Lurie 4: kundoku 1________and rhetoric

Again and again I've had to expand and revise portions of this post in conjunction with working on the next, related post. I've rendered the font of significant revisions in neon green for those who are examining this topic, with what is probably the most important passage shown against a black background for added contrast. That passage relates directly to material addressed in greater depth in the next post--the final entry in the series of posts on this book--which I should be able to put up in a day or two.

...incorporating a couple additional online documents as references, and expanding the discussion of the relationship of kundoku to the history of writing.

In light of some passages from one of the above-mentioned papers, overnight I've expanded the post by about another 900 words. There were some passages in Frellesvig's paper that deserved citing to maintain balance an fairness in this critique.

The additional passages demonstrate the degree of complexity of the subject matter, which would seem to legislate against making sweeping, unsupported statements.

Although it has occurred to me that perhaps I should at some point break this analysis and critique down into section, for the time being I'm going to move along to the next aspect of this critique: the East Asian Cultural Sphere and the role of China. 

If I do break this down at some point after having the chance to digest it and integrate the ramblings into a more coherent whole, the trends in the analysis set forth in this post would seem to indicate a separate section devoted to the content of kundoku, and another section dealing with rhetoric of kundoku in which other rhetorical dimensions of the presentation of Lurie's arguments are addressed. 

In fact, I' still in the process of grasping the material at hand in a comprehensive manner.

On p. 180, under the section subtitle of

THE SIGNIFICANCE AND HISTORY OF KUNDOKU

Lurie writes:

Much of the remainder of this book is devoted to the implications of kundoku for the history of writing, but there are four particular points to emphasize here: it is interlingual, reversible, productive, and in many cases, invisible. The interlingual difference of kundoku means that linguistic difference need not be reflected in writing, difficult though it is for us to overcome the assumption that all texts must be written in one and only one language, in the sense that this sentence is ‘written in’ English. This interlinguality has profound implications for Japanese cultural history. Many have seen that history marked by a fundamental bilingual contrast between Chinese and Japanese, but because even texts that originated in China could be read as Japanese, traditional reading practices did not necessarily involve awareness of texts written in one language or the other. This means that the diversities of literacies discussed thus far included even linguistic differences (potentially in readings of the same texts).

To say that kundoku is reversible means that it can be run in both directions: it is a method of writing as well as of reading. It was used to produce Japanese-language logographic texts (or at least, logographic texts that could potentially be read in Japanese) as well as to read/translate texts with non-Japanese origins.

Although I had sought to simply dismiss kundoku because of the weak evidence Lurie presents in defense of certain statements related to its introduction to Japan, because he makes it a central tenet of his thesis, as per the above passage, it is necessary to treat it more thoroughly, and in doing so to reposition kundoku at its proper place in the historical horizon. 

Lurie's assertion regarding the “implications of kundoku for the history of writing” strikes me as mere hyperbole, because I don't find his framing of the topic to be correct, for one, and there are significant questions left unasked. I do address one specific lapse in his treatment of the topic toward the end of the post, and that deals with the lack of a comparison of the historical development of Korean and Japanese with respect to kundoku. First and foremost, however, my take on kundoku differs fundamentally insofar as I don't agree with the assertion that kundoku per se was a method of writing. In short, I think it would be more accurate to say that the continual application of kundoku reading practices to writing texts with Chinese characters facilitated the development of the Japanese writing system. 

With regard to the mention of "interlinguality" in the portion highlighted against a dark green background, Lurie never discusses the "profound implications for Japanese culture". While that is a statement with which I would basically be in agreement, I question whether Lurie understands it in the same manner that I do. The way I see it is in relation to the hybridity of the Japanese writing system, which is a concept that is touched upon by Aldridge and which I discuss below (or in the next post).  

Before taking up various aspects in detail, I want to point out two basic fallacies with the arguments Lurie makes in the above passage.

First, kundoku is a practice of reading, not writing, and in this regard, Lurie’s assertion about texts composed in Chinese characters being written in more than one language is simply ludicrous. Japanese that were literate in the eighth century could read (comprehend) the meaning of Chinese text without vocalizing the text. Conversely, Japanese texts that were composed using Chinese characters as stand-in phonographs would have been completely unintelligible to someone literate in Chinese. 

In order to vocalize the texts written in Chinese characters in order to render their content accessible to people not literate in Chinese, kundoku reading practices could be employed in a manner such as to transpose the Chinese syntax of the text into Japanese syntax. By extension, annotation indicative of the kundoku reading could facilitate access to the texts for those not fully literate. 

It bears noting at this point that Chinese characters remain a mutually intelligible medium of written communication throughout much of East Asia among all but the youngest generation of the literate public. A Japanese person travelling in Taiwan, for example, could communicate on a rudimentary level simply through scrawling out a few Chinese characters. This is due to the logographic nature of the characters.

This takes me to the second point, which relates to the assertion Lurie makes regarding “traditional reading practices”. 

With regard to the history of writing, strictly speaking, kundoku practices were tools adopted to facilitate the adaptation of Chinese characters as the building blocks of a Japanese writing system. Perhaps this assertion could be taken to presume foreknowledge on the part of the Japanese that they were on the road to creating their own writing system. Even if I were to admit that objection, it would not change the fact that Lurie does not qualify that by "traditional reading practices" he meant the practices by which Japanese and Koreans who had first encountered writing attempted to read CHINESE TEXTS. Does anyone really think that this is what he meant to imply? Anyone attempting to make that objection would have to be prepared to respond to the objection that he is conflating a "traditional reading practices" with the deciphering of foreign texts that were not written in a Japanese writing system. Again, anyway you look at it, rhetoric is what comes out in the final analysis.

That is to say, at the point in time when kundoku practices were being used there was as yet no “tradition” per se in Japanese (or Korean, for that matter). The period in question (approximately the 5th through 9th centuries with respect to Japan) represents a developmental stage, and I assert that it is anomalous to characterize kundoku practices as “traditional reading practices”, at least without qualifying the texts that were the objects of said practices. Kundoku reading practices could be referred to as tradition reading practices with respect to hentai kanbun, which continued to be composed even after it was superseded by the advent of the full-blown Japanese writing system. Tradition is a loaded term that needs careful attention in this context.  

Accordingly, it is further ludicrous to assert that the readers of a text composed in Chinese characters had no awareness of whether the text was written in one language or another. 

I've decided to do a little Internet-based research and found two references to introduce in relation to the complex subject matter of classical Japanese, which the following author--Bjarke Frellesvig--refers to as "Old Japanese" in the following paper. The Old Japanese terminology may present an interesting possibility for an analogous contrast class with respect to Old English vs modern English. Old English was a Germanic language brought about by the Norman conquest of England, and in that sense, there is a parallel in that the written language perhaps as different from modern English as hentai kanbun is to modern Japanese. On the other hand, Chinese was not a script that was imposed on the Japanese by a conquering power.

http://conf.ling.cornell.edu/japanese_historical_linguistics/1%20Writing%20and%20Sources.pdf

In section 1.2.1 on p. 3:

… Kanbun, which properly signifies writing in classical Chinese, is often thought of as a form of Japanese which requires some transposition to be 'read' in Japanese (in reality translated into Japanese) When trying to understand this, two factors must be kept in mind: First, early writers were bilingual in written Chinese and Japanese. Second, writing itself was from the outset associated with Chinese. Thus for a person with Japanese as his native language, both writing and reading would very often involve some element of translation, into Chinese when writing and into Japanese when reading. 

A literate Japanese person in the eighth century certainly knew that his copy of the Analects was written in Chinese, and that tanka poems, even the earliest tanka poems in the Man'yoshu from the Hitomaru collection (referred to as Abbreviated Form poems, see Lurie pp. 277-2819), which are written almost exclusively in Chinese characters according to Chinese literary conventions, were Japanese. The poem itself, however, could be vocalized in either spoken Chinese or Japanese, assuming one was familiar with the kundoku protocol employed. That would not alter the fact that the reader knew it to be a Japanese poem (和歌), as opposed to a Chinese poem (漢詩).

It is a fact that Japanese incorporated devices indicative of Japanese syntax in some of their Chinese style texts that were written solely in Chinese characters, but it is my understanding that such texts are referred to as hentai kanbun,--as contrasted to kanbun--if I have understood the passages quoted from Ishigami and Frellesvig correctly. 

According to the second reference I found online:

http://faculty.washington.edu/eca1/pdf/JK-20.pdf

Analysis and Value of Hentai Kanbun as Japanese

By EDITH ALDRIDGE, University of Washington

From the publication, Japanese/Korean Linguistics 20. Edited by Bjarke Frellesvig and Peter Sells. Copyright © 2011, «GreetingLine»

At least since Motoori (1798), there has been an awareness in the field that hentai kanbun texts were intended to be read in Japanese. As for the com-position of the texts themselves, the general assumption seems to be that hentai kanbun is some sort of hybrid, containing elements of both Chinese and Japanese

I have highlighted Aldridge's use of the term hybrid here in relation to Lurie's mention of interlinguality above.

I think that this point is important enough to add the final category of Japanese text as defined by Ishigami in terms of "wabun", so I'll repost the passage from Ishigami (Centrality and Marginality of Ancient Documents, p. 43):

1.2.2 Kanbun, hentai kanbun and wabun

    In simple terms, the ancient written language of Japan embraced three partial systems.

1.       Kanbun or Chinese writing. 

This consists of its grammar, characters (Chinese characters – henceforth to be referred to, according to Japanese practice as kanji – and their Chinese pronunciations), and vocabulary.

2.       Hentai kanbun or modified Chinese writing (otherwise Japanized Chinese writing). 

This consists of its grammar, kanji (including kokuji [Chinese characters conceived in Japan], the Japanese method of using characters, and Japanese pronunciations), and vocabulary (including the Japanese method of using words).

3.      Wabun or Japanese writing. 

This consists of its grammar, kanji, the 

katakana and hiragana syllabaries, and its 

vocabulary (both Chinese and Japanese words).

Japanese tradition becomes established with the appearance of the kana syllabaries in an integrated textual form comprising kanji Chinese characters in their manifold forms of reading, which are discussed below. That is what Ishigami refers to as “wabun”, which is translated as “Japanese writing” in the above-quoted passage. It would appear that Lurie attempts to collapse hentai kanbun and kanbun, as per p. 181-2, where he states:

    The catch-all term used by modern scholars to label styles of writing that mix Japanese and Chinese word order and usage is hentai kanbun, which can be translated as “variant form Chinese writing,” although precisely because of kundoku, “Chinese writing” is a misleading way of translating kanbun. A looser rendition of hentai kanbun more in accord with the actual history of writing would be “mixed logographic writing”. The structural hallmark of these styles is simultaneous use of elements of the grammar of literary Chinese—especially particles but also word order—along with the grammar of Japanese, the influence of which is usually explicit in Object-Verb ordering of phrases and implicit in the kundoku process that governs the entire text.

Thus far in this paragraph Lurie has collapsed kanbun and hentai kanbun into one undifferentiated mass with respect to writing. It is not clear what about his definition of hentai kanbun as “mixed logographic writing” is “more in accord with the history of writing”, as it is not clear what he means in terms of a mixing of logographs. Except for the few Japanese developed characters (which are called “kokuji”, see Ishigami above), all of the written characters are Chinese logographs.

It becomes clear from what follows that Lurie is in fact trying to conflate reading with writing, and this is perhaps part of the reason he tries to force a definition of kundoku as a method of writing.

Continuing on p.182:

… Inasmuch as the arrangement of characters is at times consistent with the rules of literary Chinese and at times not, this could be called a mixture, but both arrangements are recuperated into the Japanese order when the text is read via kundoku.

Here, it is once again somewhat unclear what he is referring to as a mixture; that is to say, is it a textual mixture comprising both Chinese and Japanese (i.e., kundoku) grammatical patterns? What exactly does the phrase "at times consistent with the rules of literary Chinese and at times not" mean? 

In any case, it is clear that in the final analysis he is referring not to reading the text according to the grammar in which it is written, but to applying kundoku reading practices to vocalize the text according to Japanese syntax, regardless of the syntactical composition of the text. "Invisible" indeed! 

What I'm trying to draw out and emphasize here is that Lurie does not even offer a single point of differentiation between kanbun and hentai kanbun, and then he conflates the processes of writing kanbun with the process of applying kundoku practices to reading aloud (or even silently to oneself) in Japanese syntax.

It would seem that kanbun should be largely relegated to reading and writing according to classical literary Chinese, whereas hentai kanbun compositions manifest at least two distinct stages along the evolutionary trajectory from reading Chinese in Chinese to writing and reading in Japanese. Those stages could been roughly broken down into: 

1. Reading Chinese in Chinese (kanbun)

2. Reading Chinese with some SVO to SOV character re-ordering in mixed Chinese and Japanese syntax with both Sino-Japanese and vernacular pronunciations (hentai kanbun)

3. Reading Chinese with Japanese syntactical character re-orderings and phonographs indicative of kundoku specifics in a manner that approaches that of the Japanese writing system (senmyo-gaki). 

According to the definitions given by Ishigami and Frellesvig, kanbun was written by the Japanese in Chinese syntax, and was meant to be read in Chinese syntax.

According to Aldridge, hentai kanbun was intended to be read in Japanese syntax. Therefore, hentai kanbun is the form of writing that reflects the stage of writing most highly informed by the reading practices of kundoku in order to facilitate vocalization of the text in Japanese syntax for readers without prior knowledge of the specific kundoku protocol operative in a given text, and eventually came to incorporate logographs used as stand-ins to representing phonemes. 

Accordingly, I take the “hentai kanbun”, which is translated as “modified Chinese writing” or “Japanized Chinese writing” by Ishigami to represent an intermediary stage of development from the trajectory that spans the gamut from reading Chinese texts in Chinese to establishing a Japanese writing system that uses Chinese characters as building blocks, comprises Japanese phonetic syllabaries, and represents Japanese grammar. The highest form of hentai kanbun was senmyo-gaki, or mixed style texts, which incorporated the reduced sized phonographs employed in a superscript or subscript as an aid to reading the text, and this form developed specifically in conjunction with texts that were intended to be vocalized aloud. The practice of the reduced-size characters is still in use today in wabun. 

It is in this regard that I consider kundoku to have become an obsolete practice once the Japanese writing system was fully established and began to serve as the basis of the Japanese literary tradition. The full-blown Japanese writing system for inscribing "wabun" incorporated kundoku in the form of kunyomi readings of characters and related auxiliary scripts, such as okurigana and furigana, indicating readings.

That is to say, the term kundoku itself refers to an archaic practice that held sway only during the formative stages of the Japanese writing system; kundoku does not refer to a “method of writing” as Lurie asserts in the above-quoted passage, and repeats in the following passage. Although kundoku itself was superseded upon the appearance of the Japanese writing system proper, it continued to be used in conjunction with hentai kanbun style of composition, which remained popular in Japan through the Edo period.  

On pp.205-7, Lurie writes:

Kundoku was not the only method of writing in early Japan. In adapting the ‘Chinese’ script to inscribe the Japanese language, another strategy was to use the syllabic values of the Chinese graphs as phonographs for similar Japanese syllables.

In the above quoted passage, Lurie is referring to the pronunciation of the characters in terms of their “syllabic values”. Here, it should be pointed out that in Japanese there are two categories of readings for Chinese characters: onyomi  and kunyomi. The onyomi reading refers to the Sino-Japanese pronunciation of the character, and the kuniyomi reading refers to any of a number of Japanese readings of the same character. Note that the character for “kun” in the two-character compound of “kunyomi” is the same “kun” as the two-character compound of “kundoku”. In fact, both of the Chinese characters are the same, with the only difference being the reading of the second character. That is to say, the reading “doku” of the second character in the pronunciation of the compound as “kundoku” is the onyomi reading of the character, with the reading “yomi” being a kunyomi reading of the same character, which has the semantic content (i.e., meaning) of “reading”.

Though I am not familiar with a broad range of scholarship on kundoku (Lurie goes through some in endnote 4.7), it would seem to me that the practices of adopting certain logographs for their phonetic content would be integral to developing the pedagogical tools required in order to adapt the characters as a script for writing text in a manner that would facilitate the reading aloud of the text in the syntax of the spoken language of the country adopting the characters to form a writing system. Most of the writing I've seen, however, seems to limit kundoku to practices related strictly to what could be defined as logographic (semantic) mapping and syntactical transposition. I find that somewhat problematic.

One point that is tricky here is that with respect to Chinese character readings in Japanese, syntax in fact overlaps with semantics with respect to the different kunyomi readings of individual characters and their associated grammatical inflexion, which appear in the form of verb endings that have to be rendered in phonographs or otherwise indicated. 

I’m not sure to what extent kunyomi readings were employed at this stage of development, however. At any rate, I would thus be inclined to include it under the rubric of kundoku, as the practice of including auxiliary inscriptions composed of logographs adopted as stand-ins for representing their associated pronunciations instead of meanings was a practice adopted to facilitate “reading” in the native tongue. Lurie seems to exclude this practice from kundoku, while at the same time expanding its scope to writing otherwise in a questionable manner. 

Note that he encloses ‘Chinese’ in single quotation marks here and prefacing script, instead of writing out the heretofore used terminology of "Chinese characters". This is related to buildup for a point he attempts to make later that is taken up below in conjunction with its subsequent development by Lurie.

Continuing on pp.205-7:

… There is a long history of such phonographic use of characters in Chinese-language environments… In Chinese translations of Buddhist sutras, such characters were even used to record entire incantations, or dharani, which provided a precedent for phonographic transcription beyond the limits of individual foreign words.

     Early Chinese treatises on the culture and society of neighboring ‘barbarians‘ transcribe terms and names in this manner, and those devoted to the Wa, like the Sanguozhi description of Himiko’s Yamatai, contain phonographic transcriptions that are commonly believed to be the first recorded forms of pre-Old Japanese.

Here, Lurie is forced to introduce some form of linguistic universals relating to phonetics; that is to say, to the representation of sounds by way of tentatively adopting certain logographic characters for their associated pronunciation as stand-in symbols for phonemes instead of meanings. But then Lurie seems to deny the Japanese the agency to arrive at the use of such matter-of-course adaptive techniques. 

On pp.205-7:

          This Chinese practice (as adapted by Korean 

        scribes) underlies the first phonographic writings of 

        vernacular terms within the Japanese archipelago. 

The fifth-century inscriptions produced by scribes from the Korean peninsula transcribe numerous proper nouns: for example, the genealogy on the Sakaitama-Inariyama sword spells out nine personal names and one toponym.

…The Kita-Otsu Glossary and Kannnoji ‘Dictionary’ mokkan both use them to specify the kundoku readings of particular characters paralleling practices with a long tradition in Chinese commentary and lexicography).

    In early mokkan there are some examples of the exclusive use of phonographs to record entire texts, although they are rare, and generally involve vernacular poems or fragments that seem to be from poems… The mokkan was found along with one bearing a cyclical date corresponding to 677, but similar items have been found from mid-century sites…

Ah, but here he asserts that the use of such practices, which for all intents and purposes would appear be a universal linguistic practices that had initially been developed by the Chinese to facilitate the teaching and learning of their own characters, in Japan should be attributed to those practices “as adapted” by Korean scribes, yet again, bypassing the Japanese altogether and not even entertaining a direct introduction to Japan by Chinese “scribes”, even though there is the example of the Chinese name inscribed on the 5^th sword discussed at length in earlier posts. 

If the inscription on that sword was done by a Chinese scribe, then there is more than enough evidence to justify the assumption that such practices were also introduced--perhaps earlier than by the teachers from Paekche—by scribes from China that had settled in Japan. There would appear to be no scholarly basis for Lurie’s irrational insistence on the point relating to Koreans. I should qualify that by stating that in all likelihood, the formal study of texts does appear to have been introduced by the teachers from Paekche, starting with the Confucian classics, and then Buddhist sutras and commentaries, as per the accounts of the Nihon shoki. 

I'm just making the point about Chinese scribes as there is evidence that suggests the continued presence of Chinese in Japan from the time of Himiko (see The Cambridge History of Japan, Ancient Japan, p. 289 in relation to Chang Chen), and the inscription on one of the swords discussed in the first post on Lurie's book would appear to have been by a Chinese immigrant on the basis of the signature of the scribe, whereas Lurie offers no evidence whatsoever that the inscription on the other sword was performed by a Korean scribe. Moreover, the Sakitama-Inariyama sword dates to a period during which there were diplomatic relations between the Chinese Song court and the five kings of Wa (see Lurie p. 96). If anything, the circumstantial evidence would seem to indicate that the scribes of these swords were Chinese immigrants. The Chinese may have been based at the commanderies in Korea, where they may have learned one of the Korean languages that would have enabled them to subsequently learn Japanese more easily due to the shared syntax.

Lurie covertly attempts to build upon earlier inadequacies in a piecemeal manner, which amounts, in the final analysis, to creating a Byzantine edifice of deception, leading the reader into a labyrinth of dead ends. Lurie’s insistence on this line would seem to be influenced by some bias I’ve yet to nail down, yet I suspect there is a political background to it related to the same agenda being promoted by the Western intelligence operatives active in disseminating disinformation here in Japan, and in Kyoto in particular.

At any rate, Lurie persists with this pathological line of argument, attempting to bolster by means of reiterating in different contexts his unsupported assertions that it was Korean scribes that produced the fifth century sword inscriptions--which include phonetic transcriptions of proper names, etc.

He moves along to the seventh century (see pp. 186-7), a period for which more than three examples have been excavated at different locations around the country that demonstrate both phonetic and kundoku transcription, along with some examples of subscript/superscript auxiliary phonetic annotation. All of the Kannnoji dictionary as well as other tablets, the Kita-Otsu glossary, and the poetry fragments are dated to the latter half of the 7^th century “or somewhat earlier” (p. 187).

One other point that should be raised here is that the occurrence of writing practice exercises and the like on mokkan indicate that the people using the mokkan were not fully (functionally) literate, but were learning to read and write using Chinese characters. That is to say, they were not “scribes”. In all likelihood, the people writing on the mokkan were low level administrators, carrying out basic clerical and office work, if you like, practicing their writing so that they could attain a higher degree of literacy, perhaps enabling them to perform higher level administrative tasks.

To be clear, as far as I can tell from the text of his book, the earliest example of a Korean mokkan that he can presents as demonstrating phonetic usage of logographic characters is from the eight century, as follows.

On p. 200:

…the 52 Anapchi mokkan, which were excavated from an eighth century pond on the ground of the palace at Kyongju. Bearing two Tang era names and several cyclical dates ranging from 751 to 774, they comprise mainly memoranda about weather and courtier’s schedules, but there are also some labels and at least one mokkan that employs phonographs.

I have quoted a passage from another book about the Anapchi mokkan, and Lurie’s lack of further discussion of the mokkan he mentions in the above-quoted passage is regrettable and conspicuous in light of the exaggerated assertions he makes in relation to this point.

On pp. 201-2:

…Parallels in format and content provide evidence of connections between mokkan-based communication in late sixth- and seventh-century Korea and similar practices that emerged in mid-seventh-century Japan. Among the most important aspects of these are phonograph usage in the Anpachi mokkan… (It is also important to note that the large late sixth-century find at the southern Songsansansong site considerably predates the earliest Japanese mokkan yet found).

First of all, the discrepancies in the dates of the relevant artifacts must be pointed out. Though he refers to "mokkan-based communication in late sixth- and seventh-century Korea", he presents no archaeological evidence from Korea for that period, and there is none that I am aware of to substantiate his assertions of the “evidence of connections” claimed in the above-quoted passage. Here, it must be emphasized that with respect to phonographs, the earliest example he even references from Korea is from the eight century (Anapchi mokkan), whereas he describes several examples of Japanese mokkan that employ phonographs and date to the seventh century (p.187, etc.).

The statement relating to “parallels in format and content” is not clarified with any examples and the terms themselves are too vague in this context. That prevents an accurate assessment of whether his assertion is viable, or to what extent the content thereof should be entertained for further consideration in relation to subsequent statements. 

His reference to “phonograph usage in the Anpachi mokkan” seems to expand the scope compared to his previous reference, but again he provides absolutely no discussion, let alone concrete analysis of the text to which one assumes he is referring. Moreover, based on his earlier statement on p. 200, it appears that he is referring to at most the text appearing on a single specimen excavated from the Anapchi site. I have not been able to find any correlative information on the use of phonographs on mokkan found at that site. In any case, Lurie's practice of continuing to attempt to inflate prior weak assertions makes it seem that he is simply building a house of cards.

Furthermore, with regard to the mokkan from the “the large late sixth-century find”, the mokkan from the late sixth century site consist only of mokkan used as commodity tags for labeling packages of provisions delivered to a military outpost, as discussed in an English language version (2010) of an essay Japanese (2006) in the book Lurie fails even to cite (described above). Moreover, there is nothing of significance in the text found on those mokkan with respect to the points of kundoku and phonetics. Lurie seems to mention them simply because of the early date, attempting to associate them with practices for which there is no solid evidence whatsoever that can be dated earlier than the artifacts from Japan.

At this juncture, because I have found in the paper by Frellesvig that I just discovered some pertinent information to the question of the influence of Korean reading and writing practices on the development of the Japanese writing system, I’m going to interject that here. I’m doing this partly to demonstrate that I am not being harshly critical of Lurie because of some conspiracy theory driven paranoia, but because his text contains many assertions that are not supported and seem questionable. This passage starts at the bottom of page 4 and runs to the middle of page 5 of Frellesvig’s paper, as follows:

Chinese script may be adapted to write other languages either logographically or phonographically. Pre-alphabetic writing in Korean comprises the following three main types: (a) pure logographic writing, with sinographs used for lexical words, but with no indication of grammatical particles or morphology (the so-called sekichey ‘Gelobnisschrift’). (b) Logographic writing, with conventionalized logographic writings for grammatical elements (itwu ‘clerk readings’). (c) Logographically written lexical items supplemented by phonographically written grammatical elements (hyangchal). These types are all found in the Old Japanese text corpus. It is likely that the correspondences in specific types of writing on the continent and in Japan reflect a common continental source rather than parallel developments and there is therefore little sense in trying to reconstruct an independent course of evolution of adapting Chinese script to write Japanese. It is possible, however, to identity two writing practices which are not documented in Korean sources and which seem to be independent developments which took place in Japan: (a) writing extensive text passages entirely or mostly phonographically, reflected in the widespread use of man’yogana; (b) distinguishing orthographically between lexical and grammatical elements, as in senmyogaki.

I was somewhat relieved upon reading through this passage to see statements toward the end supporting the somewhat hypothetical observations I’ve made relating to the use of phonographs with respect to both phonetics and morphological transformations of verb endings accompanying inflexion. I was, however, somewhat dismayed at the apparent dismissal of the examination of parallel development, as this seems unwarranted. 

To illustrate why that is the case I will turn toward the discussion of hyangchal in Lurie’s book to demonstrate that the two types [(a) itwu and (b) hyangchal in the above-quoted passage] of kundoku-influenced writing that came into use in Korea did not all necessarily occur first in Korea,nor is there any concrete evidence that they were transmitted to Japan, whereas the basic tools for developing such forms of writing had already existed in Japan.

On p. 201, down the page from the above-quoted passage, Lurie writes:

    … To this can be added practices attested in later periods of Korean history that create a strong impression that scribes and scholars of the Three Kingdoms had developed a variety of kundoku-based systems to read and write early Korean languages. Perhaps the most famous atter-day method of writing is hyangch’al, a system used to record 25 Silla poems (hyangga) contained in two Koryo-period works (935-1392). Hyangch’al involves the combination of characters used logographically (for kun [Korean hun] readings of words) and phonographically (spelling out syllables using the sounds of both the Chinese and the Silla words with which the characters were associated). There are striking Japanese parallels in the inscription of the poetry collected in the eighth-century Man’yoshu.

Though I won’t examine them in depth, note the apparent discrepancies in the dates described above with respect to hyangga poems and Man'yoshu poems. Because Lurie clarifies that discrepancy to a certain extent in an endnote, I’m going to quote the endnote associated with the above-quoted passage, in which he admits that there is in fact no connection between the “striking Japanese parallels” to the inscribed form of the Korean hyangga poems and the method of inscription itself. In fact, it would appear to be more likely that the Japanese inscription of the Man'yoshu poems preceded the Korean inscription of the hyangga.

Note that his characterization of hyangch'al as a method of writing is linguistically correct, whereas the characterization of kundoku per se as as a method of writing is not. The corresponding Japanese method of writing is hentai kanbun in this case. It would not be incorrect to characterize both of said writing systems as "kundoku-based", but that would raise the question of the status of the use of phonographs, once again. 

On p. 396, Lurie writes:

    The earliest known hyangga is from 600, though most of them date from the eighth century; the texts themselves are thought to have been transcribed between the late seventh and late ninth centuries… A number of scholars have examined the parallels between this system and the mixtures of logographs and phonographs used in seventh- and eighth-century Japan… However, as there are significant differences in the phonographs used and in the way that they are combined with logographs, it is difficult to postulate a direct relationship… It is more appropriate to assume that both descended from even earlier practices of Three Kingdoms Korea that are unattested in the historical record (which seems likely to me) or that they are independent inventions (which is certainly possible).

According to Lurie's endnote, the hyangga poems are said to be "thought to have been transcribed between the late seventh and late ninth century, which would tend to indicate that they were being written down (transcribed) during a period that overlaps the release of the Man'yoshu compilation and continues for another 100 years. It would therefore seem to be likely that the inscription of most hyangga occurred later than the inscription of the most of the poems in the Man'yashu compilation, which is said to be based on earlier compilations that have been lost, and includes many different logograph-phonograph poetic formats that can be seen to indicate evolutionary stages of inscription. 

From the above-quoted passages by Frellesvig and Lurie, it is apparent that the questions are involved and require more thorough study. It is noteworthy that here Lurie at least admits that the modes of inscription found among the poems of the Man'yoshu may have been independent inventions.

On p. 203 Lurie starts a new section, entitled,

KOREA, JAPAN, AND THE EAST ASIAN WRITING SYSTEM

and begins with the following sentence:

There is both direct and circumstantial evidence that all of the techniques that revolutionized Japanese inscription in the latter half of the seventh century were grounded in practices brought by scribes who arrived from the Korean peninsula in the sixth and seventh centuries.

I repeatedly quote such passages because Lurie incessantly tries to drum this line about Korean scribes through the thick skulls of the readers of his Harvard published diatribe, in spite of the occasional appearance in an endnote of a more balanced approach to scholarship that might have made the book more worthy were it consistently embodied in the main text.

I will have to unpack the above-mentioned quote in conjunction with the following passage, however, because the rhetoric is not merely the blunt instrument it may seem. The above-quoted passage has even more in store for the unsuspecting reader. 

On p. 346, Lurie writes:

… As argued above, this commonly held medium did not necessarily conflict with local forms of expression or modes of communication (whether written or oral). Especially given the prominent role of kundoku in Japan and Korea, and signs that similar reading methods were used elsewhere also, we should not overemphasize the ‘Chineseness’ of the Chinese writing system.

The above-quoted passage relates to a number of threads of argument that have been interwoven through the course of the book, and which I will address severally to a reasonable extent below.

In this passage, which again is toward the very end of the book, he admits that kundoku, too, is not unique to Korea and Japan, which means that there is a similar universal aspect to kundoku among people originally speaking languages other than Chinese as found with respect to the practice of using logographic characters as stand-ins to represent pronunciations, which was used even among the Chinese to study their own logographic characters. 

Although I've made this point earlier in terms of the science of linguistics (insofar as I can assume the mantle), I cite the above passage from p. 346 here in order to examine rhetorical aspects of Lurie's text. 

The above-quoted passage would certainly seem to problematize the incessant insistence with which Lurie asserts repeatedly that kundoku practices as developed in Korea served as the basis for the writing system that was subsequently developed in Japan.  

There are two folds to the rhetoric with respect to this point. 

First, Lurie posits in hyper dramatic form the existence of "techniques that revolutionized Japanese inscription", yet I don't see that he sets forth a single example of such a specifically revolutionary technique. That might lead the reader to think that s/he has missed something, causing them to backtrack and retroactively substantiate the closest thing they can approximate as a revolutionary technique that would otherwise have been overlooked (by a reader less than hip to the revolutionary).

To rephrases that, the unsuspecting reader is surreptitiously redirected backward through Lurie’s text on a quest to find the missing revolutionary meaning that the careless reader has overlooked. In fact, however, Lurie has never said anything concrete, so the missing meaning that the reader thinks they find is merely the product of the desire of the reader to assign meaning to something insubstantial that Lurie has said earlier in relation to the topic, whereby the phantom meaning is retroactively instantiated in the memory of the reader but doesn’t actually exist in the text. Maybe this is a rhetorical device that could be described in terms of the oft heard phrase "the power of suggestion".

Second, since he has not set forth a single specific example of such a revolutionary technique, the claim in relation to “direct and circumstantial evidence” is used to preface the assertion of the existence of such techniques in a manner such as to obfuscate the fact that he hasn't actually introduced any, though he repeatedly states that scribes from the Korean peninsula brought them over. And this is what the reader back tracks in search of through the text, to then retroactively gestalt the phantom technique(s). 

Admittedly, this is a bit of a speculative construction on my part, as I don’t have a PhD in rhetoric or psycholinguistics. On the other hand, the account I’ve provided reflects my personal experience with the text of this book to a certain extent. I simply happened to grasp that there was nothing of substance being said at either end of some of these wild goose chase arguments, and then followed up on several of the texts Lurie refers to in his book, whereupon I was somewhat astonished by what I found—or didn’t find.

While Lurie fails to even offer any direct evidence for establishing a relation between Korean kundoku and Japanese kundoku, it doesn't really seem to matter in the final analysis, because kundoku is a set of linguistic learning tools that would in all likelihood be developed universally across cultures as a matter of course by the people of any country trying to adopt Chinese characters to write (and read) their native language. 

In this regard, it would certainly appear plausible to characterize kundoku as relating to a universal set of problems pertaining to language acquisition as applied to writing with logographs originating in a different language. With respect to solving such problems, the fact that there are commonalities in approaches developed in different places at different times doesn't necessarily rule out the historical occurrence of "techniques that revolutionized..."; however, with respect to the development of writing in Japan, I repeat, Lurie doesn't characterize a single technique as having a revolutionary impact, while he repeatedly asserts that a plurality of such techniques were introduced to Japan by Korean scribes.

Lurie mentions that Korea was the first of the states on China's periphery to adopt Chinese characters, so it may be that Koreans developed kundoku practices early. And since the Nihon shoki documents that initially a Korean teacher of Confucian texts was sent to the Yamato court and served as teacher to a prince, and that other even more prominent Korean teachers were subsequently dispatched to the Yamato court to serve in similar roles, it is highly likely that they did employ kundoku practices in the course of teaching their Japanese students how to read the meaning and the sound and write Chinese characters and sentences. And since the Korean language shares a similar syntax to Japanese, there would have been a common problem set encountered by Japanese and Koreans in the course of adopting Chinese characters as the building blocks for a writing system. It is therefore to be expected that any solutions which they respectively came up with independently might also have some overlapping aspects.

The question regarding the aforementioned reading of the ‘sound’ is what is of importance here. The Chinese characters were undoubtedly read in approximations of their Chinese pronunciations at first, which later became the onyomi Sino-Japanese pronunciations. 

It might be the case that kunyomi readings developed as a result of the process of learning the meanings of the characters by writing down descriptions of the meaning(s) in vernacular language expressed by means of logographs that had been adopted as stand-in phonographs, thereby creating dictionaries and glossaries representing Chinese meanings in spoken Japanese. Those descriptions of meanings spelled out phonetically could later be adapted as singular kunyomi readings of a character in a particular usage. Many Chinese characters have been assigned more than one kunyomi reading, with some characters having more than five separate kunyomi readings, for example.

The relationship of specific techniques purported to have been introduced by Lurie to the trajectory of the development of literacy and a writing system in Japan is not only unclear (as he defines none and there would appear to be no concrete evidence upon which to postulate any reliably), but apparently irrelevant. That is to say, what is of significance is the fact that the Japanese (and Koreans, and Vietnamese, etc.) had adapted techniques and practices for representing linguistic aspects of their native languages in conjunction with the adoption of the logographic characters of China on the way to developing a writing system. Such techniques and practices facilitated the development of a writing system tailored to the indigenous culture.

The techniques and practices comprised by kundoku as per Lurie's definition included the use of stand-in logographs to represent phonemes in the native tongue instead of their associated semantic content, changing the character ordering from that of Chinese SVO syntax to the Japanese/Korean syntax of SOV. Here, I would also include he use of reduced-size sub/superscripts to indicate the proper kundoku reading by illustrating the inflected reading of a verb, for example, which Freelesvig indicates is a practice not documented in Korean sources. 

It is noteworthy here that the use of logographs as stand-in phonographs is a specific technique that even Lurie belatedly admits was a "Chinese practice", though he then tries ameliorate the implications of that admission by qualifying the practices used in Japan as having been “adapted by Korean scribes”. Meanwhile, he later admits that other countries on China’s periphery (Vietnam, etc.) had also adopted similar practices, apparently independently.

As for the Japanese writing system, from the available evidence, it seems highly probable that most of what might be said to qualify as actual innovation occurred in Japan. I haven't seen mention of any cases in early Korean artifacts demonstrating use of an auxiliary phonetic subscript or superscript to describe the pronunciation of characters in a text, which is found on Japanese artifacts dating to the seventh century, which can be seen to represent a precursor to senmyo-gaki. In any case, the development of writing and the spread of literacy in Japan vastly outstripped that of the Korean peninsula from a comparatively early stage, resulting in a full-blown system by the ninth century, and perhaps the highest literacy rate in the world during the Edo period. It might be the case that the people on the Korean peninsula had a head start, but were overcome by the people on the Japanese archipelago from a relatively early point in history.

Here, it is noteworthy that modern Korean doesn't use kunyomi readings of Chinese characters as found in Japanese. Since I haven't studied ancient Korean texts, I don't know what the particulars of kundoku as reflected in Korean inscriptions from say the 10th century might have to teach us in regard to the evolution of Korean vis-a-vis the evolution of Japanese. However, I had studied Korean to a fairly high level in the 1980s. At that point in time, the Koreans still used Chinese characters more widely than they do today. But the Koreans only used Chinese characters in compounds, read with the onyomi reading; that is to say, the Korean use of Chinese characters was limited to compounds of two or more characters read in a Sino-Korean pronunciation. In other words, modern Korean does not use individual Chinese characters to represent native Korean language. The Japanese use of single Chinese characters to represent native Japanese language using the kunyomi system is one of the main features that differentiate modern Japanese and modern Korean, and is perhaps a linguistic feature that is unique to the Japanese writing system at this point in time.  

In fact, this would be a serious question to pursue in terms of the history of linguistics in both countries, and the history of writing in East Asia. Let me clarify that. In Japan, the archaic practice of kundoku reading of Chinese character texts was displaced by the middle of the ninth century for all intents and purposes, and what primarily can be said to have been inherited from kundoku by the mature Japanese writing system that emerged from the late eight century incorporating the hiragana and katakana syllabaries was kunyomi readings of Chinese characters. Korean, on the other hand, is a blank to me for the period in question, but there are no kunyomi readings in modern Korean. 

I would go as far to suggest that the failure to address this question represents a gaping hole in Lurie's book. And part of the credit for that error can be probably be assigned to the editorial staff at Harvard for failing to grasp or point out the issue. Apparently they don't have people that have studied both of these languages to a degree sufficient to have enabled them to hit on this potentially very significant point. 

By conducting a comparative examination of this point between Japanese and Korean usage of Chinese characters, it would be possible to determine the extent of the influence of kundoku on the development of both respective writing systems at various stages, and elucidate when the two languages diverged in relation to the manner in which they incorporated Chinese characters. There is a gap of approximately six hundred years between the development of a full-blown writing system in Japan and the development of a corresponding system in Korea, and it is only in very recent history that Korea has undertaken serious efforts to eliminate the use of Chinese characters from everyday writing.

It is not clear what relevance Lurie and his comrades in pens from Columbia University are trying to assert with respect to the Korean role in the development of Japanese culture, but though this analysis of Lurie's presentation of kundoku, I believe I may have cleared a bit of the way towards achieving a better understanding of their agenda. I think I have demonstrated that Lurie's extremely overemphasized claims regarding kundoku in relation to the assertion that Japan's development of a writing system was dependent upon Korea are fallacious, and that is part of the basis upon which I think it is fair to postulate that there is an agenda shared by the above-mentioned individuals who were apparently all mentored by Faure. 

There is yet another problematic aspect lurking in the rhetoric of the above-quoted passage. Lurie characterizes Chinese characters in terms of being a "commonly held medium" in light of the fact that they had been adopted as a writing system by numerous countries on China's periphery. He then asserts that because they enabled the people in whatever locale to express their native tongue, that they didn't "necessarily conflict with local forms of expression... (...written and oral)", in relation to which he cites the widespread adoption of kundoku practices.  In effect, he covertly asserts that because kundoku demonstrates that Chinese characters have the flexibility to facilitate their adaptation for use as a writing system by people whose native tongue is other than Chinese, that we shouldn't "overemphasize the 'Chineseness' of the Chinese writing system". This basically inverts the reality that kundoku practices were adopted because of the difference between the syntax of the Chinese and Japanese/Korean languages, and perhaps because phonetics and semantics needed to be mapped out, so to speak, using to the Chinese characters to facilitate their adoption for writing the "local forms of expression... (...written and oral)". Here, it is noteworthy that Lurie has included "oral" forms of written expression under the rubric of kundoku, whereas he has earlier attempted to characterize the use of phonographs as a separate "method of writing" from kundoku. Perhaps this is another area of unclear overlap, as there is his assertion of the "invisibility" of kundoku. I would simply again refer the examination of this question to a more thorough study of the relation of kundoku to the evolution of senmyo-gaki and beyond to the kana-based wabun Japanese writing system. It is obvious that the use of phonetics to illustrate aspects related to kundoku reading materialized at a fairly early stage (by the 7th century), so the characterization of kundoku as "invisible" is questionable on at least that front.

At any rate, kundoku reading practices had to be adapted because of the particular qualities of the Chinese characters insofar as they were logographic, not phonetic. And since Chinese characters are the only surviving form of a logographic script currently in use, I think that we can focus on their uniqueness as a logographic system with at least six types of logical schemas for forming logographs. In this respect, I can't imagine where Lurie gets the notion of "overemphasizing" the ‘Chineseness’ of Chinese characters, because the obviousness that they are Chinese is a prima fascia given. With the Roman alphabet, for example, such a question does not arise because it is a straight forward phonetic system that is more readily adaptable, as can be seen even with modern Vietnamese.

While Lurie refers to Chinese characters as "commonly held medium", he seems to covertly inflate the earlier assertion of the status of kundoku as a method of writing in the above-quoted passage, wherein he associates its use as a reading method for non-Chinese speakers that is applied to texts written in the commonly held medium of Chinese characters, in a manner such as to call into question "the 'Chineseness' of Chinese writing system". This would seem to be the other part of the reason that he tries to assert that kundoku is a method of writing, as it ties into his covert attack against Chinese characters and Chinese culture with respect to the East Asian cultural sphere. 

I take this to be an  important observation, and take this question up in the next most in greater detail. Suffice it to say that kundoku is a method of reading that influenced the development of modes of inscription that resulted in the creation of new, but related writing systems. That is to say, the 'Chineseness' of the Chinese writing system is not adversely affected by the advent of the new but related Japanese writing system, for example, of which many aspects are directly or indirectly derived from Chinese, which in turn does not detract from the 'Japaneseness' of the Japanese writing system.  

I would think that a more scientific discussion relating to the linguistics of language acquisition and writing is possible, with the Chinese character-based writing system being contrasted in terms of its being a logographic system in a comparison to phonetic systems with respect to the question of adaptation.

The following passage is from a paper by Tokio Shinkawa, entitled “Culture and Ideas Carried by Chinese Characters in Ancient East Asia, the Japanese Viewpoint” from the book Centrality and Marginality of Ancient Documents. The points made by Shinkawa in relation to pronunciation are relative to the discussion of kundoku.

On pp.67-9 Shinkawa writes:

    The site as a whole, and thus the wooden tablets, dates approximately to the late seventh and early eighth centuries AD, largely to the reign of emperor Tenmu. It consists of the ruins of the Asukaji temple and its annex, south-eastern Zen-In. The temple is known to be one of the oldest in Japan, and its sanctuary the most extensive, while the annex was founded by Do-sho (629-700) and other scholars who returned to Japan after completing their studies in China under Gen-jo (?-664), one of the most famous scholar-monks of the Tang dynasty. Thus we may expect to find a strong Chinese cultural influence on the Asian periphery where these temples were located.

    I begin with three passages written in sumi, ‘Indian’ ink, on three faces of four of a long wooden block.

    1) is obviously a reference to the Kanzeon-bosatsu sutra, the chapter of the Buddhist Goddess of Mercy within the Lotus sutra, one of Buddhism’s most important and popular texts. 2) is difficult to interpret, but may be based on a phrase typical of Buddhist sutras… 3) contains a phrase often used in the Analects, the most important Confucian text. …we think they belong to writing exercises in Chinese characters, but it should also be noted that the writer did not hesitate to inscribe the name of a Buddhist sutra and a common Confucian phrase on the same wooden material.

And on pp. 74-5:

    Let us return to the Asukaji-Ike tablets. They contain many medical allusions, for example…

   

Text 3) is a note of how to pronounce a series of medical terms, the first word [Chinese character] being pronounced as the next [two Chinese characters] ‘ha-i’… This indicates the process of learning how to pronounce Chinese characters, in other words, how the Japanese first responded to the use of unfamiliar Chinese characters…