Xu Shen's Scholarly Agenda: A New Interpretation of the Postface of the Shuowen jiezi.
Within the paradigm of traditional Chinese hermeneutics, intentionality serves as the linchpin of philological methodology. Although the Shuowen Postface has already been translated several times, rereading the Postface with this culturally specific intentionality in mind highlights the organizational principles, structure, and content of the Shuowen jiezi. This article presents a new interpretation of the lexicographic method of the Shuowen jiezi (submitted to the emperor on October 19, 121 c.E.) (1) by rereading the original text and traditional commentaries through the lens of authorial intention. The central argument I make here is that the lexicographic macrostructure and microstructures of the Shuowen are designed to prove that the changes in the writing systems are historically and graphemically observable, and consequently that the original intentions of the sages who used guwen to write the classics are literally recoverable by working backwards through the reforms and changes in writing to a proper understanding of how they classified and used their words in the guwen writing system. This was Xu Shen's scholarly agenda.
The Shuowen is one of the most important documents in the history of speculation about the nature and function of language in China--indeed, it establishes the philological tradition. As Zhu Junsheng cogently remarks, the Shuowen jiezi ... is the original beginning of philology." (2) Following traditional Chinese views of language, the sages established meaning by making words correct--following from the fact that was so, Xu Shen thus made the Shuowen," it being the case that "[Xu] relied on the meaning of the sages who made words correct." (3) In discussing the microstructures of the Shuowen, Duan Yucai even goes so far as to say: "could this be the proper meaning established by making the word correct? ... Of course [Xu] makes the word utterly correct!" (4)
The Shuowen Postface appears to be a composite work, written by Xu Shen (58-147 C.E.) and his son Xu Chong (fl. 121), and includes the imperial rescript and endorsement of the simultaneous submission of the Shuowen jiezi and Kong Anguo's Explanations of the Guwen Version of the Classic of Filial Piety to the court of Emperor An (r. 107-125 C.E.). The Shuowen jiezi itself is a monument of world lexicography, and to the present day its macrostructural innovation (bushou "radical" or "classifier" divisions) is still commonly used in Chinese dictionaries in a modified form. (5)
Like the earlier lexicons Erya and Fangyan, the Shuowen was compiled for the express purpose of helping scholars to read difficult texts, especially the classics, by means of organized and classified lists of words and explanations. The Shuowen, however, is the first dictionary in the Chinese lexicographic tradition to have prefatory material written by its compiler, albeit placed at the end of the text. And this Postface is of singular importance, as it allows us to understand the content, structure, and lexicographic method of the Shuowen jiezi--which are inevitably built upon the foundation of Xu Shen's assumptions about language.
The Postface of the Shuowen gives an autobiographical account of the compilation of the dictionary (6) and a narrative account of the history of writing itself. Xu Kai notes that "in all cases writing has differences between the ancient and modern," (7) and this is what the Shuowen sets out to explain. Explaining how the writing system works, how it writes words, is of the utmost import in reading the classics; as Gu Aiji says, "if written words are not distinguished then the classics cannot be explicated." (8) Presuming that "in language there is much that is not correct," (9) and that "as for scholars of the present day, their language is also for its part not correct," (10) the task of the philologist and the lexicographer is to correct it in such a manner.
Dai Zhen--one of the most influential Qing scholars, and teacher of Duan Yucai and Wang Niansun--discussed the history of Chinese speculation about language in the following terms:
The philology of the ancients was lost, and after this there were glossing commentaries. The methods of glossing commentary were lost, and what was thereupon transmitted became fabricated and groundless gibberish ... ah, so reprehensible! As for where the classics take you, that is the Way. As for what is used to make the Way clear, it is their words. As for what is used to define words, there has never been someone who was able to do this outside of philology and written words. From written words one can comprehend language, from language one can comprehend the intent in the hearts of sages and worthies. (11)
As this passage exemplifies, the whole point of Chinese philology was to correct hermeneutic mistakes by reading specifically for authorial intention (and the rhetoric here duplicates Han dynasty writers, including Xu Shen, furiously engaging in the project of restoring language that has been lost, broken, misrepresented, and misunderstood). Gu Aiji confirms that "the texts of the classics are disorderly in terms of their written words ...; in any particular case, when one reads books it is necessary to first have knowledge of written words; if one desires to have knowledge of written words it is necessary to first investigate graphic structures ...; the philology of antiquity has now, in modern times, been cut off." (12) This is the paradigmatic complaint of traditional Chinese philologists, and perhaps even a good definition of philological practice itself, in that in the act of doing philology per se one must always begin with the assumption that something has been lost, misread, or misunderstood beforehand.
Wang Niansun takes for granted that there is a zhengwen, an Urtext--that there must have been a perfectly understandable, perfectly transparent linguistic statement that perfectly represented the original author's precise intent. As his Miscellaneous Notes on Reading Texts states, Wang is "relying on the written text to search for the proper meaning." (13) and presuming that "one necessarily desires to search for its original meaning." (14) One of the driving forces of the Miscellaneous Notes is the desire to point out when commentators and exegetes have misread an author's intent. Some representative philological comments include: "this was not Sima Qian's intention; Ban Gu was mistaken about it," (15) "this was Yanzi's intent," (16) and "he did not understand Xunzi's intent." (17)
Wang Niansun makes historical transmitters and unskilled editors responsible for basically all textual corruption, understanding that what "textual corruption" encompasses here is everything in a text that is not perfectly and easily comprehensible to the philologically trained reader. As Wang states throughout the Miscellaneous Notes, "later people intentionally changed it [based on their own opinion]"; (18) "later people did not understand the meaning of this word and so intentionally changed it"; (19) "later people did not comprehend the ancient word and so changed it"; (20) and "later people did not understand the meaning of the written text and so changed it." (21) In other words, "the intended meaning was not understood," (22) and so something was done to the text to make it more accessible to the untrained reader.
Sloppy and egregious textual emendation drove the Qing philologists to distraction, since it further corrupted already problematic texts. Part of the damage done by such unskilled editors and commentators, beyond changing individual words, actually stemmed from deleting text (tuowen). Wang Niansun notes this frequently: "later people did not understand the meaning of the text and so intentionally deleted this"; (23) "since the modem edition excises this, the intended meaning of the language is not complete"; (24) "if one excises this word, the meaning of the written text is not clear"; (25) "if one excises this word, the meaning of the written text is not clear and the syntax is also totally incongruous"; (26) "if one excises this word, then the meaning of the language is not complete." (27) In short, if one excises words or even whole syntactic units from the original text, it damages the original intent of the author, makes the language unclear and obscure, and therefore stymies the hermeneutic process. Superfluously adding text (yanwen fift Sc) is equally harmful, as the Miscellaneous Notes says in such a case, "someone did not understand the meaning of this word and so intentionally added that word." (28)
Wang's Miscellaneous Notes is a veritable "Summa Philologica" for the Chinese tradition; in substance and in fact, his work is a set of blueprints designed to aid the reader of ancient texts in sloughing off the accretions of shoddy scholarship and the vagaries of historical transmission in order to recover the original intent of the original sage-authors as it was originally stated in the original texts. And in espousing this particular hermeneutic project, Wang Niansun is largely following a paradigm that was established by the philologists and exegetes of the Han dynasty, by scholars like Xu Shen.
One suspects that very few people would deny that the study of the classics is at the core of traditional Chinese civilization, and consequently the place philology occupies in Chinese intellectual history is monumental. As the prefatory materials to the Duan Yucai's Commentary on the Shuowen jiezi state, "philology is made clear and so the study of the classics is made clear" (29) "and philology is made clear and so the classics cannot but be made clear." (30)
In terms of analyzing language, the prefatory materials to Duan's Commentary further state that "in antiquity written words were also referred to as ming 'words' ... as for these 'words', they are what those who rule consider weighty. The sages said it is of necessity to make words correct' ... Duan Yucai has been truly able to extend and expand the purpose of the sages in making words correct." (31) That Duan's preface-writers considered his commentary to be about zhengming "making words correct," should come as no surprise, as this is what Xu Shen's dictionary was supposed to be about in the first place.
Duan Yucai was reading the Shuowen for authorial intention, engaged in the project of being as true as possible to Xu Shen's original intent. As one of the prefaces to Duan's Commentary explains, "[Duan Yucai] has himself also been true to Xu's intention ... the main purpose of Xu's book was only and ever to make clear the original meaning of written words ... Confucius said 'it is necessary to make words correct,' and likely this means that when the three aspects of graphic structure, pronunciation, and meaning are made correct only then can one speak and act. Over and above this, it is necessary that the original meaning is made clear, and only then can the three aspects of graphic structure, pronunciation, and meaning be made correct." (32) Modern scholarship has had difficulties with understanding what this benyi. "original meaning," could possibly be.
The Commentary states that "every time someone has evaluated the classics and their commentaries and alongside explicated the meaning of a word therein, it must have been the case that the meaning was not the same as the original meaning of the word." (33) This comment might really give one pause, since the claim that all the changes in the meaning of a word through time must have stemmed ultimately from a single original meaning has quite interesting implications. Following from the Confucian commonplace that words are arbitrary, socially and historically agreed upon. politically enforced artifacts created and manipulated by sages (34) (a commonplace which underpins all traditional Chinese philology and explains well why dictionaries were commonly submitted to emperors), it requires no leap of faith to postulate an originary moment when each particular word took shape in the written lexicon.
It is this historical moment that Xu Shen was so keen on explicating; he compiled the Shuowen as a guide to help readers search through the past for this defining moment in the life of each word. Duan's Commentary says that "Xu uses the original meaning of words ... Every time he explains a written word, he must be using its original meaning. Every time he explains the classics, he must be relying on the written words to search for the meaning," (35) and "every time he says this word and that word have the same meaning, in all cases it is referring to the fact that the meanings of the graphic structures [which are themselves words] out of which the written words are composed are similar to one another." (36) That the graphic structures of Chinese characters are motivated is not a particularly spectacular insight, but it is the fundament of Xu Shen's entire project. The Commentary further states: "for every word in Xu's book, he relies on the graphic structure in order to explain its original meaning. His explanation and analysis must be using the written word that has the original meaning. ... Xu's explanations of the meaning come from the graphic structure. There is a graphic structure that establishes a norm for it, and thus the meaning of the word has a single definition. The explanation and analysis which come from the fact that there is an original written graph are what he uses to define the word." (37) In other words, it is "the original meaning from when the written word was first created," (38) and "in all those cases where there has been a change in a written word from antiquity to the present [the Han dynasty], Xu's explanation is of the meaning that never changed." (39)
In terms of the written form, what is this "original meaning" associated with the benzi the "original written word"? As the Commentary explains, "the guwen word is the original written word," (40) "Xu Shen follows the guwen," (41) and "Xu Shen's intended meaning comes from the guwen." (42) As for what the term guwen could possibly mean, Duan notes: "every time he says guwen, what he is referring to is the guwen writing system created by Cangjie," (43) and "therefore he says that Cangjie created written words." (44)
In the Postface and also throughout the macrostructure and microstructures of the Shuowen. Xu Shen postulates three basic writing systems: guwen the oldest and original writing system crafted by Cangjie (the source of all original written words and their original meanings); zhouwen (a.k.a. dazhuan), the modified system created by Scribe Zhou at the end of the Western Zhou; and zhuanwen (a.k.a. xiaozhuan), the even more modified (and perhaps somehow perverted) writing system codified by Li Si. Because Xu Shen is most concerned with unearthing the oldest meaning, "therefore he uses the graphic structure of the guwen to explicate the xiaozhuan." (45) Wang Yun's comment. "it is not necessary to agree with Xu Shen's intended meaning; it is also not necessary to agree with Cangjie, Scribe Zhou. or Li Si's intended meaning," (46) goes to show that even the grand pillars of traditional Shuowenxue ("the study of the Shuowen jiezi") still took guwen to literally encode Cangjie's intention, zhouwen to encode Scribe Zhou's intention, and zhuanwen to encode that of Li Si. (47)
As for the xiaozhuan writing system. Duan 's Commentary states: "what Li Si created is called zhuanshu (a.k.a. xiaozhuan)"; (48) "that being so, Li Si changed the guwen and zhouwen and called it zhuanwen"; (49) "xiaozhuan changes the guwen by abbreviating it"; (50) and "Li Si abbreviated and changed guwen ... careless people don't know this." (51) According to the Shuowen, at a rough estimate only about 13% of the original guwen written words had their graphic structure specifically altered by Scribe Zhou and Li Si. Duan notes:
As for written words that did not change, there are many of them, and therefore Xu's xiaozhuan examples are in all these cases simultaneously guwen and dazhuan. When he does not mention that the guwen is written thus and zhouwen is written so, this means that the guwen and the zhouwen are the same as the xiaozhuan ... if, in addition to the xiaozhuan, he also says the guwen is written thus and the zhouwen is written so, then these are those xiaozhuan that have been somewhat or considerably abbreviated and changed. (52)
In other words, when there is an exact historical continuity in graphic structure between the characters in the guwen, the zhouwen, and the zhuanwen writing systems there is no need to mention it, because nothing has been lost, broken, misrepresented, or misunderstood, and the sages' desired sociolinguistic construction of the world has been maintained. Lu Deming very simply asserts that in the vast majority of lexicographic entries Xu Shen "uses official script to write out the guwen [structure]." (53) As Xu Kai notes in those cases where the guwen has not changed, "the structure of the written [guwen] graph is not different from xiaozhuan." (54) If, however, a historical intervention and break in the continuity has occurred, if some reviser has intentionally stepped in and modified the graphic structure of the guwen word by changing, deleting, adding, or otherwise modifying its graphemic components (which again literally changes the universe for humans). Xu Shen steps forward in the gloss and rhetorically says, via the chong "repeat" characters, "please carefully note that some serious change has occurred here."
The similarity of Wang Niansun's approach to the task of textual criticism to the analytical approach to graphs of Xu Shen and Duan Yucai is quite striking. As Duan's Commentary states, "in all cases he explicates what is pointed to by the original creation of the written word. His explanation and analysis must use the written word that has the original meaning ...; when there are confusions perpetrated by later people, it is necessary for us to change and correct them." (55) The text-critical endeavor that Duan Yucai engages in in his Commentary is not at all dissimilar from the graph-critical analysis that Xu Shen's text engages in; the difference comes down to a theoretical distinction between the macrocosm of the history of an entire text and the microcosm of the history of an individual written word. When the Commentary says "I suspect that all these cases are from men with superficial learning changing and creating things, that it is not the original written word that was in Xu's book," (56) or "I fear that in these cases it is not Xu's own language," (57) one could easily replace the name "Xu Shen" with "Cangjie," and then these exact statements could easily have been made by the author of the Shuowen jiezi himself.
The Shuowen Postface, for the most part a complicated historical narrative, can be divided into fourteen recognizable sections. For purposes of a general introduction to and overview of the text we here reiterate the content in a synoptic chart:
i. Proto-writing. ii. Creation of the guwen writing system by Cangjie. iii. History of guwen from Cangjie to the Western Zhou (1045-771 B.C.E.). iv. Creation of the zhouwen writing system, Eastern Zhou (770-256 B.C.E.) break in the written tradition, the sages' attempt to preserve guwen by using it to write the classics. Cultural chaos. v. Qin unification (221 B.C.E.), complete loss of guwen, creation of the zhuanwen writing system based on the zhouwen model. Eight calligraphic styles of Qin writing (based on two of the historical writing systems, zhouwen and zhuanwen). vi. Han law codes prescribe literacy for officials in the eight styles of Qin writing. vii. Western Han Emperor Xuan (r. 73-49 B.C.E.) finds only four living scholars who can still read zhuanwen. Emperor Ping (r.1-5 C.E.) summons more than one hundred scholars for a great conference on writing systems. Yang Xiong attends the conference, collates everything, and compiles a lexicographic treatise summing up all current knowledge of the writing systems, which includes the rediscovered guwen. viii. Xin Emperor Wang Mang (r. 9-23 C.E.) attempts to reform writing and systematically unify the classifiers of all known writing systems, creating errors in guwen in the process. Six types of writing (based on two of the historical writing systems, guwen and zhuanwen). ix. How guwen was rediscovered, what it is, what it means, the ignorance of jinwen scholars who consider it fake, the bad methodology of jinwen scholars in general, how politically and culturally vital guwen really is. x. Structure of, content of, and motivation for writing Shuowen jiezi, how important the Shuowen is to governmental policy as the key to understanding the history of the writing systems in general and guwen in particular, the Shuowen being the basis for knowledge of the writing systems, writing systems being the basis of the classics, the classics being the basis of all government policy. xi. The encomium of the Han. xii. Xu Shen's autobiography. xiii. Xu Chong's petition. xiv. Imperial rescript and endorsement.
In the first section, the Postface describes the proto-writing of Fuxi and Shennong. The import of these two proto-writing systems is that they form the core cultural artifacts--standardized conventional representations of information, passed down from generation to generation through history, bringing order to political affairs via their use by a bureaucracy, allowing for cultural complexity--which are ultimately to serve as the foundation of the practical social function of literacy. Hence the artifacts of proto-writing underlie the conventional, historical, political, and cultural elements of writing (i.e., all of its nonlinguistic aspects). Such "proto-writing" obviously has nothing whatsoever to do with spoken language, or language at all for that matter, and everything to do with concrete social order and its preservation through time.
The second section describes how Cangjie created writing and what the consequences of this momentous historical intervention were. As Xuanying notes, "the Shuowen says: in antiquity, Cangjie created writing, first relying on the classification xiangxing 'resemble the shape', and therefore called written words 'patterns' [i.e., rebus writing]. After this, such written words were combined together according to the xingsheng 'classifier and phonetic' classification, which are therefore referred to as 'offspring' written words [in being the 'children' of two or more earlier graphs combined together to write new words--generally one grapheme used for sound only, one grapheme used to help identify which spoken word that sound represents]. The word 'offspring' in and of itself originally meant 'something that has been born'. These 'offspring' written words grew and multiplied, gradually increasing in number [and hence the term 'offspring']." (58) Based on using a combination of conventional graphic representations as phonetics and classifiers to write words, the first effect of the creation of guwen was that it allowed for the better discrimination of things--by having human discourse on such things recorded and preserved through time--and regulated the bureaucracy that had evidently itself been created by the cultural artifacts of proto-writing.
Wang Yun argues that "as for people not having knowledge of written words, the error is due to their not being able ... to divide a single graph into multiple graphemes." (59) It was in Cangjie's original combination of graphemes that guwen was formed and established as the first writing system. It should be noted that guwen, a set of written words consisting of simple and complex graphs, was created by such a bureaucrat, and that its origin is implicated in the complex relationship between the bureaucracy and the central authority.
The historical transmission of guwen from Cangjie to the Western Zhou is explained in the third section. Although there were at least seventy-two different calligraphic styles (ti), the writing system remained guwen because the structural components were not altered, that is, the classifiers and phonetics remained the same. During the Western Zhou period and perhaps earlier, elite children were taught how to write guwen by means of the liushu, "six ways of writing."
The fourth section describes the breakdown of the ideal order of antiquity and the first attempt to reform the writing system by changing its structural elements at the end of the Western Zhou period. Although Scribe Zhou evidently created zhouwen during the reign of King Xuan (r. 827-782 B.C.E.), it was only during the third year of the reign of his son King You (r. 781-771 B.C.E.) that the acknowledgement came; the Grand Scribe ... while reading the scribal records, said 'Zhou has been utterly destroyed.'" (60) Xu Shen identifies this era as one of the three extremely problematic periods in Chinese history during which the writing system was irrevocably altered (the other two being the brief Qin and Xin periods).
In Xu Shen's account, the writings of Confucius and Zuo Qiuming reach back to pre-Spring and Autumn period culture by using the guwen writing system, and the Chunqiu and the Zuozhuan themselves are understood to be stern ethical judgments made by these sages on the decadent state of the political culture of the Eastern Zhou. The use of guwen by the sages is further seen as the reflection of a very critical and very pointed attempt on their part to return to the traditional ideal state of the Western Zhou and high antiquity. The creation of zhouwen, therefore, was one of the harbingers of the fall of the Zhou house, and hence zhouwen should be regarded as somewhat flawed, as having something of the fragrance of a doomed state about it. The efforts of Confucius and Zuo Qiuming were apparently for naught, as immediately after their era the world completely fell apart--all of the high culture of antiquity (rites, music, canons, records) was discarded, and cultural artifacts, including measurements, transportation, law, dress, language, writing, became heterogeneous, scattered, and completely lacking in order. Chaos was the watchword of the day, regicide ran rampant, and kings used the lacquered skulls of their enemies as drinking cups.
Xu continues in section five to describe what occurred after the First Emperor of Qin (r. 246-210 B.C.E.) reunified the world, when his officials homogenized all of these cultural artifacts, including standardizing the disparate writing systems by reforming them into a newly fixed system called xiaozhuan (a.k.a. zhuanwen). The consequence of this reform in the history of the writing systems is that guwen was completely lost. It should be noted here that the Qin dynasty (221-207 B.C.E.) is also purported to have destroyed all the remains of the refined ethical culture of high antiquity (rites, music, canons, etc.), and that its success in stamping out the use of guwen is tied up in this purgative activity. According to Xu, all the books that had been written in guwen were burned by the Qin and hence, except by sheer good fortune and a little planning on the part of certain individuals, might have been lost forever (see below, section nine).
Section six explains that according to Han law, court officials should be trained in the eight calligraphic styles of Qin writing. Xu Shen then complains about the laissez-faire attitude toward this policy and its historical consequences, which in section seven he states to be that very few people during the Han knew much of anything about the old writing systems and even less about the genuine historical differences between them. The role of Yang Xiong as compiler of the conference volume Xunzuan pian is then shown to be the most important philological work done during the Han. Considering that there were several Erudites (boshi) in guwen versions of the classics at that time, the explanations the Xunzuan contained must have come at least partially from guwen scholars.
Section eight decries the damage done to guwen by Wang Mang, which is further specifically identified in the main body of the text of the Shuowen with such statements as "the destroyed Xin dynasty took it to be the case that the words that had been classified under the triple ri of the jing 'bright' classifier had been used in the sense of 'great abundance', so they changed the classifier of these words to three tian FEI instead (which is not a historical classifier according to Xu Shen)." (61) One can see that for a scholar passionately concerned with the historical changes in the writing systems, and especially with how to properly interpret the original guwen of Cangjie and the sages, what Wang Mang was doing (and what Scribe Thou and the Qin reformers had done before) would be extremely disconcerting, in fact an outrage. From Xu's perspective, they were literally restructuring the world in a way the sages had not approved of. Feng Yan notes that ... Wang Mang ... edited and 'fixed' the writing systems, and thereby somewhat changed guwen." (62) If one wants to understand what classifications the sages had applied to their words, and hence how they had used them (and furthermore how these words should be used for the proper maintenance of the constructed order of the world), the last thing anyone would like to see is a committee of vulgar reformers coming along and "fixing" it. Yet according to the Postface, this is exactly what Wang Mang's officials did. The eighth section further lists out the six types of writing in use at the time of Wang Mang, based on guwen and zhuanwen.
Section nine is by far the longest and most important section of the Shuowen Postface. Xu Shen first defines what he means by guwen: (63)
1) The writing system invented by Cangjie in high antiquity.
2) The writing system used in the texts hidden away during the Qin bibliocaust and later found in the walls of the Kong family dwelling when it was demolished sometime during the reign of either Emperor Jing (r. 157-141 B.C.E.) or Emperor Wu (r. 140-87 B.C.E.), which included Liji, Shangshu, Chunqiu, Lunyu. and Xiaojing. (64) As Xu Kai notes, the meanings referred to by Xu Shen and by Kong Anguo are the same." (65)
3) The writing system used in the text of the Zuozhuan submitted to the Han court by Zhang Cang (256-152 B.C.E.) 66 and the Mao version of the Shijing (provenance unexplained by Xu Shen).
4) The writing system used on ancient bronze vessels with structurally consistent inscriptions (i.e., those whose classifiers and phonetics match up with the graphs from the above texts).
The Postface argues on the basis of these four items that guwen is authentic, internally consistent, and structurally explainable as a writing system, even if the tradition was entirely broken during the depredations of the Eastern Zhou and Qin and further confused by the ill-advised reforms of the Xin era. While acknowledging that many Eastern Han (25-220 C.E.) scholars consider guwen to be a spurious writing system created by ambitious scholars outside the mainstream seeking favor and jobs at court and by the followers of Wang Mang, by historicizing the writing systems Xu Shen explains that the jinwen scholars who make these claims have themselves actually been using Qin lishu (a calligraphic style based on the zhuanwen writing system) to graphemically analyze and interpret the words of the classics and even to adjudicate legal cases.
Because they began with the inherently flawed assumption that writing has never changed, Xu argues that jinwen scholars were therefore blindly working with what they wrongly perceived to be the genuine intentions of the sages as encoded graphemically in the writing system--that is, interpretations of words based on the structure of the characters that write them. It is almost a slap in the face of the jinwen scholars that Xu Shen presents them with the historicist argument that they have been basing at least a select portion of their exegetical and governmental policy work on the drastically harmful if not immoral alterations specifically made by Qin officials to the writing system. Hence a portion of the intentions the jinwen scholars were finding in their graphemic analysis of the classics (shuozi jiejing "explaining characters in order to explicate the classics") were the intentions not of sages, but of vile Qin criminals, one of whom was a eunuch regicide, no less. (67)
Following Xu Shen's partisan line of argument, it makes sense that the study of guwen should be made a political priority at court. It was an absolute necessity to honor and cultivate guwen; in fact, one of the key messages contained within the classics themselves is that guwen can provide a link between high antiquity and the present, and hence has the ability to make the political present more like the ideal past. As au Junsheng remarks. in reading books you must first have knowledge of written words. Once you have knowledge of written words you are able to understand the classics. Once you understand the classics you are able to cause them to be properly used." (68)
In section ten Xu Shen takes up the task of explaining the methodology he has used in compiling the Shuowen jiezi. Because zhuanwen, rather than guwen, is the underlying structure of the writing system of the Eastern Han (for the problematic historical reasons discussed above), the Shuowen uses this writing system as the apparent written form for its macrostructure. Duan Yucai explains that as for Xu Shen giving great weight to reviving the past yet not putting guwen and zhouwen first in the Shuowen layout, he wanted people to investigate further antiquity through using recent antiquity. As much of xiaozhuan relies on guwen and zhouwen and doesn't change, for that reason putting xiaozhuan first is correctly what is used to explain guwen and zhouwen." (69)
Xu Shen distinguishes what the classifiers are that he writes in zhuanwen according to their ancestry in zhouwen and guwen. In other words, using (recent antiquity) zhuanwen as the formal writing system for the macrostructure was a methodological ploy ultimately designed to ease the Eastern Han reader back to the (further antiquity) guwen writing system. Since zhuanwen was the underlying structure of the familiar official script of everyday contemporary literacy, it became the stepping-stone for getting back to guwen.
Xu Shen points out when a guwen grapheme has come down unchanged into zhuanwen but is misunderstood because the word that the classifier wrote in guwen has taken on a different graphemic structure in zhuanwen (and hence the guwen grapheme is no longer understood properly as a classifier, because the original classifier-as-character-writing-the-word has been lost). (70) Duan Yucai again clarifies: as for each and every example in the book that has the guwen or zhouwen first and the xiaozhuan after, in all cases the reason is due to the classifier." (71)
Wang Guowei further explains that Therefore, what was said in the Postface--"in the present era I have put zhuanwen in order together by means of guwen and zhuanwen"--should be talking about the lemmata Characters and not talking about the "repeat" characters. As for the guwen and zhuanwen among the "repeat" characters, it is then the case that there is a difference between the guwen and zhouwen on the one hand and the zhuanwen on the other, and this difference extends to there being a difference between the guwen and the zhouwen graphs themselves. If there are guwen and zhouwen among the lemmata characters, then it is the case that guwen, zhouwen, and zhuanwen are all this same graph. ... that being so, the Sluumen jiezi really brings together guwen, zhouwen, and zhuanwen to make a unitary text ... as for earlier people, some took it to be that lemmata characters in the Shuowen were in all cases zhuanwen and that guwen and zhouwen were only seen among the "repeat" characters. but I am afraid that this is really not the case. (72)
In discussing such unusual lemmata, Gui Fu for his part writes that as for this being a guwen word as headword, it is an example of where the xiaozhuan did not change the guwen structure ... but the zhouwen greatly changed the guwen ... and so it is not the original graphic structure." (73) This explains why the guwenlzhuanwen graph is structurally the same but the zhouwen graph, which differs, is found among the "repeat" characters.
According to the Postface, both the microstructural gloss of the classifier head-entry lemma (that is, his explanation of the bushou word) and all the glosses of the headwords listed beneath it have been explained based on what Xu Shen has heard from authoritative men who had evidence. This evidence, it seems, was their analysis of the graphemic components used to write the word in question. In every case where Xu Shen has detected structural change between guwen, zhouwen, and zhuanwen (about 13% of the total number of words), he relies on the structural components of the guwen character to give his main etymological definition gloss (i.e., how the sages were presumably using that word based on how they classified and composed it in the guwen writing system). He also plainly states that in every case he has double-checked the structure of the zhuanwen character against that of the guwen character to make sure that they correspond, and this by means of the contents of the guwen texts of the Yi, Shu, Shi, Liji, Chunqiu. Zuozhuan, Lunyu, and Xiaojing. According to his analysis, for about 87% of the words there has been no change and therefore no difference between the original guwen structure and the zhuanwen structure, and therefore he does not in those cases need to make any note of it. Wang Guowei reiterates that the Shuowen fundamentally arranges zhuanwen in order by bringing it together by means of guwen and zhouwen. In those cases where the guwen and zhouwen are different from zhuanwen it appears, but if they were the same then it does not repeatedly appear." (74)
Without exception, every single explicitly marked example of guwen and zhouwen in the Shuowen is structurally (not simply calligraphically) different from the zhuanwen--changed classifier, changed phonetic, or both, marking diachronic structural change in the writing systems--and that is why Xu Shen has listed them specifically in those cases. The guwen and zhouwetz graphs in the Shuowen are not simply "variant graphs" (yitizi), with which we are so familiar from the later lexicographic tradition, but rather are the undeniable evidence of Xu Shen's historicist method of differentiating the three writing systems. As such they represent, in every microstructure in which they appear, the overarching lexicographic principle used throughout the Shuowen jiezi, the very same principle that the author repeatedly and explicitly states in the Postface.
At the end of section ten, Xu Shen lists out the 540 classifiers of the guwen writing system in their zhuanwen forms, which exactly match zhuanwen classifiers most of the time. These classifiers, he argues, are the inherent structures of guwen. and were therefore used by Cangjie and the later sages to classify words (establishing both signifier and signified), thereby passing down their intentions through time, through all the vagaries of diachronic change in the usage of particular words. If he has understood aright the regularities of the rules of word composition and classification, the Shuowen thus becomes the perfect tool to crack the code of the historically intended message of the classics through a proper understanding of the written forms of their words. It is noteworthy that Xu Shen mentions nothing whatsoever about the liushu "six ways of writing" as part of his design of the Shuowen and in fact only mentions it in the context of juvenile pedagogy. Finally, he reiterates that his macrostructure and microstructures will help the reader understand the ultimate origins and often problematic changes in the writing systems and how these have mirrored equally problematic changes in human culture through time (from his perspective, the end of the Western Zhou, the Qin, and the Xin).
Section eleven is an encomium in praise of the Han dynasty, its awesome glory and how it supports traditional scholarship, especially, and hopefully, guwen scholarship. The twelfth section is Xu Shen's autobiography, where he narrates the story of his elite ancestry, his own fascination with the classics, and his inability to keep his nose out of canonical books and their explanations. He also notes that his scholarship is extremely careful and that he has genuinely assisted the intentions of the sages by serially arranging and clarifying their words. Finally, he asks any reader who has a better understanding of the structure of the writing systems please to correct any mistakes that he has made. Taking a look at the 182 commentaries and marginalia appended to the gulin edition of the Shuowen, one would not hesitate to say that his readers have taken this request very seriously over the past two millennia. (75)
Xu Chong's petition, which constitutes the thirteenth section of the Postface, reveals his father's scholarly affiliation to the guwen erudite Jia Kui (30-101 C.E.), who had been commanded by Emperor He (r. 89-105 C.E.) to revise and put in order the different explanations of the writing systems--not to reform writing itself, which was very dangerous, but rather to explicate its historical problems and structural differences in a comprehensible way. Xu Chong states that the Shuowen fulfills this task by giving explanations of the old glosses, and that it records a proper interpretation of how the sages intended words to be used and literally to construct everything in the entire cosmos, all in one convenient and orderly package, the ultimate in reference manuals. We also learn that Xu Shen has been compiling the Shuowen by imperial command (presumably to complete the project Jia Kui began), and that Xu Shen's parallel work on Kong Anguo's Explanations of the Guwen Version of the Classic of Filial Piety was submitted, we assume unasked for, side by side with the Shuowen jiezi. Last but not least, the fourteenth section consists of the imperial rescript and endorsement. Here we see that Xu Chong was allowed to present these two books to the Emperor in person in an informal setting at the gate of one of the palace apartments, and that he received forty bolts of cloth as remuneration for his troubles.
While the Postface has an enormous amount of material to offer us on the historical context and the macrostructure of the Shuowen--which can be quickly summed up by stating that Xu Shen's argument is essentially that the 540 bushou classifiers are the structural classificatory system of guwen (albeit written here in their zhuanwen-descended state) as it was created by Cangjie and purposely used by Confucius and Zuo Qiuming--the lexicographic microstructures of the Shuowen are also very important for a proper understanding of the Postface. The guwen classifier, including all of its glossalia, forms the bushou head-entry. All of the so-called lemmata then listed after the head-entry form the second half of a larger microstructure, which makes them microstructural units belonging to and derived from the head-entry: the classification of every X-class word in all cases comes from the X-classifier." Though it may not seem immediately obvious because the work is so large, the Shuowen therefore has only 540 recognizable lexicographic entries, only 540 true lemmata. What Xu Shen had discovered in the graphemic structure of guwen (and presumed to have been assigned to the words by Cangjie) is something nearly equivalent to the classificatory macrostructure of the nineteen word classes and thirty-nine subclasses of the Erya, which were believed by Han commentators to have been assigned to the words by the sages who wrote the text. Let us examine for a moment what the macrostructure of the Shuowen is not, so as to realize the full import of what it actually is.
In his Shuowen tongxun dingsheng (1833), the Qing philologist Zhu Jun-sheng reorganized the macrostructure of the Shuowen into eighteen Old Chinese rhyme groups according to the 1137 sheng "phonetics" that he identified, further broken down into categories by the liushu. Beyond the eighteen rhyme groups, these criteria are fully evidenced in the microstructures of the original Shuowen itself. So why, we might ask, did Xu Shen not organize the Shuowen by the phonetics that he identifies almost ninety percent of the time? Or, similarly, why did he not make the liushu part of the macrostructure, dividing up the mass of almost 10,000 words into six large but easily recognizable categories? If the liushu are to be understood as the main analytical principle Xu Shen supposedly used, why are they not foregrounded in this manner? There was certainly enough information in the text of the Shuowen on these things for Zhu Junsheng to accomplish his editorial reorga-nization--so why did Xu Shen not do it that way originally? In other words, why did he do what he did the way he did it?
It seems that Xu Shen was much less concerned with which word uses which phonetic, or which word uses which of the liushu, than with which word historically used which classifier. From the way Xu Shen discusses words in the Shuowen, both in the Postface and in the text itself, it is evident that he makes no distinction between the sound, written shape, and usage of a word--they are analytically inseparable. (76) When he lists out the structurally different "repeats" (chong), the claim may well be that these are what we would call different words--words that have undergone changes not only in the graphic structure of the written character, but also in either sound or usage or both. As he is basing his understanding of the use of a word (i.e., the historical etymology, or so-called benyi "original meaning") in large measure on the classifier, then if the classifier has changed, Xu Shen must consequently understand the usage of the word to be different. If the phonetic has changed, he likewise takes it to mean that the sound, if not also the usage, of the word has been modified in some way, which changes the world, at least insofar as humans relate to it through language.
In the microstructures outside of the bushou head-entries, so long as the graphemic structures of the guwen graph remained intact into zhuanwen, these analyses remain the same. However, when a word has changed historically in any way, that is, in every case of a "repeat," new intentions have been introduced, historically identifiable as distinct from those of Cangjie and the sages. In other words, in those cases it is rather the intentions of Scribe Zhou, the highly questionable folk of the Warring States period, the Qin or the Xin criminals damaging or distorting the words, what one might call "lexical corruption." a phenomena quite similar to the more familiar "textual corruption." For example, when zhuanwen breaks graphemically with the guwen word, it may well have been understood that Qin meddlers were purposefully trying to change the usage (and perhaps also the sound) of the original word coming down from the guwen, consequently modifying the nature of the world encoded in language by signifier and signified. According to Xu Shen this is exactly where the graphemic analyses of the vulgar jinwen scholars were mistaken, and why they were recovering tainted and mutant intentions from the texts of the classics written in Qin official script. (77)
Besides noting that Xu Shen also points out in his microstructures those places where a phonetic has been abbreviated (shengsheng) and that he gives duruo "read it out loud like" glosses, which in some instances are more than just homophones (i.e., paronomasia wherein a meaningful link in usage is implied), this discussion presents the sum total of the macrostructural and microstructural elements of the Shuowen. It is only by viewing this text through the lens of authorial intention, rather than by anachronistic application of contemporary knowledge of the history of writing in China, that the greater scope of the lexicographic principles of the Shuowen fiezi can be seen. This method of reading makes Xu Shen's scholarly agenda clear.
IV. TRANSLATION: "THE SHUOWEN POSTFACE" (78)
In antiquity, when Paoxi ruled the world as king, looking up he observed celestial phenomena in the heavens; looking down he observed regularities on the earth. He saw suitable matches between the earth and the marking-patterns on birds and wild animals. From nearby he took such marking-patterns from various parts of his own body; from afar he took them from various parts of things, and thereby created the eight trigrams (79) in order to hand down normative representations. (80) Coming to Shennong knotting cords [as quipu] (81) for the purpose of bringing order and thereby controlling his affairs, (82) the numerous professions greatly (83) flourished, ornamentation and artifice came forth like sprouts.
ii)The Scribe of the Yellow Emperor, Cangjie, observing the traces of the footprints and tracks of birds and wild animals, understood that their linear structures could be distinguished from one another by the differences between them. When he first created writing by carving in wood, the hundred officials became regulated, and the myriad things became discriminated. Cangjie probably took this from the hexagram guai (84) One of the meanings of the word guai is "to present something at the royal court." (85) This is saying that wen 3Z "cultured pattern" [writing] is a means to disseminate education and illuminate ethical influence for the king at court; it is also a means by which the gentleman bestows emoluments upon subordinates; and it is a means for storing up virtue and making clear what is prohibited. (86) When Cangjie first created writing, he probably relied on the variety of things and therefore made xiangxing "represent the shape" [i.e., rebus] graphs. Therefore he referred to such written words as wen. "written units of a cultured pattern." (87) After this, xingsheng "classifier and phonetic" were added together; accordingly he referred to them as zi "off-spring characters." (88) As for the word "offspring characters," this speaks of gradual increase through reproduction and multiplication. One refers to such writing [*trjagh] on bamboo and silk as "writing" [*hrjag]. One of the meanings of the word "writing" [*hrjag] is "to resemble" [*njag].
iii) During the ages of the Five Emperors and the Three Kings, there were changes in the distinct calligraphic styles. (89) As for the Feng sacrifices. on Mt. Tai, during the seventy-two occasions in which they occurred they were never the same in regards to calligraphic style. (90)
According to the rites of Zhou, in the eighth year one enters elementary education. (91) The Palace Protector teaches the children of the state first by means of the liushu AM "six ways of writing." (92) The first is called zhishi Ivy "point to the thing." As for zhishi, observe and one can recognize it, investigate and one sees it; shang Ii and xia are examples of this. The second is xiangxing "represent the shape." As for xiangxing, achieve the thing by drawing according to the bends and curves of its shape: ri and yue are examples of this. The third is xingsheng"classifier and phonetic." As for xingsheng, regard the thing as a ming "word" and grasp the classificatory analogy so as mutually to achieve it; jiang and he are examples of this. The fourth is huiyi "combined intentions." As for huiyi, pair up their classifications and combine their meanings [matchings] in order to see what points and waves; wu and xin are examples of this. (93) The fifth is zhuanzhu "mutual cross-gloss." As for zhuanzhu, establish a classification under one [bushou] heading, with the same intended meaning and mutually exchangeable; kao and lao are examples of this. The sixth is jiajie "loan borrowing." As for jiajie, since the word originally lacks its own written graph, rely solely on the phonetic to stand in for the thing; ling 4- and chang are examples of this.
iv) Coming down to the time of King Xuan, Scribe Zhou wrote the Dazhuan in fifteen pian, which is somewhat different from guwen. (94) Arriving at the time of Confucius writing down the six classics, and Zuo Qiuming narrating the tradition of the Spring and Autumn Annals, in all cases, even though they used guwen, their intended meaning can for the most part be obtained and explained. (95) After them, the various marklords warred with one another; (96) they were not unified under one king. They considered the "harmful" effects of the rites and music to be bad and in all cases discarded their canons and records. The realm was divided into seven states. Agricultural field divisions had different measures, carts and road-ruts different axle-widths, (97) laws and commands different rules, ritual garb different sumptuary regulations, spoken words and language different sounds, and writing different graphic structures. (98)
When the First Emperor of Qin first unified the world, Counselor-in-Chief Li Si then petitioned to homogenize these things, so as to remove what did not conform to the Qin pattern of culture. Li Si wrote the Cangjie pian. Zhao Gao, the Eunuch Palace Director of the Livery Office, wrote the Yuan ii pian. Huwu Jing. the Grand Scribe, wrote the Boxue pian. In all cases they took the writing system from the dazhuan of Scribe Zhou, with some partial simplifications and reforms--this is what is referred to as xiaozhuan. (99) And just at this time [when xiaozhuan was complete] Qin burned and utterly extinguished the classics and other [guwen] writings, washing and sweeping away the old canons. And then, in massive fashion, they sent out official troops, raised border corvee laborers, and the duties of officials and overseers--and the frequency of criminal cases--were all thus greatly multiplied. At this time lishu, "official script," first appeared, which was simplified for the purposes of speed in writing and for standardization. For these reasons guwen was completely cut off. (100) Qin writing had eight styles: the first was called large seal (dazhuan): the second, small seal (xiaozhuan): the third, tally engraving; the fourth, creature script; the fifth, copy stamp; the sixth, public official script; the seventh, spear script; the eighth, official script (lishu).
When the Han arose, draft script appeared. According to the codes of the Commandant of Punishment, when students reach their seventeenth year or more, they begin the exams, memorizing and reciting the writings to the tune of some 9000 characters, and only then do they obtain the position of Bureau Clerk. The eight styles of writing are also used to test them. The provinces send them to the Grand Scribe, who examines them on both counts [i.e., the 9000 characters and the eight styles]. The best are made Imperial Secretary Scribes. If in writing something is not correct, these Scribes abruptly report and censure it. Now, although the codes of the Commandant of Punishment exist, there are no examinations. Elementary education (xiaoxue) is not cultivated. No one has gone through the explanations for a very long time now.
vii) During the time of Filial Emperor Xuan, the Emperor summoned all those who could read the Cangjie plan. Zhang Chang had received it from someone. Du Ye, the Regional Inspector of Liangzhou, Yuan Li from Pei, and Qin Jin, Grandee for the Exposition of Education, for their part were able to recite it. During the time of Filial Emperor Ping, the Emperor summoned Yuan Li and others, more than one hundred men. He ordered them to explain writing systems and written words in the Weiyang Palace. He made Yuan Li the Paramount Serviceman of Elementary Education. The Gentleman Attendant at the Palace Gate, Yang Xiong, gathered their explanations in order to make the Xunzuan plan. In all, from the Cangjie plan to the most recent, there are fourteen pian, totaling 5,340 characters. As for that which has been recorded in the various writings, these lexicographic works have roughly preserved it.
viii) Coming down to the overthrown Xin dynasty's temporary occupation of the throne, Wang Mang ordered the Censor-in-Chief, Zhen Feng, and others, to comparatively edit the word classifications [bushou] of the writing systems used to write. Wang Mang himself regarded these as that which ought to be systematized. Zhen Feng therefore "somewhat" revised and "fixed" guwen. (101) At that time there were six types of writing. The first was called guwen, the writings found in the walls of Confucius' house. The second was called qizi, "strange offspring," that is, guwen with some things that are different. (102) The third was called zhuanshu, "seal script," that is, xiaozhuan, which was made by Cheng Miao, a person from Xiadu, who was ordered to do so by the first emperor of Qin. The fourth was called zuoshu, "civil official writing," that is, Qin lishu. The fifth was mouzhuan, "coiled seal," which was used on copy seals. The sixth was niaochongshu, "bird and creature script," which was used to write on banners and tallies.
As for the writings from the wall, King Gong of Lu (f1. mid-2nd century B.C.E.) demolished the Kong [Confucius] family dwelling and found the Record of Rites, the Classic of Documents, the Spring and Autumn Annals, the Analects, and the Classic of Filial Piety. Furthermore, the Marklord of Beiping, Zhang Cang, offered up the tradition of Zuo Qiuming on the Spring and Autumn Annals. Prefectures and states for their part often obtain cauldrons and ritual vessels from mountains and rivers [i.e., from hidden caches of ancient bronzes washed up or unearthed]. Their inscriptions are indeed the guwen of former ages, in all cases themselves structurally similar to one another. Although one cannot again see their origin and development, their particulars can always be obtained and explained in summary form. Yet among the people of this age, the great consensus is to reject and slander guwen. They consider it to come from those who like oddities, those who purposefully fake and alter the correct writing system, who from the wall falsely create writings that cannot be understood, and who change and disorder the constant and prevailing writing system so as to make themselves shine brilliantly in this age. Various scholars compete over the suitable matchings between explaining characters and explicating the classics (shuozi jiejing). (103) They consider Qin lishu to be the writing of the time of Cangjie, saying that from father to son it was transmitted one to the other--how could it receive revisions and changes? Then they rashly say the head of ma and ren together makes chang, that ren holding (104) shi makes dou, and that as for hui, it is a bent zhong. The Chamberlain of Law Enforcement, in explaining the legal codes, has reached the point of using the structures of graphs to give judgments about the laws. In reproving (ke) a person who received bribes, he wrongly considers that this offspring character is from zhi "stop" and gou "taking." Things like this are extremely numerous, in all cases not in accord with the guwen of Kong Anguo, and even erroneous according to the dazhuan of Scribe Zhou. Vulgar Ruists and despicable rustics, content with what they are familiar with, blind to that which they have seldom heard, have never seen comprehensive learning--they have not yet even once observed the regularities of the rules of offspring characters. They consider the old skills strange and consider wild talk good. They regard their own knowledge as secret subtleties--examining and comprehending the subtle intentions of the sages. Moreover, they see in the Cangjie pian the sentence "my young son has received the imperial mandate," and due to this say Cangjie pian was written by an ancient emperor--that its passages possess the art of the supernatural and the immortal due to this. Their delusions and mistakes are incomprehensible. Are they not foully perverse? The Classic of Documents says: "I desire to observe the representations of the men of antiquity." (105) This is saying that it is necessary to honor and cultivate the old writing systems and not bore through and carve them. (106) Confucius said: "I still reach back to the scribe's not writing something down. Now this is lost forever--alas!" (107) This criticizes them for not knowing and failing to make inquiries. People use their own selfish opinions; right and wrong have no correct standard; facile explanations and deviant expressions have made the scholars of the present world worthy of suspicion. Now, as for writing systems and their offspring characters, these are the root of the classics, the origin of kingly government, what former men used to hand down to posterity, and what later men use to remember antiquity. Therefore it is said: "when the root is established, the way is born," (108) and "knowing the world's extreme subtleties, there cannot be disorder." (109)
x) Recently, I have arranged in order the xiaozhuan writing system, bringing it together by means of the guwen and zhouwen writing systems. I have extensively selected [glosses] from authoritative men, both greater and lesser, and they are believable and verifiable. I have investigated and compiled their explanations for the purpose of putting in order the various classifications--so as to explicate errors and mistakes, enlighten scholars, and convey the fabulous meanings of words. I have split up and divided the explanations into word classification (bushou) locations, so they would not be jumbled up together. The myriad things all together are observed, and there are no words that are not properly recorded together under such headings. When their meaning [matching] is not clear, I make things clear by informing the reader [via the microstructures]. I have consulted the following works: the Changes of Meng, the Documents of Kong, the Poetry of Mao, the rituals of the Zhouguan [the Zhouli], the Spring and Autumn Annals of Zuo, the Analects, and the Classic of Filial Piety--in all cases guwen. Where there is something that is not known, I leave the matter open ... [bushou list excluded here]. (110) ... This dictionary has 14 sections, 540 word classifications, 9,353 graphs as units of a specific writing system, 1,163 "repeat" words with different structures that show how the writing systems have changed over time, (111) and analysis and explanation, altogether 133,444 written words.
As for their established headings, I set up the word yi "one" as the beginning. Words that are comparable are clumped together by means of classifications; Things are separated by means of such groupings. 112 What has the same (graphic) rules is placed under one classification; (113) All the graphic structures are strung together with one another. Mixed yet not transgressive, (114) Relying on graphic structures systematically joined together. Drawing out and extending them. I use them to investigate the myriad origins of words, And conclude, at the end, with the terminus hai, Knowing the transformations, sifting through the mysteries [of etymology]. (115)
xi) Now the Great Han dynasty, its sagely virtue shining and illuminating, has inherited [the mandate of] heaven and investigated Tang, (116) broadcast its virtue and regulated the center. The distant and near are covered with benefits, as it overflows with an abundance of greatness. (117) The Han broadens its endeavors and investigates the subtle--its scholar-officials know the rites. (118) Searching into the primal, inquiring into the hidden--their meaning [matching] can be transmitted.
xii) It is January 29, 100 C.E. Great-grandson of a great-grandson, I am but an insignificant child. My ancestry is from the Flaming Emperor; from Jin Yun, minister to the Yellow Emperor; from Gong Gong, who succeeded Gao Xin; from Taiyue, who assisted the Xia dynasty; from Lu Shu, who acted as border defender [for King Wu of Zhou]. He was made marklord of Xu, for generations enfeoffed, passing down this charisma. (119)From Xu my ancestors advanced to Shaoling, (120) settling in this place on the bank of the river Ru. I have presumed to look up at the great road, (121) and I have dared to direct my feet toward the gate of the sages [i.e., the classics]. What does their magnitude resemble?--lofty as the southern mountain. (122) I desired to quit pursuing this but was unable to, hence I have completely exhausted my inadequate abilities. I have taken great care with the flavor of the Way--hearing a doubt. I recorded a doubt. (123) I have expanded on and assisted the sages' intentions, putting their subtle expressions into a serial arrangement. Those who understand this are rare. Perhaps mistakes in my work will be revealed; in fact, I hope that there is someone who understands this, who puts them in order and corrects them.
xiii) Grandee of the Eighth Order, currently unemployed administrator, (124) and your retainer, I, Xu Chong from Wansui village in Shaoling, bowing my head and repeatedly doing obeisance, submit this petition to Your Majesty the August Emperor [Han Andi]. This retainer humbly observes your supernatural brilliance and abundant virtue that inherits and follows the achievements of the sages. Above, you have examined the measures in the heavens; below, you have caused transformations to flow out among the people. You anticipate heaven and heaven does not contradict you; you follow heaven and serve heaven's seasons, (125) and the myriad states are all at peace, (126) the supernatural and the people are thereby in harmony. Similarly, you also have deeply pondered the subtleties of the five classics, which in all cases constitute the Han legal code. You have extensively selected from the hidden and the distant, exhausted the structures of things, and completed inborn natures in order to reach [an understanding of] fate. (127) The former Emperor [Han Hedi] commanded the Palace Attendant and Commandant of the Cavalry. Jia Kui, to revise and put in order the distinct divergences of the different theories of the old writing systems. (128) The educational influence of kings has a single starting point--if there is something of use to the state, it is without fail gathered up in its entirety. (129) The Classic of Changes says: "thoroughly investigating the marvelous and knowing the transformations is the acme of virtue." (130) The Classic of Documents says: "as for officials who have ability and have done something worthwhile, allow them to present what they have done--then your state will be prosperous." (131) This retainer's father, Xu Shen, formerly Director of the Archival Section subordinate to the Defender-in-Chief, (132) originally received training in guwen studies from Jia Kui. Certainly sages do not rashly create new things, but in all cases have reliable evidence. (133) In the present era, the Way of the five classics is luminously illuminating and gloriously bright, and the writing systems and their offspring characters are the root from which the classics were born. From the rituals of Zhou to the Han legal codes, in all cases one must study the six types of writing to completely and thoroughly comprehend their intended meaning. Fearing that facile explanations and harmfully incorrect phrases were causing scholars to become worthy of suspicion, Xu Shen extensively questioned knowledgeable people, checked their explanations against what Jia Kui had said, and thereby wrote the Shuowen jiezi. As for the old glosses on the six classics and other various writings, this dictionary gives explanatory glosses of their intended meaning, and, as for words for heaven and earth and ghosts and the supernatural, mountains and rivers and vegetation and trees, birds and wild animals and insects and creatures, miscellaneous things and the strange and the odd, the institutions of kings and the ceremonies of ritual, the divisions of the ages of the world and the affairs of people, there are none that are not completely recorded therein. (134) In total, 15 pian and 133,441 written words.
Xu Shen formerly, by imperially proclaimed written command, was collating the Shuowen jiezi in the Eastern Pavilion and teaching the Eunuch Palace Attendants Meng Sheng Li Xi, and others. As the text [or "the writing systems and their offspring characters1 had not yet been fixed in final form, the Shuowen jiezi has heretofore not been submitted to you. Now Xu Shen is ill and has sent this retainer to present the Shuowen jiezi to you at the palace. Xu Shen also studied Kong Anguo's Explanations of the Guwen Classic of Filial Piety. As for the Guwen Classic of Filial Piety, during the time of Filial Emperor Zhao (r. 86-74 B.c.E.) it was presented by the Elders of the state of Lu. During the Jianwu reign period of Emperor Guangwu (25-56 C.E.) it was collated by the Palace Steward and Court Gentleman for Consultation, Wei Hong (fl. 25-57 C.E.). (135) and in all cases the explanations were orally transmitted. Since officials do not have these explanations, my father has attentively compiled one pian and submits it side by side with the Shuowen jiezi. This retainer, Xu Chong, is truly frightened and sincerely terrified, I kowtow and kowtow, I deserve death for this crime, death For this crime, bowing my head and repeatedly making obeisance in order to report to Your Majesty the August Emperor. Submitted on October 19,121 C.E.
xiv) We decree that the person who submitted this petition, Xu Chong from Runan, proceed to a meeting at the gate on the left side of the palace apartments, where he will be commanded to present the writings, about which he petitioned us, both together. [Imperial endorsement of receipt] November 16, 121 C.E., the Eunuch Palace Attendant, Rao Xi, according to the written proclamation of the Emperor, grants forty bolts of cloth to the Grandee of the Eighth Order from Shaoling, Xu Chong. Having on this day received the proclamation at the Vermilion Bird Gate [the southern gate of the Northern Palace], (136) you are ordered not to render thanks for this endorsement.
(1.) An erroneous calculation of this date according to the Western calendar--September 19, 121 C.E.--was unfortunately published by Paul Pe'fiat (1878-1945) in his review "Les bronzes de la Collection Eumorfopoulos publies par M. W. P. Yetis (I et II)." T'oung Pao 27 (1930): 366. Due to the extreme authority and generally unmatched erudition of Pelliot. this incorrect date has been followed without question by Western sinologists writing on the Shuowen (e.g., Roy Andrew Miller. "Problems in the Study of Shuo-wen chieh-tzu" (Ph.D. Diss., Columbia Univ., 1953], 297 n. 36; William G. Boltz, "Shuo wen chieh tzu," in Early Chinese Texts: A Bibliographical Guide, ed. Michael Loewe [Berkeley: The Society for the Study of Early China, 19931, 430; Mark Winter, "... und Cang Jie erfand die Schrift": Ein Handbuch fur den Gebrauch des Shuo Wen Jie Zi [Bern: Peter Lang, 1998], 32). But while Pelliot correctly follows the textual emendations of Yan Kejun (1762-1843) on the cyclical dates (changing jihai and wuwu to jimao and wushu according to Quan shanggu sandai qin han sanguo liuchao wen. 9 vols. (Taibei: Shijie shuju, 1969]. vol. 2. Quan houhan wen , 49.5b). one can only guess that he made a minor slip and mistook the ninth month of the Chinese calendar for "September," as opposed to the ninth month of the first year of the Jian'guang reign period of Han Andi, which ran from September 30 through October 28, 121 C.E.
(2.) Zhu Junsheng (1788-1858), Shuowen tongxun dingsheng (Beijing: Zhonghua shuju, 1998), 41a4. I use recently published separate editions of the major Shuowen commentaries rather than the standard gulin omnibus edition (published in 1930) because these newer texts are typographically and philologically far superior.
(3.) Duan Yucai (1735-1815), Shuowen jiezi zhu (Shanghai: Shanghai guji chubanshe, 2004), 14A48.7-8 and 12B7.9.
(4.) Shuowen jiezi zhu, 13B50.45.
(5.) As Xu Xuan states, "dividing words into mutually distinct bushou first began with Xu Shen" (Xu Xuan [917-9921, Songkan shuowen jiezi 6 vols. [Beijing: Xianzhuang shuju, 2005], 15A4.1-2n). Since the publication of the Zihui (1615), the lexicographic system of 214 classifiers (thereafter used in the Zhengzi tong and in the canonical Kangxi zidian) has been the standard--which is less than half of Xu Shen's original 540 bushou.
(6.) As Wang Yun points out. Xu Shen's Postface "uses the same methodology as the auto-biographical postface of Sima Qian" (Wang Yun [1784-1854], Shuowen jiezi judu (Beijing: Zhonghua shuju, 19981, 7a3-4).
(7.) Xu Kai (920-974), Shuowen jiezi xizhuan (Beijing: Zhonghua shuju. 1998), 7.3a3.
(8.) Gu Aiji (fl. 1718), Libian (Beijing: Zhonghua shuju, 2003), 1a2.
(9.) Yan Zhitui (531-597), Yanshi jiaxun huihu ed. Zhou Fagao (Taibei: Taiwan shangwu yinshuguan, 1994), 18. 125a8.
(10.) Yan Zhitui. Yanshi jiaxun huizhu, 18.121a5.
(11.) Dai Zhen (1724-1777), Dai zhen wenji (Beijing: Zhonghua shuju. OO6). 145-46.
(12.) Libian. 2b5-9.
(13.) Wang Niansun (1744-1832). Dushu zazhi (Nanjing: Jiangsu guji chubanshe, 2000), 188.8.131.52.
(14.) Dushii zazhi, 184.108.40.206-4.
(15.) Dushu zazhi, 220.127.116.11-10.
(16.) Dushu zazhi, 18.104.22.168.
(17.) Dushu zazhi, 22.214.171.124-20.
(18.) Dushu zazhi. 126.96.36.199 and 188.8.131.52-6. For a more extended discussion, see Wang Yinzhi (1766-1834), Jingyi shuwen (Nanjing: Jiangsu guji chubanshe. 2000), 32.63-67.
(19.) Dushu zazhi, 184.108.40.206-6, 220.127.116.11-20, and 18.104.22.168-16.1. Cf. "it was only later people who did not understand the ancient meaning and theretre ignorantly changed it" (Wang Yinzhi, Jingyi shuwen, I 6.20b7-8).
(20.) Dushu zazhi. 22.214.171.124-4.
(21.) Dushu zazhi, 126.96.36.199.
(22.) Dushu zazhi 188.8.131.52.
(23.) Dushu zazhi, 184.108.40.206-10.
(24.) Dushu zazhi, 220.127.116.11-3.
(25.) Dushu zazhi, 18.104.22.168-6.
(26.) Dushu zazhi, 22.214.171.124-15.
(27.) Dushu zazhi, 126.96.36.199.
(28.) Dushu zazhi, 188.8.131.52-17. Cf. the comment "later people intentionally added this" (Wang Yun. Shuowen jiezi judu. 2a10). For a more extended discussion. see Wang Yinzhi, Jingyi shuwen 32.42-44 and 32.52-63.
(29.) Shuowen jiezi zhu, np.
(30.) Shuowen jiezi zhu, np.
(31.) Shuowen jiezi zhu, np.
(32.) Shuowen jiezi zhu, np.
(33.) Shuowen jiezi zhu, 3A26.2-3.
(34.) Xunzi's Zhengming pian "Treatise on Making Words Correct" states: "words do not have a fixed meaning. Delimit the meaning in order to take command over the meaning. When the delimitation is settled, when the convention is established, refer to it as an 'appropriate meaning': if it is different from the delimitation then refer to it as an 'inappropriate meaning.' Words do not have fixed signifieds. Delimit the signified in order to take command over the signified. When this delimitation is settled, when the convention is established, refer to it as a 'word that has a signified.'" (Cf. Liezi. "signifieds have no [natural] signifier, signifiers have no [natural] signifieds. As for 'signifiers,' they are artificial and that's absolutely it" [Songben liezi (Taibei: Guangwen shuju, 1994), 7.1a13-14].) The Xunzi argues here that words are arbitrary and conventional, that signifier and signified are entirely fluid and therefore amenable to artificial manipulation and artificial fixation--that words and the things words refer to only become meaningful to humans after they have been narrowed down and exactly delineated. It was the job of sage-kings to determine conventionally correct language through prescriptively defining every single word in the lexicon and designating every thing in the world. The Xunzi further argues in the zhengming pian that this prescriptive language system. while it may have functioned perfectly in the past, had in more recent times broken down and failed: "but now the sage-kings are dead. the preservation of words is neglected, strange phrases have arisen, and words and their signifieds are in disorder." Should a new sage-king arise, his first job would obviously be to make words correct. The text goes on: "as for one who would rule as king establishing words: if words are defined and the signifieds distinguished, if this way of establishing words is enacted and the king's intentions are effectively communicated, then he will carefully lead the people and make them the same by means of correct words. Therefore, if someone splits phrases or arbitrarily creates words so as to bring disorder to correct words--causing the people to be doubtful and confused, so they have many arguments and quarrels--then refer to it as 'great wickedness.' This person's crime is similar to the crime of creating false tallies and measures [punishable by the death penalty accord to the Liji]. Therefore, among the king's people, none will dare to rely on strange phrases to bring confusion to correct words. Hence the king's people will be guileless, and if they are guileless then they will he easy to control." This is a rather stark view of language--language as a tool used by the elite to ensure the obedience of a subservient population. It is surely no coincidence that two of Xunzi's greatest students, Li Si and Hanfeizi, were advisors to the First Emperor of Qin, the first real emperor of China, who conquered all the other states and whose policies unified and standardized spoken and written language, among other things. This view of language integrally informs the length and breadth of jingxue and forms the ultimate basis of traditional Chinese philology. Chinese lexicographers and commentators were "harmless drudges" in no way, shape, or form--they were engaged in the deeply serious Confucian social engineering work of making words correct.
(35.) Shuowen jiezi zhu. 9A22.9-10.
(36.) Shuowen jiezi zhu, 5A25.6-7.
(37.) Shuowen jiezi zhu, 15A8.5-9.
(38.) Shuowen jiezi zhu, 2A30.13.
(39.) Shuowen jiezi zhu, 2B33.9.
(40.) Shuowen jiezi zhu, 5A11.4.
(41.) Shuowen jiezi zhu, 3A3.6 and 9A27.15.
(42.) Shuowen jiezi zhu, 7B38.17.
(43.) Shuowen jiezi zhu, 1A1.10.
(44.) Shuowen jiezi zhu, 844.12.
(45.) Shuowen jiezi zhu, 14B18. 18.
(46.) Shuowen shili (Beijing: Zhonghua shuju. 1998), 2b8-9.
(47.) What is meant by "encode an intention" is the historical individual's highly motivated choice of graphemic components used to construct the characters that write a particular word (all aspects of this process having profound sociopolitical implications according to the aforementioned Confucian view of language). The larger point is that advocating specific meanings of a word through graphemic selection can literally, insofar as humans are concerned. maintain or change the nature of the universe through manipulation of signifier and signified.
(48.) Shuowen jiezi zhu. 5A3.1.
(49.) Shuowen jiezi zhu. 10A34.1.
(50.) Shuowen jiezi zhu. 5B5.14.
(51.) Shuowen jiezi zhu. 13B 15.18.
(52.) Shuowenjiezi zhu, 15A10-16. Cf. "when the ancient graphic structures of the characters in classical texts were transmitted, it was not necessarily the case that they were always changed so as to accord with the graphic structure of zhuanwen" (Zhang Can [fl, 776], Wujing wenzi, in Ganlu zishu wujing wenzi [Beijing: Zhonghua shuju. 1985], 5a5).
(53.) Lu Deming (550-630), Jingdiun shiwen 2 vols. (Jinan: Shandong youyi shushe. 1991) 3.2a8.
(54.) Shuowen jiezi xizhuan. 39.3b2.
(55.) Shuowen jiezi zhu. 15A2.2-3.
(56.) Shuowen jiezi zhu, 2A19.5.
(57.) Shuowen jiezi zhu. 10A8.15.
(58.) Xuanying (fl. 645), Yiqie jing yinyi ed. Zhou Fagao (Taibei: Taiwan shangwu yinshuguan. 1992), 24.5. Cf. "in antiquity there were few written words ... but in later ages events became more complicated, and the number of written words changed and gruw. having been combined together as xingsheng" (Shuowen jiezi xizhuun, 39.1a4).
(59.) Wenzi mengqiu (Taibei: Yiwen yinshuguan. 1995), 1.1-2.
(60.) Sima Qian (145-86 B.C.E.). Shiji (Beijing: Zhonghua shuju, 1999), 4:147.
(61.) Shuowen jiezi zhu, 7A23.15. Presumably, three "agricultural fields" seemed like a better way of classifying words meaning "great abundance" to the Xin reformers--better, at least, than three "suns." As Xu Shen's illuminating glosses of the five words listed under the jing classifier (Shuowen jiezi zhu, 7A22-23) show, three "suns" (i.e., the word jing "bright") actually makes perfect historical sense as a classifier for these words.
(62.) Feng Yan (fl. 756). Fengshi wenjianji jianzhu ed. Zhao Zhenxin (Beijing: Zhonghua shuju. 2005). 2.8.2. The particle meaning "somewhat" here might also be adverbially read as "in a cockeyed fashion" but that may be over-reading Xu Shen's own etymological glosses into the prose of his Postface, which Feng Yan is rephrasing here.
(63.) Western scholars who have worked closely with the Shuowen recognize that there is something odd going on with guwen. For example, "the Shuo wen postface clearly contrasts ku wen script (dated to the pre-Ch'in period) with the small and large seal script types sponsored by Ch'in ... thus not all texts written in pre-Han script were labeled ku wen" (Michael Nylan, "The Chin Wen/Ku Wen Controversy in Han Times," T'oung Pao 80 : 87 n. 13): few are willing to state definitively what this might mean. One of the exceptions seems to be Serruys, who, following Wang Guowei (Wang Guowei, "Shizhou pian xulu" in Wang guanzang xiansheng quanji 16 vols. [Taibei: Wenhua shuju. 1968]. 7:2371), argues that guwen represents eastern Chinese dialects: -this would mean that when Shuo-wen gives a ku-wen graph, where the structure suggests a different reading than the hsiao-chuan graph, we can suspect a dialect difference between the central and the eastern States" (Paul Leo-Mary Serruys. "Prolegomena to the Study of the Chinese Dialects of Han Time According to Fang yen" [Ph.D. diss.. Univ. of California, Berkeley, 1955], 290 n. 91). While this is an interesting hypothesis, I only bring it up here as a contrast to what Xu Shen says in the Postface.
(64.) This rare find is extensively documented in Han historical records: see. for example. Ban Gu (32-92 C.E.), Hanshii (Beijing: Zhonghua shuju. 1997), 30:1706. 36:1969, 53:2414, and 88:3607.
(65.) Shuowen jiezi xizhuan. 16.7b6.
(66.) As Zhang Cang had been a high official under the Qin, his personal copies of texts (like the Zuozhuan) would presumably have been spared the flames of both the bibliocaust itself and the destruction of the Qin imperial library when the capital burned to the ground. Hence Zhang Cang becomes one of the living conduits through which the guwen tradition is passed across the otherwise impenetrable historical barrier of the Qin.
(67.) This parallelism in the Postface to the title of the work provides a strong argument for understanding the original title of the Shuowen jiezi to mean something like "analyzing the [three distinct] writing systems in order to explicate their [component] offspring characters."
(68.) Shuowen tongxun dingsheng. 2a3-4.
(69.) Shuowen jiezi zhu, 15A21.17-18. Cf. Wang Guowei. "Shuowen 'jin xu zhuanwen he yi gu zhou' shuo" in Guantang jilin 4 vols. (Beijing: Zhonghua shuju. 1959), 2:317-18.
(70.) See. e.g., the guwen classifier in Shuowen jiezi zhu, 7B16.
(71.) Shuowen jiezi zhu, 15A22.3.
(72.) Guantung jilin, 2:319-20.
(73.) Gui Fu (1736-1805), Shuowen jiezi yizheng 2 vols. (Beijing: Zhonghua shuju. 1998). 1.6b10-7a2.
(74.) Wang guantang xiansheng quanji, 7:2377-2378. Cf. "no variant [chong] indicated that the [zhuanwen] is the same as in [guwen] and [zhauwen]" (Xue Shiqi, "Chinese Lexicography Past and Present," in Lexicography: Critical Concepts, ed. R. R. K. Hartmann, 3 vols. [London: Routledge. 2003], 2:162).
(75.) Ding Fubao (1874-1952). ed., Shuowen jiezi gulin 66 vols. (Shanghai: Shangwu yinshuguan, 1930).
(76.) As Serruys remarks in passing, "if they were different in graphic structure or in phonetic composition they may be different words" ("Prolegomena to the Study," 291 n. 98).
(77.) That this was generally recognized by the tradition is further exemplified in the following question (as one has to first assume it to even begin to question it): as for the xiaozhuan, could it really have the original meaning that Li Si intended in creating those written words?" (Gui Fu, Shuowen jiezi yitheng, 1.3a5).
(78.) The Shuowen Postface has been partially translated several times in Western languages: Miller, "Problems in the Study of Shuo-wen chieh-tzu," 273-97; Kenneth Lawrence Them, Posy'ace of the Shuo-wen Chieh-tzu: The First Comprehensive Chinese Dictionary (Madison: Dept. of East Asian Languages and Literature, Univ. of Wisconsin, 1966), 8-78; Nils Goran David Malmqvist, "Xu Shen's Postface to the Shuo Wen Jie Zi," in On Script and Writing in Ancient China: Translations with Annotations, ed. David Pankenier (Stockholm: Foreningen for Orienta-liska Studier, 1974), 48-53; Francoise Bottero, Semantisme et classifiction dans l'ecriture chinoise: Les systemes de classement des caracteres par cles du Shuowen jiezi au Kangxi zidian (Paris: Institut des Hautes Etudes Chinoises, College de France, 1996), 17-43 and 58-60. It has also been completely translated once: Winter, "... und Cang Jie erfand die Schrft," 557--6. As my interpretations are often somewhat different. I translate it here anew, using the Song woodblock edition of Xu Xuan (op. cit.) us base text.
(79.) Significant portions of these first two paragraphs are direct citation of the Yijing (in Ruan Yuan [1764-1849]. Shiscmjing zhirshu 2 vols. [Shanghai: Shanghai guji chubanshe, 1997], 1:86b). There are also large chunks lifted from the "Yiwenzhi" with some modifications: see Honshu, 30:1704-6 and 30:1719-22. Sporadically throughout the Postface Xu Shen and Xu Chong also quote and make allusions to the Shangshu, Shijing, Zuozhuan, Guliang zhuan, and Lunyu.
(80.) There is probably an allusion here to the "Xici" passage "heaven hangs down celestial phenomena: upon observing what is auspicious and inauspicious in them, the sages represented it [by means of the hexagrams!" (Shisanjing zhushu, 1:82b), and perhaps also to the Shangshu passage "in order to hand down norms to your descendents" (Shisanjing zhushu,1:227b). What is being handed down by Paoxi (a.k.a. Fuxi), therefore, is a normative manner of representing omenological phenomena, to be used in divinatory practice. The eight trigrams are a standard (and standardizing) method of representing the distinctions and nuances between good and bad omens, upon which divination using the Yijing is then based (the punning polysemy of the word xiang here on the one hand "celestial phenomena" and on the other "to represent." "representation." provides the ultimate justification for this). Compare the syntax of this clause to similar passages in other Han writers, such as Zhao Qi (107-201 C.E.), "Mengzi was ashamed that he would die without being known by anyone; for this reason he handed down normative words in order to teach his message to later people" Shisanjing zhushu. 2:2662), and Yang Xiang, "the great measures examine and reexamine; in heaven displaying celestial phenomena, it hands down its exemplars. The Fathoming (autocommentary) says: as for 'the great measures examining and reexamining.' this means heaven hands down that which represents the correct" (Dim Chuek Lau et al., ed., Taixuan jing zhuzi suoyin [Xianggang: Shangwu yinshuguan. 1995], 38; cf. Michael Nylan, tr., The Canon of Supreme Mystery by Yang Hsiung: A Translation with Commentary of the T'ai Hsilan Ching [Albany: State Univ. of New York Press, 1993], 321). Malmqvist's translation. "to hand down the regular patterns" ("Xu Shen's Postface to the Shuo Wen Jie Zi," 48), shows that his understanding of the syntax is similar.
(81.) Quipu--an English word etymologically derived from Quechua Mayan khipu--is the modern anthropological term for the practice of knotting cords in order to record events, keep accounts, etc.
(82.) Yijing citation (Shisanjing zhushu,1:87b).
(83.) Following Duan Yucai's reading of It [*gj3g] as [*gjag] (Shuowen jiezi zhu, 15A1.11). Cf. the comments in Winter. "... und rang Jie erfand die Schrift," 558 n. 4.
(84.) Yijing citation (Shisanjing zhushu. 1:87b ).
(85.) Yijing citation (Shisanjing zhushu, 1:56c).
(86.) Yijing citation (Shisanjing zhushu, 1:56c). I am here following Gui Fu's suggested emendation of to based on Wang Bi's and Kong Yingda's reading of the Yijing passage (Shuowen jiezi yizheng, 49.2b9: Li Xueqin Shisanjing zhushu zhengli ben 46 vols. [Taibei: Taiwan guji chubanshe, 2001], 2:212), and also Wang Yun's suggested interpretation of (Shuowen jiezi judu, 29.2b8-10: Shisanjing zhushu zhengli ben, 2:212).
(87.) This "therefore" syntactically stretches back all the way to the first wen--this polysemic definition of wen is simultaneously the "simple" written forms of particular words and the ethical norms (within the social hierarchy) that words as ming are intended to portray. Even the historical invention of writing, Cangjie's first use of "simple" rebus-graphs to write down a limited number of spoken words, is inextricably bound up here in the Postface with the complicated ritual behavior, including language, prescriptively required, for example, of social superior and inferior.
(88.) All zi "offspring characters" must be derived, according to this, from a specific wen "writing system" that intrinsically bears profound cultural implications (guwen, zhouwen, or zhuanwen), the "offspring" of the mating of at least two graphs of the parent wen. This is the earliest explanation of why the word zi "offspring" also means "written word"--and it seemingly only means "character" as graph because the graphemic structure of each zi is understood to have been parturited from the structural components of the parent wen--produced first by combining graphs together to form complex graphs. Although she unfortunately uses the English word emblem. Bottero's explanation is not radically dissimilar (see Francoise Bottero, "Revisiting the wen and the zi The Great Chinese Character Hoax," Bulletin of the Museum of Far Eastern Antiquities 74 : 14-33).
(89.) Ti is here a calligraphic distinction rather than a structural one. In other words, according to Xu Shen the individual written characters may have appeared dissimilar during the stretch of time spanning from the Yellow Emperor to the Western Zhou, but they all relied on the same structural pattern of a specific writing system that had cultural implications (i.e., guwen). The stated claim is that although the calligraphic appearance changed, guwen as a writing system never changed its set of graphic components, nor its manner of putting them together, nor its cultural implications (the role of the politically ordained meaning of the otherwise arbitrary word--encoded in the writing system with classifiers by Cangjie--in social activity and its continuation through time). Thus guwen is considered to be an unbroken tradition from Cangjie down to the Western Zhou. Hulsewe also interprets ti in this manner; see Anthony Francois Paulus Hulsewe, "The Shuo-wen Dictionary as a Source for Ancient Chinese Law," in Studio Serica Bernhard Karlgren Dedicata: Sinological Studies Dedicated to Bernhard Karlgren on his Seventieth Birthday. October Fifth, 1959, ed. Soren Egerod and Else Glahn (Copenhagen: Ejnar Munksgaard, 1959), 248.
(90.) On the seventy-two sacrifices performed on Mt. Tai, see Sima Qian, Shiji, 28:1355-1404.
(91.) Wang Anshi similarly mentions that "as for writing being used in later ages and for a really long time, the early kings established schools in order to teach it" (Wang Anshi [1021-1086]. Jin zishuo biao in Wang anshi zishuo ji ed. Zhang Zongxiang and Cao Jinyan [Fuzhou: Fujian renmen chubanshe, 2005] 161.3).
(92.) Cf. Zheng Xuan OM (127-200), Zhouli zhengzhu 2 vols. (Taibei: Taiwan shangwu yinshuguan, 1967), 14.3b4. This liushu "six ways of writing" is not described in the Shuowen Postface as a set of graphemic analytical principles (which it becomes later in the lexicographic tradition, and was often used anachronistically to interpret the microstructures of the Shuowen). but rather as a pedagogical tool used to teach illiterate children how to begin to approach guwen. in order to learn how to write from the ground up. I have therefore translated them here in the prescriptive active. The other liushu"six types of writing" in the Postface is an entirely different list of the writing systems prevalent from the Xin period up to the time of Xu Shen. As Duan Yucai notes, the later list is the same word as the liushu of the Palace Protector according to the rites of Zhou but a different signified" (Shuowen jiezi zhu, 15A16.11).
(93.) Strictly speaking, a huiyi is graphemically a combination of two or more classifiers.
(94.) According to the Postface, this is the first cultural break from the longstanding guwen tradition and has simply massive historical consequences. It must be remembered that Scribe Zhou is supposedly compiling this revisionary lexicon right at the end of the Western Zhou period (in fact only a few short years before the capital was sacked and burned, and the line of legitimate succession broken), and his tradition-breaking lexicographic work is implicated in the cultural collapse of the Thou state. As Sima Qian notes, during this time "Zhou is about to be utterly destroyed" (Shill. 4:145).
(95.) Xu Shen is implying here that Confucius and Zuo Qiuming purposely chose the conservative ancient writing system, guwen, in opposition to the somewhat morally suspect and altered system of Scribe Zhou (which had been in circulation for some three centuries by their time), and further, that if one wants to properly understand the intentions of such sages as Confucius and Zuo Qiutning. one must perforce understand guwen, the writing system these two writers specifically chose to use to write down the classics.
(96.) Following Gui Fu and Wang Yun's emendation of to (Shuowen jiezi yizheng, 49.6a5: Shuowen jiezi judu, 29.4a10).
(97.) Cf. "when those who ruled as kings regulated the world, they necessarily caused cart axles to be the same width and writing to use the same writing system; by these means they educated the people" (Zhang Can, Wujing wenzi, 1b4-5). If cart axles are the same width then the wheel-ruts in dirt roads will be the same everywhere, facilitating practical ease of transport across long distances.
(98.) The Postface is here ascribing cultural unity to the pre-Eastern Thou period. The breakdown of the political body and the material and social culture that unified it has a direct correlation to the chaotic condition of having no fixed standards of measurement, transportation, laws, dress, speech, language, and writing. Scribe Zhou's break with guwen (as opposed to Confucius and Zuo Qiuming's attempt to uphold the old standard) has prefigured the general breakdown of order during the Spring and Autumn and the Warring States periods.
(99.) Attempting to break entirely from the past in all things (at least according to this narrative), the Qin reformers seem to have recognized that basing the new standard writing system on zhouwen would be best--guwen was simply too conservative and traditional, and too deeply associated with the classics as written by Confucius. Thus Xu Shen is here claiming that they did to zhouwen what Scribe Zhou had done to guwen. i.e., more simplification and reform of the structural elements, which have cultural and even constructed-cosmological implications. These three writers (who, it must be remembered, were implicated in burning books, murdering scholars, and committing the greatest cultural crimes in the history of early China according to Han historiography--indeed, they were considered to have almost succeeded in entirely wiping out reliable knowledge of an idealized past) were generally feared and demonized by Han literati.
(100.) At exactly this point, the Postface has historicized the three distinct writing systems basic to the macrostructure and microstructures of the Shuowen: guwen, zhouwen (dazhuan). zhuanwen (xiaozhuan). Each of the major writing systems after guwen is a simplification and reform of the structural elements of the prior one, and thus represents a visible break in the tradition and a cultural change. By the time of official script, which is a calligraphic style of zhuanwen (based on the third historical writing system), the structural break has become so profound that guwen is for the most part unrecognizable and unreadable to those literate only in lishu.
(101.) Again, perhaps in a "cockeyed fashion" (this according to Xu Shen's Shuowen etymological gloss).
(102.) Presumably this is what Wang Mang had obnoxiously created by "fixing" guwen, or Eastern Zhou mutant graphs that did not correspond exactly to Xu Shen's postulated guwen. See the discussion in Zhang Nengfu Lidai yuyanxue wenxian duben (Chengdu: Bashu shushe, 2003). 13 n. 22.
(103.) Again, since this four character parallelism exactly matches the syntax of the title Shuowen jiezi, it reiterates that Xu Shen's main purpose is not, as was true of other Han scholars, to argue directly from the graphic components of Qin lishu--or even Qin xiaozhuan--in order to interpret words in the texts of the classics (e.g., the apocryphal "broken-character" readings prevalent during the Han), but rather to analyze characters by first historicizing the different writing systems of which they are the "offspring"--which obviously further leads to a better understanding of the original intentions of the sages who specifically wrote the classics with the oldest writing system (guwen).
(104.) Accepting Xu Kai's emendation of [*djeg] for Xu Xuan's [*djeg]; see Shuowen jiezi xizhuan. 28.8a5: Shuowen jiezi zhu, 15A.19b3.
(105.) Shangshu citation (Shisanjing zhushu. 1:141c).
(106.) That is, not to make a hush of them for one's own agenda--as was done by Scribe Zhou. the Qin officials, Wang Mang, jinwen scholars, etc.
(107.) Lunyu citation (Shisanjing zhushu, 2:2518a).
(108.) Lunyu citation (Shisanjing zhushu, 2:2457b).
(109.) Yijing citation (Shisanjing zhushu, 1:79b). In other words, if the emperor understands and respects the history of the writing systems (i.e., guwen, zhouwen, and zhuanwen)--"the root established"--and supports guwen scholarship on the classics--"the extreme subtleties"--the state will function as it should: morally, smoothly, and without disruption.
(110.) For a description, discussion, and full analytical translation of the 540 classifiers, see Paul Leo-Mary Serruys, "On the System of the Pu Shou in the Shuo-wen chieh-tzu," Bulletin of the Institute of History and Philology, Academia Sinica 55.4 (1984): 651-754.
(111.) This innocuous-seeming number is the most important single piece of lexicographic information we have in the Shuowen Postface. Xu Shen is here telling us that, according to his calculations, of the 9.353 units of the particular writing system examined, he has been able to track changes in the structure and give examples of such for almost 13% of them. And while 13% may not seem to be a large amount of structural change overall, imagine a Han dynasty scholar fundamentally misunderstanding the meaning and usage of about one out of every ten words in the texts of the classics. The significance of the figure 13% should now be clear.
(112.) Yijing citation (Shisanjing zhushu, 1:76a).
(113.) Accepting Xu Kai's emendation of for Xu Xuan's (Shuorven jiezi xizhuan. 30.1b1; Shuowen jiezi zhu, 15B.1b5).
(114.) Yijing citation (Shisanjing zhushu, 1:89a).
(115.) Yijing citation (Shisanjing zhushu, 1:87c).
(116.) That is, ancient sage-king Yao, from whom the Liu house claimed to descend.
(117.) With five characters in a row containing the water classifier, it makes one almost want to read this passage as "distant and near are covered over and enriched, drenched to overflow by an abundance of heavy rain [good for the crops and hence for everyone]"--a kind of water-filled encomium of the greatness of the Han--which is exactly the reading you get if you look up the Shuowen gloss of each of the five graphs. Sometimes it is difficult to tell whether or not Xu Shen is using a normal meaning of a word or using his own etymological glosses to inform his lexical choices in the Postface.
(118.) Lunyu allusion (Shisanjing zhushu, 2:2500b). (in Tang Wen, Zheng xuan cidian [Beijing: Yuwen chubanshe, 2004], 228).
(119.) On interpretive cum philological grounds Duan Yucai argues that this [*ling] "spirit" or "charisma" should be understood as (*ljingh] "good"; Miller follows this reading ("Problems in the Study," 282 and 282 n. 25). Here, however, we take no umbrage with an elite ancestor worshiper talking about a vital charismatic "spirit" of excellence passing down through generations in particular aristocratic families, especially his own.
(120.) I follow Duan Yucai on the identification of the Shao here as Shaoling (Shuowen jiezi zhu, 15B4.7).
(121.) Shijing allusion (Shisanjing zhushu, 1:482b).
(122.) Shijing citation (Shisanjing zhushu, 1:440a--b).
(123.) Guliang zhuan allusion (Shisanjing zhushu, 2:2374b).
(124.) I follow Duan Yucai's interpretation of caomang; see Shuoiven jiezi zhu, 15B7.10. Miller translates it as "rustic minister," while stating that the term means "unemployed" ("Problems in the Study," 285 and 285 n. 2).
(125.) Yijing citation (Shisanjing zhushu, 1:17b).
(126.) Yijing citation (Shisanjing zhushu, 1:14b).
(127.) Yijing citation (Shisanjing zhushu, 1:93c).
(128.) Jia Kui, from whom Xu Shen received his training, was an important guwen scholar who participated in the White Tiger Hall conference of 79 C.E. He was a major player at court who almost single-handedly convinced the Emperor to promulgate the Edict of 83 C.E.--establishing the Zuozhuan, Guliang zhuan. and the guwen texts of the Shijing and Shangshu as official educational curriculum--which was a historic victory for the guwen advocates (as, following this political masterstroke, nearly all Jia Kui's disciples, evidently including Xu Shen, received at least temporary positions at court). See Fan Ye (398-445 C.E.). Hou hanshu (Beijing: Zhonghua shuju, 1996), 36:1234-41; Tjan Tjoe Som, Po Hu T'ung: The Comprehensive Discussions in the White Tiger Hall, 2 vols. (Leiden: Brill, 1949-1952), 1:153-64.
(129.) Zuozhuan citation (Shisanjing zhushu. 2:2143c).
(130.) Cf. Shisanjing zhushu. 1:87c-88a.
(131.) Cf. Shisanjing zhushu, 1:190a. More than likely Xu Chong selected this exact passage from the "Hongfan" because it suits the rhetorical needs of the petition (i.e., these two books are worthwhile things done by an official with ability). The rhetorical force of this citation is that Emperor An's acceptance of the books will help to cause prosperity throughout the Han empire.
(132.) I follow Duan Yucai's reading here, which emends ge to he; see Shuowen jiezi zhu, 15B9.7; Miller, "Problems in the Study." 288-91 n. 16.
(133.) Lunyu allusion (Shisanjing zhushu. 2:2483b).
(134.) This is a direct allusion to the macrostructure of the Erya.
(135.) Wei Hong was a guwen scholar of the Shijing and Shangshu and likely wrote the "Great Preface" to the Maoshi. See Fan Ye, Hou hanshu. 27:936 and 79:2584: Tjan Tjoe Som, Po Hu T'ung, 1:150.
(136.) On the Vermilion Bird Gate. see Hans Bielenstein, "Lo-yang in Later Han Times." Bulletin of the Museum of Far Eastern Antiquities 48 (1976): 33.
|Printer friendly Cite/link Email Feedback|
|Publication:||The Journal of the American Oriental Society|
|Date:||Jul 1, 2013|
|Previous Article:||Lucrarile Simpozionului International Cartea, Romania, Europa. Editia a 11-1, 20-24 septembrie 2009.|
|Next Article:||Etana in Eden: new light on the Mesopotamian and Biblical Tales in their semitic context.|