For the nearly fifty years that I have been studying Classical Chinese the one constant but unsatisfied desideratum of English-speaking students and scholars has been for a Chinese-English dictionary that focuses specifically on premodern texts. In the absence of such a reference work, those seeking such help have usually had to resort faute de mieux to the 1931 Mathews' Chinese-English Dictionary, prepared originally for the China Inland Mission and heavily indebted to Herbert Giles' 1892 Chinese-English Dictionary. The inadequacies of Mathews' (and Giles') dictionary for this purpose are well known and do not need detailed rehearsal. Probably the most troubling fact is that, although mostly focusing on late imperial and modern usage, it indiscriminately mixes together vocabulary of all periods, from the early layers of the Shang shu to early twentieth-century merchant and missionary vocabulary, and snatches of much else in between, with the unhappy result that students infer all terms and meanings to be equally applicable throughout three thousand years of Chinese history. The seemingly random arrangement of various meanings for any particular word is likewise unfortunate, leading to the pick-and-choose approach so maddeningly familiar to any instructor who hears students explain that after all they "got" this or that definition from Mathews', so how could it not be correct?
The magnitude of compiling a dictionary that might rectify or at least improve the situation has been a ready deterrent for a long time. The abandonment in 1955 of the Harvard-Yenching Institute project for a comprehensive Chinese-English Dictionary served to depress hopes substantially.1 Yet the scope of our wishes did not shrink, as what was usually spoken of as most desirable was an English equivalent of massive, encyclopedic dictionaries such as Morohashi Tetsuji's 諸橋轍次 Dai Kan-Wa jiten 大漢和辭典 (1955-60) or the Zhongwen dacidian 中文大辭典 (1962-68, rev. ed. 1973)2 or the Hanyu dacidian 漢語大詞典 (1986-94). Such a dictionary would be ideal, of course. But its completion would call for the dedicated work of dozens of scholars for more years than any of them would be willing to sacrifice, given the lack of funding for such a large and lengthy project and the realities of professional imperatives in American higher education currently.
Nevertheless, something better than Mathews' and the other available options is needed, especially for students in their first years of studying Classical Chinese. It is no use if we allow the perfect to be forever the enemy of the good. The Pareto principle may apply here. Hence, the present work which, imperfect as it is, hopes to make a start toward the ultimate dictionary we all desire.
A Student's Dictionary of Classical and Medieval Chinese is meant for practical use in reading and translating. It is not an etymological dictionary, nor is it a phrase dictionary (cidian 辭典). As a lexicon of some 8,200 individual graphs (thus, a zidian 字典), it aims to provide immediate assistance for the interpretation and translation of words in certain contexts. In this regard perhaps it is worth stressing that a Chinese graph, or character, is not itself a word. It is the representation of a word. Moreover, one and the same graph may represent different words, words that may either be related in meaning or that may be totally unrelated. Where possible, I have tried to arrange the contents of individual entries in a manner that may suggest a certain development of meanings or understandable progression from a basic sense to various derived meanings. But when a single graph is used to represent fundamentally different words, this is not possible and should not be expected.
The term "Classical Chinese" itself as normally used by scholars has two different references. The first is as a general, inclusive term for what might otherwise be called premodern Chinese, as when we speak of teaching a course in (that is, with texts written in) Classical Chinese. The second, as used in the title of this book, is as a narrower term for a specific stage of premodern Chinese, covering roughly the language used in texts dating from the Warring States period (481–221 BCE) through the Qin (221–206) and Han (206 BCE–220 CE) dynasties. Complementing this usage, "Medieval Chinese" refers to the language used in texts dating roughly from the third century to the tenth century, that is, from the Wei-Jin-Nanbeichao period (220–589) through the Sui (589-618) and Tang (618-907) dynasties.
Specialists in Chinese historical linguistics make finer distinctions and divide the language of this millennium and a half into several more stages. But the twofold division into "classical" and "medieval" Chinese, with its transition during the third century CE, is widely used by scholars in Chinese history, literature, and religion, and is adopted here. From the third century on, many new meanings came into play for existing graphs (or to put it more accurately, many new words came to be represented by existing graphs) and many new graphs were developed. Although we cannot be strictly precise in these matters, because such a large percentage of texts known by name (and doubtless even more that are unidentified) have been lost to us, it is helpful for readers of texts to have a sense of this general periodization and to recognize, for example, that a meaning attested for a particular graph only from medieval times is not applicable to the graph when it appears in, say, a text from the Warring States or Western Han eras. Where such distinctions can be made with reasonable reliability, they are marked parenthetically in the dictionary as "(med.)." New characters and meanings of existing characters that came into general use after the tenth century are not included in this dictionary. I hope that someone soon might take up the task of compiling a successor dictionary for the late imperial period, from the Song dynasty through the Qing, with its multitude of new coinages deriving from popular culture and vernacular speech.
The three determining factors in making sense of any text are grammar, context, and vocabulary. That order reflects the relative importance of these factors in reading classical and medieval Chinese. This dictionary will provide substantial help with vocabulary. Context of course is text-specific and cannot be prescribed. However, there is a certain amount of general cultural knowledge that hovers around particular words, which one learns from long experience but which may also be learned in advance. Many entries in the dictionary therefore include comments of a discursive nature, in addition to definitions. This is the kind of information I impart orally to students in my courses, and I hope some of it will be of use here in written form.
Grammar is the most important of the three elements mentioned above. The grammar of classical and medieval Chinese depends crucially on the so-called "empty words" (xuzi 虛字) that indicate grammatical relationships and aid in the proper construal of syntax. For these words, it does not suffice to learn a translation. One must learn for each the grammatical function it serves and be prepared to translate as needed, within the bounds of that function. These words are marked in the dictionary as "(GP)," that is, grammatical particles, and the entries for them center on explanation of their function. Unlike other entries, these usually include sample quotations which give a more extended illustration of their usage. All of the detailed GP entries, or GP segments of entries, were drafted by William G. Boltz, for whose contributions I am most grateful.
Every reader of texts needs assistance in identifying technical terms, starting with plants and animals. There are discrete dictionaries available for all manner of specialized fields, and a seasoned sinologist will eventually accumulate dozens of them. But it is useful to have much of this information accessible in a single place. Hence this dictionary includes hundreds of entries giving accurate identifications of plants and animals, as well as items in various other areas of common experience, such as astronomy, the bureaucracy, foreign names, etc., that a reader of historical and literary texts is likely to encounter. Certain important terms drawn from Buddhist texts and from religious Daoism, both fields exercising major cultural influence in medieval times, also find a place here.
Another frequent difficulty, especially in poetry, is the binome (lianmianci 連綿詞) or two-syllable word, usually alliterative or rhyming in structure. The graphs representing such words are often variable, the most important feature of the word being its phonetic shape. Sometimes a binome includes a graph that is a bound form, appearing only as an element of that binome. Often the "normal" meaning associated with a particular graph included in a binome (that is, its meaning as an independent word) has no relationship with the meaning of the binome. Binomes have been appropriately characterized as Gestalt constructions, comprising more than the sum of their parts, or as impressifs gesturing toward a certain imagistic effect but without a rigidly fixed semantic core. Like some other scholars, I typically render binomes by a pair of (usually) alliterative or assonantal synonyms, in order to suggest something of their original phonetic effect. However, this does not mean that a binome is at root a "synonym compound," nor is a Chinese phrase that is in fact a synonym compound rightly called a binome. Because binomes often present particularly challenging problems in translation, the dictionary includes several hundred of them, with suggested renderings for various contexts.
A special feature of the dictionary is the inclusion of the Middle Chinese reconstructed pronunciation, marked as "MC," of every word. Too often we read premodern Chinese texts as though just decoding a semantic puzzle. But the sound of a word is important as well as its sense,3 and having even a rough idea of the actual sound of a word in earlier times, not just its pronunciation in Modern Standard Chinese, is always worthwhile. This is especially, though not only, the case when reading literary texts that are most keenly attuned to the conscious blending of sound and sense. Of course we know that the sound of words in Classical Chinese (understood inclusively) changed through time. Here again, although finer distinctions can be made, there are two primary divisions applicable to the 1,500-year period with which we are concerned, usually called Old Chinese and Middle Chinese. "Old Chinese" normally refers to the linguistic reconstruction of the language of texts from the Warring States period through the Han dynasty, in other words the language of what we earlier defined (in its more narrow sense) as "classical Chinese." On the other hand, "Middle Chinese" is a linguistic reconstruction largely based on the language represented in the Qieyun 切韻 rhyming dictionary of 601 CE, dating from approximately the midpoint of the period we have defined as that of "medieval Chinese."
Why, one may ask, are only the Middle Chinese readings supplied in the dictionary, and not the Old Chinese readings too? The simple answer is that there is a relatively stable consensus about Middle Chinese readings (though specialists argue about certain points of detail), whereas the linguistic reconstruction of Old Chinese continues to generate broad dispute and widely divergent representational models. Moreover, Old Chinese itself is in large part a back-projection from the established phonological values of Middle Chinese and is thus largely an abstract outgrowth (or perhaps we might say pre-growth) of it. Given the relative certainty of Middle Chinese reconstruction and the relative uncertainty of Old Chinese, it has therefore seemed the safest course in a dictionary meant for general use to provide only the Middle Chinese readings for words. Of the several systems for Middle Chinese reconstructions that have been proposed over the years, that of William Baxter and Laurent Sagart enjoys the widest acceptance among scholars. For the MC readings in this dictionary I make use of the Baxter-Sagart reconstructions as given in their online posting of 20 February 2011. Any errors in the reconstruction of words included in the dictionary but not in their 2011 list are entirely my fault.
In addition to forty-plus years of my own notes about various words and their usage, I have incorporated in the dictionary definitions and interpretations gleaned from numerous other sources. It is a pleasure to acknowledge how much I have learned—and continue to learn—from scores of other sinologists, past and present, who have commented in useful and convincing ways about one or another word or term in books and articles. I have freely made use of their insights here. Thus, much of what will be found in the dictionary is a gathering of the wisdom published in various forms and outlets during the past hundred years of sinological study. Three scholars whose works I have most benefited from are Richard B. Mather, David R. Knechtges, and Edward H. Schafer. Countless other colleagues will also find their own suggestions regarding particular words or terms echoed here. In this sense the dictionary is far more of a collective production than it appears from the title page.
Six friends and colleagues, whose names appear with mine on the title page, have assisted materially in the compilation of the dictionary. Each supplied drafts of up to a hundred entries, and their commitment to the project has been integral to its success. Without their contributions and constant encouragement, my burden would have been much heavier. Each of them deserves significant credit for the dictionary's merits. I am responsible for the final form of all entries, and all mistakes should be laid at my door.
No one is more aware than I of the inadequacies and deficiencies of the present work. But a start must be made, if we are ever to have the ideal dictionary we eventually wish to see. I hope this dictionary will be a first step toward that goal and might ease to some degree the labors of those engaged in the teaching and studying of premodern texts. We are all partners together in the great enterprise of sinology. It is only through the common and accumulated research of all scholars—past, present, and future—that our understanding of classical and medieval Chinese can advance.
Paul W. Kroll
1 Especially with realization of what would ultimately be required, seeing that the "preliminary print" of a fascicle giving the entry on the single graph zi 子 ran to sixty-eight double-column pages.
2 Despite what is often said, this is not just a "Chinese version" of Morohashi.
3 Consider how naturally, in composing a written text, we decide on one among several alternative phrasings because to our inner ear it "just sounds better."