laid out by Sin Chaeyŏng.
The assumption of a dual system is of interest because there is little reason to suggest such a system for Chinese – there are many more forms that range somewhere in between; kun 君 being the obvious example. In Korean, the matter is even more complicated. It seems likely that the dual system is an import from the grammar of European languages, such as German or French, which have a relatively clearly defined dual system of formal and informal personal pronouns.
Sin Chaeyŏng's analysis is most likely based on the grammar of Western (Indo-European) languages, which is why he uses such a binary system. Such a system is found in languages such as French or German, where a binary differentiation between two sets of forms of address exists and is easily defined by different morphological forms. Literary Sinitic, on the other hand, is much more complex in its expression of social relations and closer to Korean or Japanese – even though Literary Sinitic expresses these relations through the choice of words and not through grammatical forms, as Korean and Japanese do. It seems that Sin Chaeyŏng’s idea here is informed by his knowledge of Western grammar, which is then – mistakenly – applied to Hanmun.
Thus we have two surprising choices for a Hanmun textbook: The form kwiha 貴 下, which is not Chinese and the binary system of address, which is not Chinese either. Apparently, both these forms seem useful enough to Sin Chaeyŏng to include them, even though they are not Hanmun in the narrow sense.
Cases and null subjects
Kang Mae explains casus (
kyŏk 格) as part of his treatment of the word classes. Interestingly, he explains the casus of nouns and pronouns separately, even though there is obviously no difference between a noun or a pronoun acting as a subject or object.
88 It is not quite clear where he gets this idea from; possibly this is due to grammars of Indo-European languages having separate lists for casus in nouns and pronouns.
Another interesting feature is his treatment of null-subject sentences, which he calls the “empty case”
hŏgyŏk 虛格. He provides several explanatory examples, which give the impression that he may also be including other forms of ellipsis into his
hŏgyŏk category.
89
Most of the textbooks also use some form of word order markings as a didactic tool. The idea is to give the reader the word order that the sentence would have in Korean, so as to make translation easier. This is very likely to be based on Japanese traditions of annotating Literary Sinitic texts, such as the use of kaeriten 返り点 marks to express word order.
The
Hanmun kyosu ch‘ŏpkyŏng has a very simple system, only hinting at the reading. For example:
90
(He) sees swallows and (he) hears a hawk
見 (二)燕(一)聞鷹
The marks 一 “one, first” and 二 “two, second” are meant to explain the sentence structure in the Korean translation. In the first example, the verb 見 comes before the object 燕 in Chinese, but the structure would be inverted in Korean word order, where the verb (見) would stand at the end of the sentence, while the object (燕) would come before the verb. The second clause (聞鷹) is syntactically identical to the first clause and is not explained again for this reason. The reason why the markings are added here apparently is that the two clauses could also be misread as attributive constructions otherwise, in which a verbal attribute (見 or 聞) would be attributed to a noun (燕 or 鷹), thus being “the seeing swallow” (or: “the swallow that sees”) and “the hearing hawk”.
Another example on the same page is
91
peaches and plums
lacquer trees and willows
桃(一)與(二)李(三)梧及柳
Again, the second clause is left without markings, as the reader is assumed to understand that the examples are meant to explicate the function of 與 and 及 as markers of coordination (“and”). The word order is unchanged, as it would be the same in Korean. The markings are nevertheless necessary to avoid 與 and 及 being misread as verbs (“to give” and “to arrive”), however unlikely that may be.
Context and grammar teaching
Chŏng Ik is of the opinion that
hŏsa are best taught integrated into the texts and explained en passant.
92 This is a very modern approach, even though the practice in his
Hanmun kyosu ch‘ŏpkyŏng does not deliver on its promises, as explanations are lacking in most parts of the textbook.
Chŏng Ik gives an overview of his vision on the teaching of hŏsa in his chapter on “grammar” (munbŏp 文法). Teaching should be by two didactic tools: Example sentences and patterns.
The example sentences are explicated by inserting t’o 吐 annotation in han’gŭl script into the Hanmun text. T’o annotation is a system by which Korean grammatical forms (verb forms, casus markers etc.) are are inserted into a Hanmun text to make it easier for Korean readers to parse the grammar. T’o also helps with ad-hoc translations of the Hanmun text into Korean, since it provides most of the information needed to form a Korean sentence; with complete t’o given for a Hanmun text, only the word order needs to be readjusted to form a Korean text.
Patterns are explained by giving the
hŏsa and its relative position to other syntactical elements. These are given in direct reference to the example sentences, thus also serving as a reminder of what parts of the example sentences are
hŏsa. For example:
93
[1] As for learning, its value lies in thoroughness. And that by which thorough knowledge is achieved, is due to having books available (...)
[2] Thus 〇〇 is because of 〇〇〇
[1] 學은貴乎博하니所以博識者난以其有書也ㅣ라 (...)
[2] 所以〇〇者以〇〇〇也
In this example, the explanatory second line replaces all non-hŏsa words with a null symbol (〇), indicating a position where silsa can be inserted into the pattern. While it may seem strange that Chŏng Ik insists on a specific length for the elements to be inserted, this is a very useful explication in his model: Since his explanation of Hanmun syntax is in the form of fixed-length sentences, he emphasizes the stylistic and metrical dimensions of Hanmun. Thus they even take precedence over the explanation of grammar itself, which could be taken as a case of highlighting aesthetics as a primary goal of Hanmun education.
Although it would arguably be more useful for modern Hanmun learners to be introduced to more generalised patterns, the idea that hŏsa should be taught with attention to the patterns in which they appear in sentences remains a highly useful approach. Chŏng Ik’s textbook is a useful repository of such patterns, complete with example sentences, even if these are arguably too difficult in some cases.
Hŏsa/silsa and stylistics
Style and genre are important categories to decide the appropriate use of language. And these are highly regulated in Hanmun. It is most likely for this reason, that the 1921 textbook by the Yŏngch’ang sŏgwan p’yŏnjip pu includes a chapter on genre.
94 Hŏsa are what can be left out for prosodic purposes (虛 as in the 虛格 category above).
To illustrate the relevance of Hanmun for present-day Korea, let us take an example from contemporary use of Hanmun in South Korea:
In 2024, readers of the
Kyosu Sinmun have picked
to ryang pal ho” 跳梁跋扈, meaning “acting madly while abusing one’s power” as the expression that best describes Korean society in 2024. The background for this decision obviously is the reign of president Yun Sŏkyŏl’s
95 administration. The results of the election were published on 9 December 2024, just six days after president Yun Sŏkyŏl had tried – and failed – to declare martial law in the country. Thus, the classical language (Chinese, not Korean) is used to comment on and to voice criticism concerning contemporary politics.
The sŏngŏ scoring the second place was hu an mu ch‘i 厚顔無恥, meaning that someone has, a thick face and no shame“, another obvious reference to president Yun Sŏkyŏl.
This was followed in the third place by the sŏngŏ sŏk sŏ wi ryŏ 碩鼠危旅, which means “a single vermin endangers the whole army”, which, again, seems to be a reference to the political situation.
Our example comes in fourth place: ka chŏng maeng ŏ ho 苛政猛於虎. This is the most overtly political choice among the top-scoring sŏngŏ: “cruel government is worse than a tiger”. This is exactly as blunt a statement as it reads and there is no hidden meaning here. While this type of sŏngŏ is typically made up of four characters, this one has five characters. This is possible because the character ŏ 於, which is a hŏsa, can be ignored for the purposes of prosodic conformity, so that this sŏngŏ can be considered to be a four character phrase.
For completeness, the fifth and last place, scoring 5.7 percent of the vote, is the only
sŏngŏ in the list that is not overtly political.
pon rip to saeng 本立道生 means “if the roots are firm, a way will be found”. This can be understood in a number of ways, possibly also as a a critique of president Yun’s government.
96 Alternatively, it could be interpreted as a call for stability as the foundation on which all further progress is built, and for the people to remember their common interest.
Here are the sŏngŏ again in an overview:
跳梁跋扈
厚顔無恥
碩鼠危旅
苛政猛於虎
本立道生
Obviously, 苛政猛於虎 stands out from the rest by being one character longer. This is possible because 於 is a hŏsa and thus does not strictly need to be counted for prosodic purposes.
It is important that students of Hanmun are at some point introduced to the concept of hŏsa and silsa, so that they can understand usages such as in the sŏngŏ 苛政猛於 虎: “Cruel gouvernment is fiercer than a tiger.”
As we have just seen, aesthetics, and accordingly stylistics play an important role in the composition of Hanmun texts. The authors of the
Silji ŭngyong changmun taebang 實地應用作文大方 (1921) accordingly devote much attention to style, and especially to genre – since stylistic conventions differ between the various genres of Hanmun literature. The following genres are included:
97
|
Genre (Original) |
Genre (English translation) |
Notes |
|
non 論 |
|
Described as a systematic inquiry into a topic, discussing every aspect of it. |
|
A large number of sub-categories is given, which deal with more specialized subjects, such as 史論 or 文論 – these sub-categories do not have their own entries. |
|
sŏl 說 |
explanation |
Described as a text which explains a subject, either of an abstract or of a material nature, and helps to clarify its meaning. |
|
chŏn 傳 |
biography |
Explained as the description of a person’s life, deeds, and significance, often also including moral aspects. |
|
ki 記 |
records |
Explained as a form of descriptive text whose function is to prevent that the even or things described may be forgotten. |
|
Travelogues (遊記), records on events (事 記), war records (戰 記), and miscellaneous records (雜 記) are mentioned as subcategories, even though they have their own entries (see the entries below). |
|
yugi 遊記 |
travelogue |
Explained as a description of noteworthy sights and events during a travel, especially a travel to visit famous places from history or from classical literature. |
|
sagi 事記 |
records (on an event) |
Explained as a description of noteworthy events, especially politics, extraordinary moral deeds, calamities, or unusual events and rumors of all sorts. Emphasis is on a focus on the important aspects and their detailed description. |
|
chŏn’gi 戰記 |
war records |
Described as a record of military numbers, strategies, battlefield formations, battles, commanders, etc. It is encouraged to add commentary to these records. |
|
War records is a very unusual genre, which is unexpected in this list. Presumably, the events of World War I and the public interest in war news may be the reason to include this as a genre. |
|
chapki 雜記 |
miscellaneous records |
Described as a genre of texts on artifacts, buildings, plants, animals and other “things” (!), their characteristics and their use and application. Again, the prospective author is encouraged to add commentary to this description. |
|
sŏ 序 |
preface |
Described as an explanatory text that accompanies other texts and which may also include appraisal or criticism. |
|
pal 跋 |
afterword |
Described as a genre that accompanies another text and comments on that text, providing corrections and also including appraisal and criticism. |
|
Interestingly, there is no mention of the complementary function of prefaces and afterwords, nor any mention on how they differ from each other. |
|
che 題 |
explication |
Described as a type of explanatory text that is placed at the beginning of another. It serves as an introduction and provides important facts about the text to the reader. |
|
sa 辭 |
verse |
Described as a genre that belongs to the category of songs. It usually expresses feelings such as anger, regret, sorrow, or longings. |
|
mun 文 |
(condolence) text |
Refers to condolence texts, such as 祭文 and 吊文 (both could be translated as “condolence texts”). Described as a text expressing sorrow, usually in verse. It is interesting that the very general 文, which could simply mean “text” in most contexts, is explained to exclusively refer to the very specific genre of condolence texts. |
|
sŏ 書 |
letter |
Refers to the genre of letters, addressed to other persons. The genre can also include writings addressed to the public (in the form of open letters). |
|
ch’an 賛 |
eulogy |
Described as a genre of praising a person’s achievements and fame, aiming at instilling a feeling of reverence in the reader. |
Some of the entries also have some historical remarks on when that genre started and what the formative texts are.
In addition, the genres of
song 頌
98 (also best translated as “eulogy”) and
myŏng 銘
99 (“inscription”, “epigraph”) are not in the list of explained genres, but they have entries in the later section of example texts. This is followed by a list of minor genres, which also includes epigraphs and commentaries. This final list appears to be much less systematic and seems to have been added for use in classes by experienced instructors, who would provide their own explanations. It is also likely that this final section was composed ad hoc with insufficient time for systematic editing.
Despite the shortcomings of the final section, the educational value of such a genre list cannot be overemphasized, since knowledge of genres is of paramount importance in Hanmun education. Despite this importance, it is very difficult to find a descriptive list of genres that is extensive and detailed enough to provide a comprehensive overview of Hanmun literary genres while also being concise enough to be practical for use in teaching Hanmun. While the textbook does not claim this list to be exhaustive — in fact, a number of genres are explicitly listed as missing
100 – it is still one of the most complete (and concise!) lists of Hanmun literature genres published in Korean.
The list is still useful for teaching Hanmun today. Although it needs to be translated from the original Kukhanmun style into a modern language, it is a valuable resource for students of Hanmun and should be included in some form in Literary Sinitic textbooks and courses. Even the very short English-language overview of the content provided here should suffice to illustrate this point.
Chŏng Ik also acknowledges the importance of style. He begins his textbook with a collection of phrases, presumably for rote learning. These phrases, some nominal or verbal and some full sentences, are divided by length, from two to eight characters
101 While not very systematic, they are building blocks for more complex structures, and learning them by length automatically instils in the student a sense of the rhythm and stylistic peculiarities of Hanmun writing.
Although this method of rote learning vast quantities of text is not acceptable to contemporary tastes, especially since some of the choices are dubious, the Hanmun kyosu ch‘ŏpkyŏng is still a treasure trove of example phrases that could be used in modern Hanmun courses. They also serve as a reminder that style trumps grammar, and that many genres of Hanmun allow for considerable freedom. This is illustrated by the following couplet of two phrases:
The fish travels the water.
The bird sings (in) the forest.
鳥啼林
The second sentence (魚游水) is unproblematic and adheres to a simple subjectverb-object pattern. The first sentence (鳥啼林), however, should include a locative marker (like 於) in standard prose usage. However, this is not standard prose usage, but either a poetical use of language, where the “forest” (林) is the object of the verb “to sing” (啼), or “forest” is in fact intended to be in the locative, but the marker has been dropped to achieve the desired three-character phrase.
Obviously, the goal of teaching with the help of such phrases is not an analytical understanding of grammar, but rather an intuitive usage focusing on the stylistic and aesthetic aspects of writing. Although this focus on style, especially metrics, is unusual for most modern readers, it is typical of Hanmun writing. It highlights the need to prepare students of Hanmun for apparently ungrammatical constructions, which are best explained by the necessities of style.
T’o, kugyŏl and reading Hanmun
Finally, a brief note on Korean glossing is in order. The early 20th century was a time when everything could be found. It is possible to find Hanmun with no glossing, with heavy t’o glossing, and even kugyŏl can still be found in some textbooks.
With the possible exception of kugyŏl, which has fallen out of use, the question of whether or not Korean Hanmun should use t’o remains unresolved to this day.
A method that we can find in some of these grammars, which seems to have been popular in the early 20th cent. is that ‘f re-arranging the word order of the Chinese text. This is very similar to the Japanese method of reading, kanbun kundoku 漢文訓読. It seems highly likely that the popularity of this method in Korea is due to Japanese influence. Another possible source is the Korean tradition of idu – the idea of using Chinese characters to depict a text in Korean language and thus in Korean word order would not have been totally alien to Korean Hanmun users.
The system, as it is laid out in Sin Ch‘aeyŏng‘s 1916
Hanmun ŭidok chahae 漢文 義讀自解 is called
ŭidok 義讀, as in the title.
103
To explain the method of re-ordering the text, he introduces a pair of terms:
sun 順 and
yŏk 逆, or
sundok 順讀 and
yŏkdok 逆讀. The concepts are explained as follows:
104
“[There are two methods for] reading Hanmun according to the meaning: [These] are with the flow and against the flow.”
漢文義讀有順有逆
“Reading with the flow means that the characters are lined up in descending order, following the flow. Reading against the flow means that the characters are lined up in ascending order, going against the flow.”
順者字列之降順也逆者字列之昇逆也
Sin Ch’aeyŏng then proceeds to give a list of how word classes correspond to these
sunja 順字 and
yŏkcha逆字, according to whether they are read with or against the flow:
105
|
myŏngsa 名詞 |
nouns |
sundoksa 順讀詞 |
|
taesa 代詞 |
pronouns |
々 |
|
hyŏngyongsa 形容詞 |
adjectives |
々 |
|
t’ansa 歎詞 |
exclamations |
々 |
|
pusa 副詞 |
Adverbs |
々 |
|
chŏpsoksa 接續詞 |
conjunctions |
々 |
|
chadongsa 自動詞 |
intransitive verbs |
々 |
|
chonggyŏljodongsa 終結助動詞 |
sentence final adverbs |
々 |
|
t’adongsa 他動詞 |
transitive verbs |
yŏktoksa 逆讀詞 |
|
kadongsa 加動詞 |
*function words |
々 |
|
chodongsa 助動詞 |
*function words |
々 |
|
kajodongsa 加助動詞 |
*function words |
々 |
Conclusion
Although the hŏsa/silsa paradigm is an outdated grammatical concept, it still has some didactic value and should be taught in introductory Literary Sinitic courses. In practical terms, it is useful because it provides a flexible framework for initial analysis. Understanding hŏsa and silsa can solve most fundamental problems in understanding the structure of Hanmun texts, making it highly useful.
Apart from the hŏsa/silsa paradigm, these old textbooks contain other valuable resources, such as useful diagrams and compilations of usage patterns for grammatical forms. These textbooks also provide example sentences for all purposes of Hanmun teaching. Once the reader has navigated the antiquated language, they will find the explanations remarkably clear. The ideas about pattern recognition and the tendency of some hŏsa to appear together are especially useful and should be kept in mind when planning new textbooks.
It should also be noted that these old textbooks are not just literature on Hanmun; they are also valuable sources of the language, written or compiled by highly competent native speakers. So they are also literature in Hanmun. While they can provide useful example sentences for general Hanmun teaching, they can also be used to investigate Hanmun usage in the early 20th century.
As I have noted, these grammars are not easy to understand, but I do not intend to mock the clumsiness of these early grammars – quite the contrary. I want to emphasize that we are deeply indebted to these pioneers and their work. Without it, no progress would have been possible. Furthermore, I would like to highlight that we can still learn from them today, and even their mistakes — or perhaps their zeitgeist-driven inventions — can serve as valuable sources of inspiration and materials for teaching and research.
The fact that early 20th-century grammars grappled with many of the same issues as contemporary grammars is a useful reminder that some basic didactic problems in the teaching of Hanmun remain unsolved. Notably, the issue of how to describe Literary Sinitic grammar and relate it to Latin grammar remains unresolved.
Meanwhile, the situation has changed in that knowledge of Latin cannot be presumed among Western students. English has taken over the role of Latin as the universal language of scholarship. Alongside the reality of contemporary language teaching, this means that many Western students start their Hanmun classes without any systematic training in language grammar. While this presents many challenges, it also offers an opportunity to liberate Hanmun education from the constraints of the Latin tradition. Early East Asian grammars of Literary Sinitic, such as those discussed in this paper, offer insight into how such a system could be structured.
When working with learners who have a background in Korean or Japanese, it is often much more useful to explain Literary Sinitic grammatical features in reference to similar phenomena in those languages than to look for an explanation in terms of Latin grammar traditions.
Although the concept of hŏsa/silsa encompasses a wide range of ideas in early modern texts, there is still some common ground. For all authors, silsa are those words that add substantial meaning to the text and cannot be removed without profoundly altering its semantics. Hŏsa, on the other hand, are words added to provide grammatical clarity or fulfill aesthetic rules. Slighty exaggerating, one could say that the difference between hŏsa and silsa is that hŏsa can be removed without altering the core meaning of the text, whereas silsa constitute the text itself. Obviously, removing hŏsa changes the explicit syntactical structure of the text and thus increases the number of its possible interpretations of the (implicit) syntax. In extreme cases, this may even result in ungrammatical text – but the semantics expressed by the silsa would still be there.
We also need to consider the aesthetics of Hanmun texts and their influence on syntax, particularly with regard to the use of hŏsa. The rhetoric and requirements of literary form can be just as influential as grammar in determining the use of hŏsa. Therefore, pre-modern and early modern Hanmun pedagogy took style and form very seriously. Although the aim of our Hanmun classes is not usually to compose texts, it is still useful for students to understand the general principles that govern the composition of Hanmun texts. Some of the old grammars also seem to mix up grammatical, formal and aesthetic categories. This is because that is how the correctness of a text was judged. This is not to say that grammar is unimportant in Hanmun, but rather that we should discuss whether and to what extent aesthetic and stylistic conventions should be considered to be a part of Hanmun grammar.
Notes
Reference List
- Bruneton, Yannick. Hanmun. Grand Manuel de Chinois Classique. Malakoff: Armand Colin, 2024.
- Chŏng Sŏng’im. “Kungnae kanhaeng hanmunbŏp kyogwasŏ-ŭi pusa pŏmju yŏn’gu.” Han Chung inmunhak yŏn’gu 69 (2020): 337-356.
- Dobson, W.A.C.H. [William Arthur Charles Harvey]. A Dictionary of the Chinese Particles: With a Prolegomenon in which the Problems of the Particles are Considered and they are Classified by their Grammatical Functions. Toronto: University of Toronto Press, 1974.
- Fuller, Michael. An Introduction to Literary Chinese. Rev. Edition. Cambridge: Harvard University Asia Center, 2004.
- Von der Gabelentz, Georg. Chinesische Grammatik. Leipzig: Weigel, 1881.
- Gassmann, Robert/Behr, Wolfgang. Antikchinesisch. Bern etc.: Peter Lang, 2011.
- Hong Yubin. “Ch‘odŭng changmunbŏp-ŭi hŏsasŏl-e tae-han kŏmt’to-wa ihae (1) - myŏngsaryu-but’ŏ pusaryu-kkaji-rŭl taesang-ŭro” 初等作文法 의 虛詞說 에 대한 검토와 이해 (1) - 名詞類 부터 副詞類 까지를 대상으로. Taedong hanmunhak 大東漢文學 69 (2021): 311-341.
- Im Sangsŏk. “1920 nyŏndae changmun’gyobon Silji ŭngyong changmun taebang-ŭi kukhanmunch’e kŭlssŭgi-wa hanmunjŏnt’ong” 1920 년대 작문교본 實地應用作文大方의 국한문체 글쓰기와 한문전통. Uri ŏmun yŏn’gu 39 (2011): 463-489. ——. Hanmun, hanchagyojae-ro ponŭn singmint’ongch’i-wa han’gŭljŏnyong, ‘Hanmun kyosu ch’ŏpgyŏng sŏ’ (1929), ‘Hanjo yŏksa ch’ŏnjamun’ (1928), (Taehan min’guk) Saech’ŏnmun (1951)-ŭi kyŏng’u” 한문, 한자교재로 보는 식민통치와 한글전용, 「漢文敎授捷徑 序」(1929), 朝鮮歷史千字文(1928), (대한민국)새千字(1951)의 경우. Han’guk hanmunhak yŏn’gu 韓國漢文學研究 76 (2019): 105-134.
- Kim Yonghan 金容漢. “Ch’ogi kanhaeng hanmun munbŏpsŏ-e tae-hayŏ I, Ch’odŭng changmunbŏp-ŭl chungsim-ŭro” 初期 刊行 漢文文法書 에 對 하여 I, 初等作文法 을 중심으로. Kyonam Hanmunhak 嶠南漢文學 [=Taedong hanmunhak 大東漢文學] 3 (1990): 93-128. ——. “Hanmun ŭidok chahae-ŭi munbŏp yŏn’gu” 漢文義讀自解 의 文法 研究. Taedong hanmunhak 大東漢文學 8 (1996): 17-54. ——. “Hanmunbŏp cheyo-ŭi puja yŏn’gu” 漢文法提要의 副字 硏究. Taedong hanmunhak 大東漢文學 9 (1997a): 5-20. ——. “Hanmunbŏp cheyo-ŭi tongja yŏn’gu” 漢文法提要의 動字 硏究. Tongbang hanmunhak 東方漢文學 13 (1997b): 1-16. ——. “Hanmunbŏp cheyo-ŭi kudongnon yŏn’gu 1, kudok-ŭi sŏngnip-ŭl chungsimŭro” 漢文法提要의 구독론 연구 1, 구독의 성립을 중심으로. Taedong hanmunhak 東方漢文學 10 (1998a): 303-321. ——. “Hanmunbŏp cheyo-ŭi kaeja yŏn’gu” 漢文法提要의 介字 硏究. in Tongbang hanmunhak 東方漢文學 15 (1998b): 1-21. ——. “Hanmunbŏp cheyo-ŭi yŏnja yŏn’gu” 漢文法提要의 連字 硏究. in Tongbang hanmunhak 東方漢文學 16 (1999): 1-19. ——. “Hanmunbŏp cheyo-ŭi kudok hyŏngsik yŏn’gu” 漢文法提要의 句讀形式 硏究, Tongbang hanmunhak 東方漢文學 17 (2002): 257-278. ——. “Ch’ogi kanhaeng hanmun munbŏpsŏ-ŭi t’ongsa iron” 初期 刊行 漢文文法書의 統辭 理論. Hanmun kyoyuk yŏn’gu 漢文教育研究 25 (2005): 455-484.
- Kim Yŏngju 金英珠. “Kaehwa’gi hanmunkyojae-ŭi p’yŏnch’an ŭisik-kwa hyŏnjaejŏk sisa - Wŏn Yŏngŭi-ŭi hanmunkyojae py’ŏnch’an-ŭl chungsim-ŭro” 개화기 한문교재의 편찬 의식과 현재적 시사 元泳義의 한문교재 편찬을 중심으로. Tongbang hanmunhak 東方漢文學 70 (2017): 107-131.
- Mair, Victor. “Ma Jianzhong and the Invention of Chinese Grammar.” Journal of Chinese Linguistics Monograph Series 10 (1997): 5-26.
- McDonald, Edward. “The Challenge of a ‘Lacking’ Language. The Historical Development of Chinese Grammatics.” Chinese Language and Discourse 8, no.2 (2017): 244-265.
- Namgung Wŏn. “Kaehwagi hanmun munbŏpsŏ Ch’odŭng changmunbŏp-ŭi chŏjak paegyŏng-gwa ŭiŭi” 開化期 漢文文法書 初等作文法의 著作 背景과 意義. Hanmun kyoyuk yŏn’gu 漢文教育研究 26 (2006a): 329-256. ——. “Han-Il happyŏng chŏnhu kyogwasŏ kŏmjŏng-ŭl t’ong-han hanmun’gwa kyogwasŏ kaebal ŏkje silt’ae yŏn’gu.” Hancha hanmun kyoyuk 漢文教育研究 17 (2006b): 217-248.
- Pulleyblank, Edwin. Outline of Classical Chinese Grammar. Vancouver: University of British Columbia Press, 1995.
- Yi Kunsŏn 李君善. “Hanmun-gwa kyoyuk-esŏ hŏsa-ŭi taep’yohun munje.” Tongbang hanmunhak 55, (2013): 1-15.
- Yi Ŭnsŏn. “Wŏn Yŏngŭi-ŭi ‘Kungmun kwabon’ (1908)-ŭi t’ŭksŏng-gwa Hawai kug’ŏ kyogwasŏ-ro-ŭi hwalyong yŏn’gu” 원영의(元泳義)의 『국문과본(國文課本)』(1908)의 특성과 하와이 국어 교과서로의 활용 연구. Ŏmun nonjip 語文論集 9 (2022): 7-31.
- All the sources are cited from the Yŏkdae Han’guk munbŏp taegye 歷代韓國文法大 系series, which was published in 1986 by Kwangmunsa 光文社 in Seoul. In the following, they are cited by their original year and place of publication.
- Chŏng Ik 鄭瀷. Hanmun kyosu ch‘ŏpkyŏng 漢文教授捷徑. Kyŏngsŏng 京城: Unhyang sŏru 芸香書樓, Shōwa 昭和 4(1929a).
- Chŏng Ik 鄭瀷. Hanmun kyosu ch‘ŏpkyŏng chusŏk 漢文教授捷徑註釋. Kyŏngsŏng 京城: Unhyang sŏru 芸香書樓, Shōwa 昭和 4(1929b).
- Kang Mae 姜邁. Hanmunbŏp cheyo 漢文法提要. Kyŏngsŏng 京城: Pangmun sŏgwan 博文書館, Taishō 大正 6(1917)/.
- Sin Chaeyŏng 辛在英, Hanmun ŭidok chahae 漢文義讀自解. Kyŏngsŏng 京城: Sin Chaeyŏng ka 辛在英家, Taishō 大正 5(1916).
- Wŏn Yŏngŭi 元泳義. Ch‘odŭng changmunbŏp 初等作文法. Kyŏngsŏng 京城: Kwangdong sŏguk 光東書局, Yunghŭi 隆熙 2(1908).
- Yŏngch’ang sŏgwan p’yŏnjip pu 永昌書館編輯部. Silji ŭngyong changmun taebang 實地應用作文大方. Kyŏngsŏng 京城: Yŏngch’ang sŏgwan 永昌書館, Taishō 大正 10(1921).
News
