1 “Riben dao ge”: Its Inclusion in Ouyang Xiu’s Collected Works and Reason for Creation

For the sake of discussion, I will quote the “Riben dao ge” here:

In the Ming dynasty (1368–1644), it was found that the poem had already been reproduced in the collected works of Sima Guang 司馬光 (1019–1086). Ouyang Xiu’s 歐陽修 (1007–1072) authorship was also called into question during the Qing dynasty (1644–1911), but the first person to clearly articulate the view that the poem was written by Sima Guang was a Japanese scholar (more anon). The poem is a work whose authorship is disputed, yet because it has been included in the Waiji 外集 of Ouyang Xiu’s collected works for such a long time, and its influence has been enormous, the Ouyang Xiu text is the main focus here.

Descriptions of or songs about swords are a long-standing tradition in China. These include Tongjian zan 銅劍贊 by Jiang Yan 江淹 (444–505) and Gujian pian 古劍篇 by Guo Zhen 郭震 (656–713). Yet the uniqueness of the “Riben dao ge” lies in how the poet, through the treasured sword, expresses a yearning for its land of origin, Japan, and heaps praise on that country with such lines as “its soil is fertile and its customs fine” and “its scholar-officials are skilled in poetry and rhetoric.” Yet, of even greater importance to the author is that “the Qin dynasty had not yet burned the books when Xu Fu set out (for Japan), / so the one hundred chapters missing from China are still preserved there.” This refers to the one hundred chapters of the Shangshu, which Xu Fu 徐福 (b. 255 BCE) had taken to Japan and placed there for safekeeping before the Qin Shi Huang 秦皇 (r. 221–210 BCE) ordered the burning of books. One cannot help but shed grateful tears when thinking that “the sea is vast and impenetrable.” Contemporaneous with Sima Guang, Mei Yaochen 梅堯臣 (1002–1060) wrote “Qian Junyi xueshi Riben dao” 錢君倚學士日本刀, a poem praising the elegance of Japanese swords as being akin to that of the Tai E 太阿 sword forged by Gan Jiang 干將. The substance of Mei’s poem is its expression of the view that “the ancients were experts in both civilian and military affairs.”⁴ Even though the two poems speak of the same type of object, they differ greatly in intent. According to Zhu Dongrun 朱東潤, Mei’s poem was composed in the third year of the Jiayou 嘉祐 period (1058); the “Riben dao ge” is considered to have been composed contemporaneously.

The “Riben dao ge” is included in the extant Waiji of Ouyang Xiu’s collected works, and was included in an edition collated by Zhou Bida 周必大 (1126–1204). By the middle of the Northern Song dynasty (960–1127) at the latest, the poem was already considered to be Ouyang Xiu’s work. This information was recorded in scroll forty of the Yinchuang zalu 吟窗雜錄. It has not been cited by scholars previously, so it is provided here:

Minister Lü asked, “It is said that Koryŏ holds the one hundred chapters of the Shangshu. Is it true?” None of the guests dared to reply. Wang Ledao 王樂道, who happened to be seated at the end, then answered, “I’m afraid so.” Lü asked, “How do you know?” Wang said, “Ouyang Xiu wrote in his poem that ‘the Qin dynasty had not yet burned the books when Xu Fu set out (for Japan), / so the one hundred chapters missing from China are still preserved there.’ He must have had a reason for writing that.”⁵

The fifty scrolls of the present Yinchuang zalu were completed in the fifth year of the Shaoxi 紹熙 period (1194). The information recorded therein was extracted before the Song court fled south of the Yangtze River,⁶ so the content cited above must have been written in the Northern Song dynasty. In terms of the style of the content, the text seems to have been taken from some remarks on poetry or note-form stories. Unfortunately, the original work on which it is based has been lost, and there is no means of re-examining it. Since Zhou Bida’s collated edition of Ouyang Xiu’s collected works was first published in the second year of the Qingyuan 慶元 period (1196), it did not form the basis for the Yinchuang zalu. Different editions of Ouyang Xiu’s collected works “were widely distributed, but there was no reliable edition”⁷ ; the poem must have been included in some Bieji or Waiji of Ouyang Xiu’s collected works before the Zhou Bida edition was published.

Minister of Lü refers to Lü Gongzhu 呂公著 (1018–1089). According to the Songshi 宋史, he “resigned his post, and was appointed Minister of Works with the rank of Minister of Military Affairs in the fourth month of the third year (of the Yuanyou 元祐 period).”⁸ Wang Ledao compiled the twenty scrolls of Ouyang Xiu’s Bieji with assistance from his son Wang Xingzhi 王性之 (dates unknown). The Bieji was completed in the fourth year of the Zhenghe 政和 period (1114), and “none of it was included in the family collection.”⁹ This Bieji constitutes the main part of the present-day Waiji (twenty-five scrolls). If the record quoted in the Yinchuang zalu is true, the “Riben dao ge” was regarded by contemporaries as Ouyang Xiu’s work not long after his death. It was included in Ouyang’s collected works probably by Wang Ledao (or his son Wang Xingzhi). This is of course a relatively early and important piece of evidence. Nevertheless, we are still unable to assign the authorship of this poem to Ouyang Xiu; as explained below, it should be considered Sima Guang’s work in light of various sources.

It is noteworthy that although Wang Ledao quoted from the “Riben dao ge,” his remark referred to how Koryŏ was preserving the missing chapters. This brings us to the historical context in which people during the Song dynasty sought to recover books from overseas.

The pursuit of books located in Koryŏ and Japan had already been undertaken on a large scale by the state of Wuyue (907–978) during the Five Dynasties period (907–960), yet all books concerned were Buddhist scriptures. In the autumn of the sixth year of the Xiande 顯德 period (959) of the Later Zhou (951–960), Koryŏ dispatched envoys to present one scroll of the Biexu Xiaojing 別序孝經, eight scrolls of the Yuewang Xiaojing xinyi 越王孝經新義, one scroll of the Huangling Xiaojing 皇靈孝經, and one scroll of the Xiaojing cixiong tu 孝經雌雄圖, which can be found in scroll two of the Koryŏsa 高麗史 (“Shijia Guangzong” 世家光宗). This was the first time that classical books from foreign countries had been offered to China. In the eighth year of the Taiping Xingguo 太平興國 period (983), the Japanese monk Chōnen 奝然 (938–1016) came to China and presented the Xiaojing 孝經 annotated by Zheng Xuan 鄭玄 (127–ca. 200) and fifteen scrolls of the Yuewang Xiaojing xinyi 越王孝經新義, books missing from China (see the “Riben zhuan” 日本傳 of the Song shi). In all likelihood, Chōnen’s presentation of the Xiaojing, a cultural event, brought great excitement to the Song dynasty sovereign and scholars at court and elsewhere. It is perhaps for this reason that the Song people had high expectations for the existence of similar texts, such as the full version of the Shangshu, in collections abroad. Aside from the “Riben dao ge,” written in the third year of the Jiayou period (1058), and the rumors, as described in the preceding, heard by minister Lü (circa 1074), we may also cite the catalog of book requests in the eighth year of the reign of King Sŏnjong 宣宗 of Koryŏ (that is, the sixth year of Song Zhezong’s 宋哲宗 Yuanyou period [1091]), as contained in scroll ten of the Koryŏsa (“Shijia Xuanzong” 世家宣宗). This catalog records 128 books, the first of which is the “hundred chapters of the Shang Shu.” According to scroll fifty-two of the Yu hai 玉海, “on the nineteenth day of the fifth month of the seventh year of the Yuanyou period, the Palace Library (mishu sheng 秘書省) said that the editions of the books presented by Koryŏ differ from those of our country; they are not seen on our shelves. An imperial edict stated that two of the books were to be collated and corrected, then copied and deposited in the Taiqing lou 太清樓 and Tianzhang ge 天章閣.”¹⁰ In scroll five it is recorded that “in the eighth year of the Yuanyou period, Koryŏ presented ten scrolls of the Jingshi Zhouyi zhan 京氏周易佔, which may be from the twelve scrolls of the Zhouyi zhan 周易佔 recorded in the ‘Jingji zhi’ of Suishu.”¹¹ It may be seen that Koryŏ truly did supply quite a large number of different versions at the request of the Song dynasty, some of which have been lost in China.

Why would Sima Guang have written about a Japanese sword from the perspective of missing texts? I believe that this is related to the positions he held. Consider that in the second year of the Yuanyou period, Sima Guang “obtained a new appointment. He switched from the role of academician at the Court of Imperial Sacrifices (taichang boshi 太常博士) to a role at the Palace Library (mige 秘閣).”¹² The Palace Library was a Song dynasty institution for the storage of books; when Sima Guang held his post there, he naturally had a certain understanding of the classic texts offered by overseas donors, and it was only natural that he should have written the poem from this perspective. Furthermore, early in the Jiayou period, Qian Gongfu 錢公輔 (1021–1072), the holder of the Japanese sword, transferred from the role of local magistrate in Yuezhou 越州 to that of editor at the Jixian yuan 集賢院. This position is referred to as “Tianlu” 天祿 in the line “after returning from Tianlu, he showed the sword to his friends” in Mei Yaochen’s poem.¹³ “Tianlu” is a place where books are collected and collated. Although Qian held this position, he was only interested in the sword acquired by his predecessor, which may not have been appreciated by Sima Guang. Therefore, at the end of the “Riben dao ge” comes the line, “What does this dull, rusty blade offer in comparison?” This is not insignificant. We know that in the Song dynasty, Yuezhou was an important port for trade with Japan and Koryŏ and many artifacts and books were imported from there. Sima Guang could not have been unaware of this. It was clear to him that the sword was not worth comparing to texts such as “the decrees of the sage rulers of the past.” To a certain extent, the poem conceals Sima Guang’s mockery of Qian Gongfu. In one respect, the poem reflects the former’s culture-based values.

2 On the Authorship of “Riben dao ge” and Late-Qing Sino-Japanese Intellectual Exchange

Since the 1980s, the primary discussions on the authorship of the “Riben dao ge” have been “‘Riben dao ge’ zuozhe xinkao” 《日本刀歌》作者新考 by Tan Bi’an 譚彼岸, “‘Riben dao ge’ yu hanji huiliu” 《日本刀歌》與漢籍回流 by Wang Shuizhao 王水照, and “Guanyu ‘Riben dao ge’ zuozhe jiqi yingxiang de kaozheng” 關於《日本刀歌》作者及其影響的考證 by Ning Qundi 寧群娣. Tan believes that Qian Gongfu was the author, but this is patently false. Wang indicates that Sima Guang was the author, based on the following three reasons. First, the poem is included in several extant collections of Sima Guang’s poems and essays. Yet the titles, of which there are two, are slightly different: “Junyi Riben dao ge” 君倚日本刀歌 and “He Qian Junyi xueshi Riben dao ge” 和錢君倚學士日本刀歌. Next, considering the relationship between Sima Guang, Mei Yaochen, and Qian Gongfu, it is far more likely that Sima Guang wrote the poem than Ouyang Xiu. Further, in terms of editions, the poem was included in Ouyang Xiu’s Jushi waiji 居士外集, but the Waiji is a compilation supplemented by others, and scholars have long pointed out that it contained forgeries. The poem was also included in numerous editions of Sima Guang’s collected writings, among them Sima wenzheng gong chuanjia ji 司馬文正公傳家集, compiled by Sima Guang himself. The reliability of this edition is far greater than that of the Waiji. Ning developed his article primarily based on Wang’s work, and enters into a slightly more detailed discussion.

The first Chinese person to advance the Sima Guang theory was Yang Shoujing 楊守敬 (1839–1915). His handwritten “Riben dao ge” scroll (now in a private collection in Japan) bears the following caption:

On the right is Sima Guang’s “He Qian Junyi Riben dao ge,” included in scroll forty-seven of the Wenzheng gong ji. Yet Ouyang Xiu is also said to have written this poem, so few people knew that it was written by Sima Guang. Examining the Wenzheng gong ji, there is also the “He Qian Junyi tengchuang shi’er shou” 和錢君倚藤床十二首, which indicates that Qian Junyi has always exchanged poems with Sima Guang. The Ouyang Xiu version of the poem is not labeled with the name of Qian, and there are many typographical errors. Thus, there is no doubt that the poem was composed by Sima Guang. I wrote this passage in the first month of the second year of the Xuantong 宣統 period to correct a past mistake. Yang Shoujing of Yidu 宜都, aged 72.

This account is abridged. Yet based on Sima Guang’s collected works, his circle of friends, and the erroneous characters in the Ouyang Xiu edition, Yang concluded that “there is no doubt that the poem was composed by Sima Guang.” In 1880, Yang was sent on a diplomatic mission to Japan. He spent four years there searching for books, and his trip was a success. In his “Riben fangshuzhi yuanqi” 日本訪書志緣起, Yang said: “In the Gengchen 庚辰 year (1880), I went to Japan. When I thought upon Ouyang Xiu’s words that the hundred chapters were still preserved there, it became my ambition to collect lost books … Although none of the books I found were among those that had survived the book burning of the Qin dynasty, some of them were books that Chōnen did not present to the Song dynasty.”¹⁴ Apparently, he took the poem as Ouyang Xiu’s work before changing his opinion.

Yang’s attribution of the poem to Sima Guang is in fact derived from the Japanese Meiji-era (1868–1912) scholar Kusaka Hiroshi 日下寬 (1852–1926). Kusaka was a historian and sinologist also known by his literary name Shakusui 勺水, and he belonged to department B of bureau two of the Office of Historiography (Shūshi-kyoku 修史局), a pioneer of Japanese historical research. His authored works include Hō kō ibun 豐公遺文, Hōkō nenpu 豐公年譜, and Roku yū sō bunshū 鹿友莊文集. In the third month of the sixteenth year of the Meiji period (1883), Kusaka’s essay “Sho Nihontō ka” 書日本刀歌 was published by Seishō-sha in volume 82 of Kokon gogon kaidoku 古今語言解讀. He writes:

It is said that the “Riben dao ge,” transcribed on the right, is the work of the Song scholar Ouyang Xiu. Other people say Sima Guang also wrote a poem with the same title, but it is the same poem, I’m afraid. We probably owe this idea to Kaibara Ekiken 貝原益軒. In Kakubutsu yowa 格物餘話, Kaibara compiled a broad list of the similarities and differences among the words and expressions used in the two pieces, but did not mention which are right and which are wrong. By chance, editor Hoshino was leafing through the Chuanjia ji 傳家集 and showed it to me. We compared it with the poem recorded in the Wenzhong waiji 文忠外集. It’s true what Kaibara said: it’s exactly the same work.

Thus, books such as Songshi chao 宋詩鈔 and Yongwu shi xuan 詠物詩選 all say that it is Ouyang Xiu’s poem. However, the Chuanjia ji collected in the Song baijia shi cun 宋百家詩存 does not include the poem. Is it really not a work by Sima Guang? Consider that the Chuanjia ji is a collection compiled by Sima Guang himself, while the Wenzhong waiji was compiled by later generations. Furthermore, Qian Junyi lived during the same period as Sima Guang, but a little later than Ouyang Xiu. Now, we cannot see the original work that this poem responds to, but in the Chuanjia ji we can often see the matching poems. Comparing the poem in two editions, and referring to this one, I can say it is definitely not the work of Ouyang Xiu.

Alas! The poem is so popular, but over the past centuries no one has pointed out that it was written by Sima Guang. The Yongwu shi xuan 詠物詩選 is an anthology that was compiled by a group of scholars under an order from the emperor Kangxi 康熙 of the Qing dynasty. But they attributed the poem to Ouyang Xiu; how serious a matter it is to pass something off for what it is not! Dongpo 東坡 [Su Shi 蘇軾] said that it’s a pity that the poetry and literature of Li Taibai 李太白 [Li Bai 李白], Han Tuizhi 韓退之 [Han Yu 韓愈], and Bai Letian 白樂天 [Bai Juyi 白居易] were being mispresented by common people. We might well ask whether this is the only poem whose origin is contested among all the works of dubious authenticity from the past and the present! I am righting a wrong on behalf of Sima Guang, and eliminating fraudulent claimants and impostors on behalf of Ouyang Xiu. Sima Guang’s seven-syllable lines in ancient style have an enduring charm and a mellow and generous character, which is poles apart from the plainness and simplicity of Ouyang Xiu’s poems. Anyone with the ability to ponder this will naturally understand their difference.

In scroll four of Kusaka’s work Roku yū sō bunshū, the following is included, with a remark from Yang Shoujing:

Yang Xingwu 楊惺吾 [Yang Shoujing 楊守敬] said: No one ever considered the poem not to be the work of Ouyang Xiu. The Chuanjia ji is not an uncommon book. After Shakusui [Kusaka] pointed this out, we felt ashamed.

Yang returned to China in 1884. It is thought that Kusaka Hiroshi sought Yang’s advice on his essays one year prior to this. Yang thus offered his remarks to Kusaka. The reason for Kusaka’s interest in the poem may be related to Yang’s search for books in Japan, the assistance he gave to Li Shuchang 黎庶昌 (1837–1898) in block-printing the Guyi congshu 古逸叢書, and often speaking of the poetry of Ouyang Xiu. From the above, we can clearly discern the mutual influence and encouragement that marked Sino-Japanese intellectual life in the late Qing period.

3 The “Riben dao ge” and the Edo-Era Concept of Lost and Surviving Books

It is now known when the “Riben dao ge” was introduced to Japan. The Song edition of Ouyang Xiu’s collected works is now held by the Tenri Central Library 天理圖書館, and it is said that the collection was transferred to the Kanazawa Bunko 金澤文庫 in the Kaiqing 開慶 year (1259). Nevertheless, it did not have an impact on literary circles, and therefore it can be disregarded. In 1339, the fourth year of Japan’s Engen 延元 period (1336–1340), Kitabatake Chikafusa 北畠親房 (1293–1354) stated in the Jinnō shōtō ki 神皇正統記 that: “The first emperor of the Qin dynasty loved to seek immortality and pursued the elixir of immortality in Japan. In China, Japan sought the surviving books of the three sovereigns and five emperors, and the first emperor of the Qin dynasty gave all of them to Japan. Thirty-five years later, the Qin burned its books and buried alive its Confucian scholars. The complete works of Confucian classics were then all kept in Japan.”¹⁵ This passage differs from the “Riben dao ge,” and there are no means of confirming a direct relationship between the two. However, it shows that by that time Japanese people already believed the myth that “the full version of Confucian classics were then all kept in Japan,” and used it to define an orthodoxy of their own. There is some corroborating evidence from Chinese texts. For instance, in Yang Weizhen’s 楊維楨 (1296–1370) Song seng gui Riben 送僧歸日本, it states that: “I would like to go to the Eastern Barbarian regions in search of texts, and after returning to China, collate the complete works of Confucian classics.”¹⁶ This is a well-known example. Considering the large number of monks who came to China during the Song and Yuan dynasties, it was a commonplace view in the Song through Yuan (1206–1368) periods that China’s “complete” Confucian texts and a large number of lost books were held by Japan.¹⁷

In the Edo period (1603–1868), the “Riben dao ge” received further attention from Japanese scholars. Matsushita Kenrin 松下見林 (1637–1704) included the poem in the third entry of the first scroll of the Ishō Nihon den 異稱日本傳, published in 1693, the sixth year of the Genroku 元祿 period (1688–1704). This was the first work to include the poem in historical sources of Sino-Japanese relations, and deserves our attention. Subsequently, the poem appeared in section one, scroll two of Itō Matsu’s 伊藤松 (dates unknown) Rinkō Chōsho 鄰交徵書, expanding the poem’s influence even further.

The survival of the one hundred chapters of the Shangshu is admittedly an unrealistic figure of speech. Yet the “Riben dao ge” instilled pride in Japanese scholars for their country’s preservation of surviving Chinese classics, and eventually put this into practice. Verifying the existence of classic texts was no easy matter, and it depended on one’s knowledge of antiquarian cataloging. It was only by the Edo period, particularly the mid-Edo period, that such knowledge gradually evolved.

The Buddha was revered throughout Japan. War generally did not destroy Buddhist temples; classic Buddhist texts were therefore preserved relatively well and their survival was the first to be confirmed. The publication in China of the Yongle bei zang 永樂北藏 and the Jiaxing zang 嘉興藏 reflected the transmission of Buddhist scriptures in the Ming dynasty. Japanese monks made comparisons and discovered that a considerable number of scriptures from outside the Chinese Buddhist canon existed in Japan. Although the Fusō zō gai genzon mokuroku 扶桑藏外現存目錄, by Hōtan 鳳潭 (1654–1738), was not published, it reflected a mastery of surviving Buddhist scriptures among Japanese monks.

In 1701, the Kanotomi 辛巳 year of the Genroku period, the Kanazawa 金澤 monk Tanzui 潭瑞 (dates unknown) wrote the following in Batsu shin ku moku shō go 跋鋟孔目章後, at the end of scroll four of the Kegon gyonai shō mon tō zōku moku 華嚴經內章門等雜孔目:

In Ouyang Xiu’s “Riben dao ge” it says: “The Qin dynasty had not yet burned the books when Xu Fu set out (for Japan), / so the one hundred chapters missing from China are still preserved there. Yet they were strictly forbidden from being sent back to China, / and nobody could see them.” Oh! Is this true or not? Without definite proof, it is not easy to tell. But in terms of the Buddhist classics we have preserved, there are indeed some books missing from China. Recently, I was asked by the publisher to collate the Kongmu zhang 孔目章, so I have written this as a postscript.¹⁸

The Huayan jing neizhang men deng za kongmu zhang is a Huayan 華嚴 scripture written by Zhiyan 智儼 (602–688) of the Tang dynasty (618–907) and has been lost to China for a very long time. It has been left out of major collections, at least until Japanese scholars included it in their compilations of the Xu zangjing 續藏經 and the Dazheng zang 大正藏. In the opinion of Tanzui, the fact that Buddhist texts were extant in Japan in the early eighteenth century was as expected, so he quoted from the “Riben dao ge,” to which his response was one of immeasurable pride.

Such understanding and pride concerning the presence of Buddhist texts in Japan eventually manifested as action. In his article “Riben chuanlai foshu yi yu bi zhe, jizeng Da Qingguo, qing nazhi minglan, yiwei xuejiang guijian zhuang” 日本傳來佛書逸於彼者, 寄贈大清國, 請納之名藍, 以為學匠龜鑑狀, contained in scroll four of the Hoku zen isō 北禪文草 and written in 1793, the fifth year of the Kansei 寬政 period, Daiten Kenjō 大典顯常 (1719–1801) made a request to the bakufu 幕府 to present the Qing dynasty with Buddhist scriptures. It states:

In Japan we have venerated Buddhism for a long time. … These ancient texts that have been lost to their country are nonetheless extant in ours … so we are making concerted efforts to go through various books and pick out hundreds of scrolls. If they can be sent by boat as gifts, received by well-known temples, and read by Buddhist masters, wouldn’t it be impressive? If the people of the Qing dynasty were to transcribe and block-print these scriptures, or re-purchase from our country, then the books would be immortalized, enshrining the Buddha’s teachings forever.

The Qianshu mulu 遣書目錄 listed by Daiten relates to such Buddhist sects as the Tiantai 天台, Huayan, and Faxiang 法相. In all, it contains one hundred sections arranged into six hundred and ninety-two volumes.¹⁹ Although the bakufu did not accede to his request, it was nonetheless a conscious effort to transmit a large number of surviving Buddhist texts to China, which is worthy of praise.

Let us now examine a well-known example from the history of the Confucian classics. In his Jū koku kobun kōkyō jō 重刻古文孝經序, written in 1731 (the sixteenth year of the Kyōhō 享保 period) as part of the Shundai sensei shi shi en kō 春台先生紫芝園稿, Dazai Shundai 太宰春台 (1680–1747) wrote:

There are many ancient Chinese books missing from that country that are nonetheless extant in our country of Japan. From the Song dynasty, a poem by Ouyang Xiu acclaims that “many ancient books and records survived.” Formerly, the monk Chōnen went to Song China, and presented Zheng Xuan’s annotations of the Xiaojing to the emperor Taizong 太宗. Sima Guang and the others were overjoyed. Today more than seven hundred years have passed, and more than a few ancient books have been lost; and yet, the Confucian classic Kobun kōkyō 古文孝經 remains here in our country, Japan. Isn’t that remarkable?

Dazai was a representative figure of the study of archaic words and phrases (Kobunji gaku 古文辞学) of Japan’s Edo period, a school that sought to compete with Neo-Confucianism (songxue 宋學) by unearthing and restoring ancient scriptures. Dazai cited the “Riben dao ge” as a case of “ancient Chinese books missing from that country” which are “nonetheless extant in our country of Japan.” Although we have seen the early Song-period example of Chōnen’s donation of texts, it was the Edo period that became quite conscious of surviving texts. Dazai evidently subsumed the Confucian classics in the westward return of the Guwen xiaojing into the historical context of the return of surviving books. When it spread to Qing territory, the book was soon incorporated into the Zhi bu zu zhai congshu 知不足齋叢書 and generated an enormous response.²⁰

The Isson sōsho 佚存叢書 is another famous example. This collection was compiled by Hayashi Kō 林衡 (1768–1841), the head of the academy (Daigaku-no-kami 大学頭), and it brings together seventeen surviving books. The first of this series was printed in movable type in 1799, the eleventh year of the Kansei period. The preface states:

Ouyang Xiu’s “Riben dao ge” says: “The Qin dynasty had not yet burned the books when Xu Fu set out (for Japan), / so the one hundred chapters missing from China are still preserved there.” However, our country does not have the so-called one hundred chapters of the Shangshu, and I do not know what the evidence for this is. Was it conjecture? … If the hundred chapters of the Confucian classics indeed found their way to our country, I would ensure they would not be scattered and lost. I have read literature from the Tang and Song dynasties and thereafter; hence, I know that a substantial amount of their country’s literature is no longer extant. That is why I think of books that exist only in our country, and if the people of our country were to lose them, then those books would disappear from the world. Would that not be a great pity? So I compiled them together, borrowed from Ouyang Xiu’s poems and named the collection the Isson sōsho.²¹

The Siku quanshu 四庫全書, compiled during the Qianlong 乾隆 period (1735–1796), was at the time the largest book series in East Asia and attracted great attention. The series was only in hand-written form, so its general catalog the Siku quanshu zongmu tiyao 四庫全書總目提要 became an important means with which mid-Edo-period scholars verified the existence of books outside the canon. The fact that the Isson sōsho was able to be compiled was dependent on this.

We may see from the aforementioned that many Japanese scholars from the Edo period did not believe the theory that the hundred chapters of the Shangshu were in Japan. However, they were still moved and inspired by the “Riben dao ge.” In this way, they consciously identified and published surviving ancient texts in Japan, and exerted a major influence on Qing intellectuals.

4 The “Riben dao ge” and Chosŏn’s Goodwill Missions

In the preceding, we have surveyed the influence that the “Riben dao ge” had on Japan. Its secondary influence on Chosŏn’s goodwill missions (t’ongsinsa 通信使) cannot be overlooked either.

In Itsusho 逸書, scroll thirty-six of the Hayashi Razan bunshū 林羅山文集, Hayashi Razan 林羅山 (1583–1657) states that:

A memorial to the emperor from a Chosŏn envoy made this request: It is said that when Xu Fu went to Japan, he took with him pre-Qin texts. That is why Ouyang Xiu says, “The Qin dynasty had not yet burned the books when Xu Fu set out (for Japan), / so the one hundred chapters missing from China are still preserved there.” Presumably, Japan has the Shangshu written in tadpole seal script (kedou zhuanzi 科鬥篆字) as well as other Confucian classics and commentaries. I am honored to visit your country, and if permitted to view the ancient books, I would indeed be fortunate. This has been a lifelong wish, and I hope that this memorial is sufficient for a response. My heart is sincere.

The Ambassador of Chosŏn

The envoy from Chosŏn asked Hayashi Razan about Ouyang Xiu’s “Riben dao ge,” hoping to see the ancient classics and rare books missing from China. It was naturally impossible to see the hundred chapters of the Shangshu. We do not know how Hayashi responded. Yet this draws attention to the fact that when Chosŏn’s goodwill missions went to Japan, the topic of lost books that revolved around the “Riben dao ge” was probably quite a popular one.

In part three of Ang Yŏp ki 盎葉記, in scroll fifty-six of the Ch’ŏngjanggwan chŏnsŏ 青莊館全書, Yi Tŏng-mu 李德懋 (1741–1793) writes:

When Sin Yu-han 申維翰, known by his literary name Ch’ŏngch’ŏn 青泉, went to Japan, he asked Amenomori Azuma 雨森東 (note: the surname is Amenomori, the first name is Azuma, and he was secretary 書記 of Japan’s Tsushima province): Xu Fu began crossing the sea before the Qin emperor had books burned, so it is said that Japan has these authentic ancient texts. Now, for thousands of years, these books have not seen the light of day. Why is that? Amenomori said: Ouyang Xiu also spoke about this. It’s completely unreasonable. The Confucian classics and commentaries are of course the most valuable treasures between heaven and earth, and not even the gods or the ghosts can keep them a secret. Therefore, the Shangshu in the old script could be found on the walls of Confucius’ home or the head of the ship. Although Japan lies far across the sea, there is no reason that the Confucian classics preserved there are not seen by the world. The Japanese people are very boastful. If there is a legacy of the sages that is hidden there on its own, it can be regarded as a rare treasure that has existed for thousands of years, so even if a ban on maritime travel was established, there would be nothing to prevent its resale. And there was no such ban in the first place.²²

In 1719, the fourth year of the Kyōhō period (1716–1736), Shin Yuhan (1681–1752) accompanied envoys to Japan. His notes on his visit were titled Hae sa dong yu nok 海槎東遊錄 and Hae yu munkyŏn chapnok 海遊聞見雜錄.²³ Amenomori Hōshū 雨森芳洲 (referred to above as Amenomori Azuma) was a mid-Edo Confucian scholar. His taboo name was Shunryō 俊良, his common names were Tōgorō 藤五郎 and Azuma Gorō 東五郎, his literary name was Hōshū 芳洲, and his style name was Hakuyō 伯陽.

In the Ch’ŏngnyŏng gukchi 蜻蛉國志, scroll sixty-four of the Ch’ŏngjanggwan chŏnsŏ, it states:

In the Guiwei 癸未 year of Emperor Qianlong’s reign, the secretary from Chosŏn, Wŏn Chunggŏ 元重舉, asked Japanese Confucian scholars such as Kamei Ro 龜井魯: Did Xu Fu really bring the six Confucian classics in the old script with him? They replied: We also read about that in Ouyang Xiu’s poem the “Riben dao ge,” but this country knows nothing of it. Furthermore, our country’s people are always so boastful, and have been in communication with your country for a long time. Even if the country proscribes it, there is no reason why it should not be leaked. What’s more, the country’s people cannot truly understand the value of the six Confucian classics. If they knew its value, it would have to be shared with all the nations of the world. If they didn’t know how valuable it was, why would they keep it a secret?²⁴

Wŏn Chunggŏ (1719–1790), who bore the style name Chajae 子才 and the literary name Hyŏnch’ŏn 玄川, was sent to Japan as an envoy in 1764 and appointed deputy secretary. He wrote books such as Hwagukchi 和國志. Kamei Nanmei 龜井南冥 (1743–1814) was an Edo-era Confucianist; his taboo name was Ro 魯, his style name was Dōsai 道載, and he was often referred to as Shusui 主水, while his literary name was Nanmei 南冥. He was the originator of Kimon studies (Kimon gaku 龜門學).

From another aspect, these questions from Chosŏn’s goodwill missions undoubtedly generated greater interest in the “Riben dao ge” among Japanese Confucian scholars. Due to the prevalence of Neo-Confucianism in the Chosŏn dynasty, insufficient attention was given to the Kogaku school (Kogaku ha 古學派) that began to appear in Japan in the mid-Edo period, and it was even regarded as delusional, highlighted by the brushtalk between the goodwill mission and the Kogaku school in 1748. Thereafter, in 1764, while the Japanese Confucian scholar Taki Chōgai stated in his poem Zeng Cheng Longyuan 贈成龍淵 that “the Japanese people cannot write poems matching Wani’s songs. / The fires of Qin burned all books, except those taken away by Xu Fu. / For the ancient classics that remain today, / we rely on you to pass them on to other places.”²⁵ Most of the goodwill ambassadors never saw the original books with their own eyes,²⁶ and were never able to efficiently deliver to their country the important scholarly information of the ancient classics that existed in Japan.²⁷ From this standpoint, given the objective factors to which Chosŏn scholars were subject, as well as the limitations and restrictions of Neo-Confucian concepts, the idea of lost books that formed around the “Riben dao ge” was not able to grow. Moreover, Japan’s Kogaku school developed further, and by unearthing and publishing Japanese texts, it spurred that country’s intellectual development and made important contributions in that respect. As excitement grew over the release of new texts in Japan, Chinese scholars also began to pay greater attention to collections held outside China. The search by Yang Shoujing and others for books in Japan may be seen as an important response to the “Riben dao ge.” The dissemination of the poem within China, Japan, and Korea is indicative that alongside mutual inspiration, deeply complex relations exist between the three countries.

5 The “Riben dao ge” and “Searching for Lost Rituals among the People”

Ultimately, the “Riben dao ge” is a cultural issue of “searching for lost rituals among the people.”²⁸ Regardless of whether it is Japan or the Korean peninsula, the concepts, topics, and behaviors they generated around the “Riben dao ge” cannot be separated from China, which lies at the heart of the Sinosphere.

As we all know, the historical facts surrounding the return of surviving books can be traced back to the state of Wu (907–937) during the Five Dynasties period and to the Song dynasty. This was a period of momentous change in premodern China, and it was no accident that book donations from Koryŏ and Japan, and China’s pursuit of books outside its borders, occurred at this exact time. From the perspective of the so-called Tang-Song transition, the subject of the “Riben dao ge” is of particular importance.

In the study of non-Chinese civilizations in the Tang dynasty, we find a large number of imports, including implements, flora, and fauna; everything except Sinographic texts. This was because China itself was the founder of Han Chinese culture (including the Sinographic texts that were a vehicle for that culture). This was especially true of the Tang dynasty, which without doubt played a long-term pivotal role in East Asia. From this point of view, it was obviously difficult to imagine or accept that China could import Sinographic texts (symbols of culture) from so-called barbarians. This all gradually changed during the mid-Tang period and even more so during the rebellions of the late Tang. Literature, a vehicle of culture, suffered large-scale destruction by fire in that period. The state of Wu in the Five Dynasties period was governed by the Tiantai sect, which knew that many of the Tiantai texts were held outside China, and spent vast amounts of money purchasing them. When China was unified in the early Song dynasty, a policy of “revering the civil arts and suppressing the martial arts” (Chong wen yi wu 崇文抑武) held sway, due to an urgent need for cultural restoration. This was an adaptation to the cultural demands of the dynasty. The return of lost books and the pursuit of books outside China emerged as demanded by the times.

From this it would appear that the pursuit of books outside China, or the offering of such as tribute, are to a great extent acts of the state with strong political and diplomatic overtones. This led to a peculiar phenomenon not previously seen in the Tang dynasty. That is, China, which had long been a cultural sovereign, began to seek de facto aid from neighboring Japan and the Korean peninsula, which it had viewed previously as “barbarians” and vassal states; in other words, it purchased or was offered, texts as a way to restore the nation’s cultural traditions. It may be said that the “Riben dao ge” is a cultural manifesto from Sima Guang, standing on China’s side, in which he “searched for lost rituals among the people” of peripheral cultures. Through literary language, the poem frankly acknowledges the fact that Chinese texts, with their representative “canon by sovereigns past,” lay hidden in foreign parts. In the wake of changes in East Asia, the cultures of the three nations gradually evolved, leading to psychological differences in terms of book diplomacy. In this way, it is easy to understand why Chosŏn’s goodwill ambassadors frequently referred to the “Riben dao ge” in their conversations with Japanese Edo-era scholars, and closely inquired as to the location of the Guwen Shangshu. These are issues related to cultural orthodoxy.

Another dimension to this is that the “Riben dao ge” has been a topic of heated debate among scholarly experts. For instance, there are many references to the poem in literati’s notes from the Ming dynasty. These include “Waiguo shu” 外國書 in scroll nine of Zhang Dingsi’s 張鼎思 (1543–1603) Langya daizui bian 瑯琊代醉編, “Woguo yi shu bai pian” 倭國逸書百篇 in Huang Yu’s 黃榆 (1426–1497) Shuanghuai suichao 雙槐歲鈔, and “Zhuzi bu zhu Shangshu” 朱子不注尚書 in Du Mu’s 都穆 (1458–1525) Tingyu jitan 聽雨紀談. From these examples, it can be seen that discussions of the poem were a popular trend.²⁹ Most of these notes believed that the Shangshu in the old script was preserved in Japan. There were also other records, such as “Zhu shu” 著述, in supplementary scroll four of Shen Defu’s 沈德符 (1578–1642) Wanli yehuo bian 萬曆野獲編, which mentions that Liu Yuanqing 劉元卿 (1544–1609), in charge of the Board of Rites (libu 禮部), delivered a message to the emperor requesting to use these “rediscovered” texts from Japan to fill in lacunae in Fu Sheng’s 伏生 (268 BCE–178 BCE) version of Shangshu. The introduction of these texts also had their own impact on Japan and the Korean peninsula, where Chinese writings were closely read.

This dimension was more evident in academia. For instance, Confucian scholars of the Qing dynasty were fond of discussing the “Riben dao ge” (but the majority took a negative view); the Haedong yŏksa 海東繹史, written in the early nineteenth century by the Chosŏn historiographer Han Ch’iyun 韓致奫 (1765–1814), offers a range of understandings of the issue. Han’s book was a compilation of documents from the Song, Yuan, Ming and Qing dynasties related to the existence of the Shangshu in old script in Japan and the Korean peninsula. Han concludes that:

The people of the Song dynasty always suspected that Koryŏ had the one hundred chapters of the Shangshu since these texts had not been handed down in China. Additionally, because of what Ouyang Xiu wrote in “Riben dao ge,” they went so far as to search for them outside China, so much so that Emperor Shizu of Yuan 元世祖 (Kublai Khan) asked the Crown Prince of Koryŏ about the matter. In the Wanli 萬曆 period of the Ming dynasty, Ye Chunji 葉春及, whose literary name was Jiongzhai 絅齋, submitted a memorial to the emperor requesting permission to make multiple demands to Japanese special envoys for the Shangshu in old script. In the early years of the Kangxi period of the Qing dynasty, a certain salaried scholar surnamed Cai also submitted a memorial to the emperor requesting permission to travel abroad to obtain the Shangshu in old script, again without success. Japan did not have it, however. Gu Tinglin 顧亭林 has already said so, and scholars such as Zhu Zhutuo 朱竹坨 and Yan Qianqiu 閻潛邱 have also explained it at great length. When Shin Yuhan went to Japan, he asked Amenomori Azuma … in the Guiwei year of Emperor Qianlong, the envoy secretary from Chosŏn, Wŏn Chunggŏ, asked the Japanese Confucian scholar Kamei Nanmei … Based on these accounts, we understand that neither Koryŏ nor Japan had the Shangshu in old script. This is enough to resolve that eternal question.³⁰

Han Ch’iyun primarily cites the works of Qing scholars from the so-called Qian-Jia 乾嘉 period (1735–1820). These include works such as the Rizhi lu 日知錄, Jingyi kao 經義考, Xihe ji 西河集, and Qianqiu zhaji 潛邱札記. On this basis, he refuted the notion that Koryŏ or Japan were in possession of the Shangshu in the old script. It may be seen that where discussion of the “Riben dao ge” was concerned, Chinese scholars, particularly those of the Qing dynasty, attracted the attention of Korean historians, and this appears to be one of the primary reasons that the poem attracted the attention of Chosŏn and Japanese scholars.

6 Conclusion

The “Riben dao ge” was a work written by Sima Guang during his appointment at the Palace Library. The poem expresses his cultural expectations as well as some satirical remarks about Qian Gongfu. At some point thereafter, it may have been Wang Ledao and his son who added the poem to Ouyang Xiu’s collected works the Bieji, and so it has been mistakenly identified as the latter’s work for nearly one thousand years. It was only in the late nineteenth century that the Japanese scholar Kusaka Hiroshi proposed that it was the work of Sima Guang.

After the seventeenth century, following the eastward spread of the Siku quanshu zongmu tiyao, it became feasible to confirm the existence of ancient texts. The “Riben dao ge” attracted sustained interest in intellectual circles in Japan and Chosŏn. It even became an important theme of Japan-Korea diplomatic occasions, the context of which was a response to, and an interest in, China’s cultural issue of “searching for lost rituals among the people.” Despite their discussions on the matter, Chinese, Japanese, and Korean scholars basically recognized that the survival of lost books, of which the “Riben dao ge” speaks, was a misrepresentation. Yet this pursuit inspired Japanese scholars to collect and organize lost and surviving texts; simultaneously, it also created a chain reaction in Qing intellectual circles. The spread of the “Riben dao ge” around China, Japan, and Korea provides a vivid historical imagery of East Asian cultures inspiring one another.

Translated by Damien Kinney