Swapping odes in a sacred language: the Kanshi exchange of Natsume Sōseki and Masaoka Shiki and its meaning

Abstract

Natsume Sōseki and Masaoka Shiki, two of the towering figures in modern Japanese literature, exchanged kanshi poems for eleven years starting in 1889 when they were students. What Sōseki and Shiki enacted in their kanshi exchanges was not simply an admiration for Chinese culture, but rather a performance of literati cultural exchange. In the personae that these two writers adopted in these exchanges, and in the poetic voices that each writer meticulously honed, they were achieving a return to a cultural homeland and to an “imagined community” in the Sinosphere. Further, the exchange of poetry in a language that was simultaneously both foreign and hauntingly familiar demonstrated a performative quality that reflected their appreciation of the dynamics of poetic exchange in China but also of the yose theater and of the rakugo performances that they frequented as students in Tokyo. In fact, for Sōseki and Shiki, kanshi composition and exchange served two paradoxical purposes: it offered both the challenge of poetic expression in a foreign language and a return to an imagined community and to the familiar rhythms and conventions of a sacred language.

Introduction

The production and consumption of kanbun 漢文 and kanshi 漢詩 constitute a tradition in Japan that stretches back 1300 years. Although the capacity among Japanese to read and write in Chinese has diminished in the modern era to the point that only a handful of enthusiasts carry on that tradition, a facility with kanbun was still a mark of the educated person at the beginning of the modern era in the Meiji period (1868–1912). Meiji leaders in fields as diverse as politics, culture, education, and the military produced kanbun texts or kanshi to commemorate public occasions or to convey something of import.

This essay proposes that the exchange of poems in Classical Chinese contributed in crucial ways to the respective development of Masaoka Shiki 正岡子規 (1867–1902) and Natsume Sōseki 夏目漱石 (1867–1916) as writers while it consolidated their friendship. Communicating in a non-native language paradoxically allowed these two novice writers to express themselves candidly and to experiment with both form and style. Along with a shared love of kanshi, Shiki and Soseki also shared a passion for yose 寄席 theater “popular variety shows” and particularly rakugo 落語 “comic raconteurs,” and as students they frequented yose theaters. Their poetic exchange consequently exhibited a performative quality, and they sometimes adopted personae that highlighted their youthful passion for Chinese culture while underscoring their sensitivity to the conventions of literati cultural exchange. Admittedly, Classical Chinese was hardly “foreign” to these writers, educated as they were in the Edo terakoya 寺子屋 tradition. As with the kanshi poets who preceded them in Japanese cultural history, Shiki and Sōseki used poetry as a means of self-cultivation but also as a way to critically comment on contemporary society. For Shiki and Sōseki, as for other Meiji intellectuals, kanshi composition and exchange served as a kind of antidote to Bunmei kaika 文明開花 “Civilization and Enlightenment” and to the headlong rush to embrace Western culture.

Kanshi exchange in the context of Meiji culture and Bunmei Kaika

By the time that Masaoka Shiki and Natsume Sōseki had initiated their exchange of kanshi poetry, Japan already could boast a venerable kangaku tradition of well over a millennium, and it participated in the Sinitic cultural sphere of influence that encompassed East Asia from what is now Tibet in the west to Vietnam in the south. Japanese kangaku scholars were conscious participants in the Sinosphere, and for generations of Japanese thinkers and writers, traditional bunjin 文人 “literati” (Chinese wenren) culture constituted an “imagined community,” a term used by Benedict Anderson to refer to the nation state as a socially constructed community comprised of individuals who perceive themselves as part of that shared community based on linguistic and cultural affinities. In this essay, I will posit producers and consumers of kanbun and kanshi in Japan, who fashioned themselves in the image of Chinese wenren participating in the Sinosphere, as an “imagined community.” Poetry in the societies that comprised the Sinosphere served among the cultural elite as a tool to celebrate connections with the values of the past and to demonstrate affinities with particular cultural epochs and figures.

One important function of poetry among the literati in the Sinosphere was poetic exchange, a tradition embodied in the practice of changhe 唱和 (Japanese shōwa), or “singing in harmony,” which constituted a kind of dialogue in poetry in which one poet responded to another's poem according to strict prosodic rules (Li 2011, p. 216). Chang means “to sing” and he means “to echo,” and the he poem, by convention, responded both in terms of content and form to the chang poem (Zheng 2012). The practice of changhe poetic exchange, which evolved as a communicative activity between individuals with equivalent cultural backgrounds, has a history that stretches back to the Jin Dynasty (266–420) and included Tao Yuanming 陶淵明 (365–427) among its famous practitioners (Ibid).

The changhe exchange of poetry and the practice of sharing poetry for mutual critique reflects an important strain in Chinese cultural history and was commonly practiced throughout the Sinosphere. In the centuries that followed its origins in the Jin Dynasty, the changhe practice continued to evolve, and by the mid-Tang Dynasty (618–907) new conventions governing the production of changhe poems required that the he poem match the original chang poem with corresponding meter and rhyme schemes and with identical Chinese characters ending the first, second, and fourth lines (Ibid). Due to the challenges associated with these generic developments, by the height of this practice in the Tang and Song Dynasties, changhe poems had largely abandoned the communicative function and had been reduced to a tool to showcase poetic skills and erudition and to consolidate literary relationships (Ibid). As this practice spread throughout the Sinosphere, the conventions accompanied the practice of poetic exchange. In the modern era, when Masaoka Shiki and Natsume Sōseki engaged in this poetic exchange using kanshi, no practice better demonstrated one's commitment to the traditional values of the Sinosphere than did the custom of changhe exchange.

The exchange of kanshi between Shiki and Sōseki is most profitably viewed not only against this tradition of poetic exchange in China but also against the long tradition of poetic exchange in Japan, including both kanshi exchange and exchanges of traditional Japanese poetic forms. Direct exchange of poems between poets in the form of correspondence as part of the larger phenomenon of communal production and exchange has historically played a central role in Japanese poetic production. Robert Tuck uses the term “poetic sociality” to refer to a range of practices in Japan that treat poetry as a communal experience including but not limited to poetry contests, linked verse composition, master-disciple exchanges, poetic critique among peers, and poetic correspondence of the type practiced by Shiki and Sōseki (Tuck 2012, p. 3). Tuck's scholarship builds upon prior scholarship including Ogata Tsutomu's pioneering Za no bungaku and subsequent scholarship by Haruo Shirane, Eiko Ikegami, and H. Mack Horton that have investigated the role of poetic exchange within medieval and early modern modes of Japanese literary production.

From the Nara period (710–794) onward, such communal poetic practices were conducted both in Japanese and Classical Chinese. The “rhyme-matching” or “rhyme-following” (Japanese wa'in 和韻 or ji'in 次韻; Chinese heyun or ciyun, respectively) that Shiki and Sōseki engaged in was but one example of the practice of poetic sociality that had a venerable lineage in the Japanese literary tradition but which was enjoying renewed interest in the Meiji period (1868–1912) due to significant developments in print culture. Shiki and Sōseki were readers of Narushima Ryuūhoku's 成島柳北 (1837–1884) kanshi column in Chōya Shinbun 朝野新聞 and of a similar kanshi column that appeared in the newspaper Nippon called Hyōrin 評森 (Forest of Criticism), which featured kanshi exchanges by contributors, and they had absorbed the lesson that rhyme-matching was an essential element in the social rituals of kanshi practice (Ibid., p. 102).

Like other writers of their generation who sought to retain a space for kanbun composition in the new Meiji literature, Natsume Sōseki and Masaoka Shiki developed approaches to literary production in which kanbun composition coexisted with and complemented their wabun composition. In fact, the first poems of Masaoka Shiki to appear in the literary section of the newspaper Nippon were neither haiku nor tanka but rather kanshi, and both Shiki and Sōseki would continue to compose kanshi alongside Japanese poetry and prose throughout their lives (Ibid., p. 172). For Shiki and Sōseki, as for many Sinophiles among the intellectual elite in the Meiji period, the mission of fashioning a modern national identity was inextricably intertwined with the task of maintaining ties to the Sinosphere. While the tradition of kanbun and the cultural indebtedness to China were subject to periodical scrutiny during the Meiji period, it is important to acknowledge the simultaneous emergence of discourses that represented the kangaku tradition not as antithetical to but rather as an indispensable element of the modernization process (Hedberg 2020, p.3).

In Meiji Japan, the practice of changhe poetic exchange involved a tacit acceptance of the practices that defined the exchange throughout the Sinosphere. The kanshi scholar Saitō Mareshi has demonstrated that the continued tradition of kanbun production constituted an important and vibrant cultural strain in the Meiji Period. In fact, despite a lull in interest in kanbun as the Japanese sought to satisfy a growing hunger for Western cultural forms, exposure to kanbun texts actually increased in Meiji Japan with rapid developments in printing technology (Saitō 2007, p. 119). In Meiji Japan, kanbun and kanshi production and consumption were conducted against the backdrop of a reassessment of the value of Sinitic cultural values and practices. How Japan should modernize and what should be retained and what abandoned of its cultural heritage in the process of modernization were questions that consumed Meiji intellectuals. The preservation of traditional cultural values against the onslaught of Western culture would become something of a national obsession in the Nihon e no kaiki 日本への回帰 “Return to Japan” Movement in the 1920s and 1930s, but those concerns were already the subject of vigorous debate in intellectual circles in the Meiji period (Hutchinson and Williams 2007, p. 1).

Kanbun production in Meiji Japan took place in the context of a debate about the viability of kangaku in a society that was increasingly looking to the West for its cultural cues. Between 1880 and 1895, intellectuals in Japan sympathetic to Sinitic culture engaged in a running dialogue concerned with finding a space for kangaku in contemporary intellectual discourse. The debate about the status of kanshi in Meiji intellectual life often involved members of the Japan Academy, which was founded in 1879, and a variety of views on the subject were shared in the pages of the Academy's journal, Tōkyō Gakushikaiin Zasshi 東京学士会院雑誌 (Tokyo Academy Magazine). Among the topics discussed by these scholars were whether kangaku might not serve as an alternative to the Western-centric educational values being promulgated by the new Meiji government and whether or not to focus solely on Classical Chinese as a dead language – a “moribund and dying corpus to be jettisoned” – or to approach Chinese in the same way educators were approaching Western languages, as a “progressive discipline with revolutionary potential” (Ibid., p. 6). This debate was enacted, of course, at precisely the moment that relations between Japan and the Qing Dynastic government were deteriorating in the years prior to the Sino-Japanese War of 1894–1895, further complicating public attitudes in Japan toward kangaku.

In fact, the exchange of kanshi between Shiki and Sōseki took place at a moment when some intellectuals such as Shiroi Kunitsuna 城井国綱, the editor of the contemporary kanshi anthology collection Meiji meika shisen 明治名家詩選 (Anthology of Famous Poetic Masters of the Meiji Period) were defining a “distinctly Meiji era approach to kanshi” (Lynn 2003, p. 105). Robert Tuck, for example, describes kanshi composition and sociality during the Meiji period as “transnational.”¹ Rather than drawing a strong line of demarcation between kanshi and waka composition, Shiki envisioned the two as complementary and rejected the idea that poetic composition was intrinsically connected to national identity (Tuck 2012, p. 18). Shiki and Sōseki, like others of generation, consciously adapted traditions received from the Sinosphere to the emergence of a national identity.

By the Meiji period, after over a millennium of engagement with kangaku, the Japanese had refined and modified reading and writing practices to the extent that kanbun was a fully integrated dimension of Japan's own cultural and intellectual heritage. Although kanbun remained Chinese in origin, it served for the Japanese as the “flesh and blood” of cultural life, and kanshi, which constituted the marriage of Chinese prosody and Japanese aesthetic sensibilities, represented the best of both cultural traditions (Makizumi 2013, p. 99). Nevertheless, in choosing to articulate contemporary concerns in kanshi, Meiji intellectuals were painfully aware of the irony inherent in expressing their ideas at this critical historical juncture in a form that was rapidly slipping into irrelevance.

Kanbunmyaku as imagined community

The Japanese term Kanbunmyaku 漢文脈 closely parallels the English term “Sinosphere” and refers to the lineage of Chinese language expression in Japan and to the community comprised of Japanese producers and consumers of kanbun. In the Meiji period, Shiki and Sōseki were writing as part of that community, which was in crucial ways an “imagined community,” to appropriate Benedict Anderson's term. Benedict Anderson's concept of imagined communities is surely familiar to many, but it nevertheless bears reminding along with a simple explanation of how I am applying that concept in this study. Anderson uses this term in the context of the emergence of the nation state in the nineteenth century and the accompanying emergence of a consciousness of nationalism and national identity, but the term also points to the shared values surrounded extended cultural communities “distinguished by the style in which they are imagined” (Anderson 2006, p. 6). It is in this latter sense that I am using this term to describe Masaoka Shiki's and Natsume Sōseki's consciousness of themselves as members of the community of producers and consumers of Classical Chinese in the Kanbunmyaku.

Imagined Communities, as defined by Anderson, also refer to global or extended communities that coalesce around religious and cultural traditions, which embrace texts written in what he refers to as “sacred silent languages” through which shared cultural values are transmitted (Ibid., p. 14). One of the examples of “sacred silent languages” that Anderson cites is that of the community of consumers and producers of Classical Chinese. At the turn of the twentieth century, writing in Chinese by Japanese writers and other non-Chinese writers in the Sinosphere constituted an example of exophonic literary practice that had garnered conventions over a 1300-year history.

“Exophony,” and its derivative adjective “exophonic,” describes the phenomenon of writing produced by non-native users of a language. The term first emerged to describe the works of African literature written in Western languages, particularly in French (Wright 2008, p. 39). The expression becomes more widely disseminated in academic circles as a result of the critical success of a collection of essays called Exophonie by Tawada Yōko 多和田葉子 (1960-), which treats the topic of leaving one's mother tongue in order to write in an adopted language. Tawada had become familiar with the term at a conference about exophonic literature in Senegal in 2002 and recognized in this concept a reflection of her own struggles to define herself as a writer who had produced works in both German and her native Japanese (Ibid., p. 39).

Samuel Beckett, who was at once one of the twentieth century's most famous practitioners of exophonic literature and an advocate of literary production in a non-native language, had the following to say about his own experience of writing in French:

More and more my own language appears to me like a veil that must be torn apart in order to get at the things (or the Nothingness) behind it. Grammar and style. To me they seem to have become as irrelevant as a Victorian bathing suit or the imperturbability of a true gentleman.²

Traditionally, Classical Chinese was not only for the Japanese a “sacred silent language,” it was also, as Mizumura (2014, p. 106) asserts, an “external language” divorced from the critical day-to-day communicative function. In this way, Kanbun represented an important, time-tested manifestation of exophonic literature in Japan. Traditionally, the capacity for a Japanese to compose prose or poetry in Chinese using the grammar, syntax, and conventions of Classical Chinese was the sine qua non for the cultured individual in the Sinosphere, and by the Heian period (786–1185) kanbun had come to constitute the foundational core of Japanese culture (Makizumi 2013, p. 99). In particular, the ability to compose a kanshi poem by a Heian courtier was both the mark of breeding and an expression of engagement with East Asian intellectual traditions. During the Heian period, the hierarchy between Classical Chinese and Japanese was an “absolute given,” and male bilinguals in Japan would continue to be highly conscious of the perceived superiority of Classical Chinese to convey important official information all the way through the dawn of modern Japan (Mizumura 2014, p. 110). In choosing to incorporate kanshi exchange into their correspondence, Shiki and Sōseki, like generations of Japanese writers preceding them, were choosing to share their most intimate thoughts in a foreign tongue, embracing the practice of exophonic literature.

Some commentators feel uncomfortable treating Meiji era kanshi as an expression of exophonic literature or indeed of approaching Classical Chinese as a “foreign” language at all. Admittedly, Japan long participated in the Sinosphere, and by the modern era, Classical Chinese had been embedded in Japanese culture already for a thousand years. Nevertheless, Classical Chinese is most properly treated as a form of exophonic literature in Japan. Mizumura refers to Classical Chinese as an “external language” even for the Chinese themselves, divorced from the modern language and traditionally from specific regional dialects (2014, p. 110). For the great majority of Japanese readers and writers of Classical Chinese starting in the Asuka period and continuing into the modern period, regardless of their facility with Classical Chinese and its conventions, kanbun constituted non-native literary production. Moreover, it is apparent that the Meiji writers themselves recognized Classical Chinese as “foreign.” The Meiji kangaku scholar Shigeno Yasutsugu 重野安繹 (1827–1910) and like-minded scholars endorsed treating the study of Chinese the same way that Meiji Japanese were tackling the acquisition of Western languages, as a living language, a foreign tongue that should be approached via study of the spoken language (Hedberg 2020, p. 7).

The recognition that Classical Chinese remained for Meiji era writers and readers of kanbun a truly foreign language is manifested, for example, in the desire on the part of Meiji kanshi poets to engage with native Chinese poets such as visiting Qing dynasty (1616–1912) scholars and diplomats including He Ruzhang 何如璋 (1838–1891) and Huang Zunxian 黄遵憲 (1848–1905), as well as private figures such as teacher Ye Songshi 葉松石 (1839–1903), as a way to legitimize their own kanshi compositions and to give weight to their kanshi columns and journals (Tuck 2012, p. 24). These Chinese visitors, well known in the Japanese kangaku community, were invited to contribute to kanshi journals and were frequent guests at poetic gatherings in Tokyo. The desire to engage with these Chinese poets – indeed to receive their stamp of approval – is an indication of the desire on the part of these Japanese poets to maintain ties to traditional Sinosphere but also constitutes a manifestation of exophonic angst and a desire for the recognition of the native speaker poet.

As with other Exophonic writers, those Meiji era Japanese who produced and consumed Classical Chinese were choosing “to live in more than one language” (Slaymaker 2007, p. 3). Among the latter who choose to write simultaneously in two languages, some may write in the second language for certain purposes, and their second language writing may be of a very different character than their writing in their first language; others may produce literature in a second language that bears the trademarks of their body of writing in their native language. The kanshi produced by Natsume Sōseki and Masaoka Shiki reflect both of those tendencies.

Shiki and Sōseki and Kanshi exchange

By the time that Shiki and Sōseki met as students at Ikkō 一校, the First Tokyo Middle School, when both were twenty years old, each had already received a thorough kangaku education and produced both kanbun prose pieces and kanshi. Along with their enthusiastic participation in the Kanbunmyaku, what the two young aspiring writers shared was a pride in familiarity with the Chinese literary tradition and with an education that resulted in a facility with kanshi production. This shared interest in kanshi resulted in the beginning of an exchange of poems that would last from the early days of the two writers' friendship in 1889 until shortly before Shiki's death when Sōseki sent Shiki a final kanshi before departing for England in 1900. Of the sixty-seven kanshi poems that Natsume Sōseki penned in the eleven-year period from his first poetic exchange with Masaoka Shiki until the latter's untimely death, nearly all were shared with Shiki (Xu 2005, p. 7). Shiki is one of only a handful of poets with whom Sôseki practiced rhyme-matching. Of these, Shiki is the only one with whom he exchanged more than one verse, a fact which demonstrates both the depth of their friendship and the highly social nature of their correspondence (Tuck 2012, p. 128). This exchange, conducted in classical Chinese, was mutually enriching and was predicated on the shared expectation of peer critique.

It was almost inevitable that Masaoka Shiki, the most important Japanese poet in the modern era, should have remained committed to the production of kanshi poetry throughout his life. Shiki received his early education from his grandfather, Ōhara Kanzan 大原観山 (1818–1875), a Kangaku scholar and teacher in the Matsuyama Domain. Provincial Kangaku scholars and teachers in that era like Ōhara were important intellectual figures, highly respected in their communities (Ibid., p. 7). After his initial education with his grandfather, Shiki continued this education with another Confucian scholar in Matsuyama named Tsuchiya Kyūmei 土屋久明. Shiki would later recount that from the age of eleven he composed one five-character jueju 絕句 (Japanese zekku) quatrain per day under the tutelage of Tsuchiya (Satō 2020, p. 191). Shiki ultimately would leave Matsuyama before completing junior high school in order to enroll in Ikkō in preparation to entering Tokyo Imperial University. It was during this period that Shiki would meet Sōseki and begin the exchange of kanshi.

As with Masaoka Shika, Natsume Sōseki too was among the last generation of Japanese to receive a traditional kangaku education. Sōseki received an informal education in kangaku as a child and was proficient enough by age eleven to compose an essay about Kusunoki Masashige 楠木正成 (1294–1336), the loyalist who had fought on behalf of Emperor Go-Daigo 後醍醐天皇 (1288–1339) in the early fourteenth century (Nathan 2018, p. 12). At age fifteen, Sōseki entered Nishō Gakusha 二松学舎, a traditional terakoya-style academy of Chinese Studies, where he read and commented on the Analects, Mencius, Chinese dynastic histories, and Tang and Song dynasty poems. Sōseki would later contend that he had memorized thousands of poems during what amounted to little over a year at Nishō Gakusha (Ibid., p. 12). By the time that Shiki and Sōseki met, both were not only familiar with the tradition of kanshi poetic exchange, but each had already engaged in the production of kanbun and kanshi. The exchange of kanshi between Shiki and Sōseki began in 1889 when Sōseki agreed to comment on kanshi composed by Shiki that were included in a manuscript that the latter had self-published called Nanakusashū 七草集 (Collection of Seven Herbs), which included pieces in seven different genres including kanshi. It was with this critique, written in kanbun, of Shiki's work that Natsume Kin'nosuke first used the nom de plume Sōseki (Natsume 1993a, pp. 72–73).

In that same year, while Natsume Sōseki was enrolled at Ikkō, he traveled during his summer break with several friends on a journey that started in Shizuoka Prefecture and ended in Chiba Prefecture. In the fall, he put together a kanshi cycle entitled Bokusetsuroku 木屑録 (A Record of Wood Chips). Sōseki later commented that this collection, evoking the aesthetic ideal of fūryū 風流, which emphasized surprising encounters with beauty in unexpected places and allusions to shared cultural values, was written while thinking of Masaoka Shiki, who had by that time dropped out of Ikkō and returned to Matsuyama (Satō 2020, p. 194).

Sōseki's command of the forms and the thematic concerns of kanshi poetry is already on full display in this early cycle, as demonstrated in poem number 13 entitled “Remembering Friends Who Remain in the Capital,” which opens with the line:

魂飛千里墨江邉 The soul flies for a thousand miles across the banks of the Sumida River.³

This first line of the poem with its image of a soul in flight, in keeping with the playful elegance of the fūryū aesthetic tradition, is jarring and unexpected. The remainder of the poem is directed to the college mates Sōseki has left behind in Tokyo, which until recently had included Shiki, and expresses the speaker's inability to find solace in either the intoxication afforded by alcohol or poetry (Kamiyama 2011, pp. 57–58). The tone also, one of melancholy at separation from like-minded comrades, reflects one of the thematic concerns of Tang Dynasty poetry.

The shared passion for kanshi that the two aspiring writers discovered in one another was accompanied by a shared interest in popular yose theater and particularly in rakugo raconteur monologues. One quality that defined the exchange of poems between Sōseki and Shiki was its theatricality, the adroit adoption and maintenance of personae reflecting the conventions of the Sinosphere. Yose, which originated in the Edo period and continued to flourish in the Meiji period, were vaudeville-like theaters that featured acts designed to reach a mass audience including among other performative forms, manzai 漫才 “omic dialogues,” naniwa-bushi, 浪花節 a type of narrative singing originating in the Kansai region, and most prominently rakugo. Rakugo, which developed in the Edo Period when it was known as karukuchi 軽口, had its origins in Buddhist sermons. Rakugo became a mainstay of popular urban entertainment in the new capital of Tokyo. Rakugo, which continues to flourish in Japan today, is characterized by a crisp, charismatic speaking style, punctuated by a comic punchline that often requires the raconteur to adopt several personae and voices to convey a story and to engage with the audience. Along with a shared taste for kanshi, from the time that they met in 1889, Shiki and Sōseki recognized in the other a mutual fascination with rakugo, and the two often would attend performances together at a yose theater in the Nihonbashi district called Isemoto (Ikkai 1996, p. 44).

At times earnest, at times playful, Shiki and Sōseki developed dramatic personae in these exchanges, donning masks lifted from the long tradition of changhe poetic exchange in the Sinosphere. Among the common personae who appear in Chinese poetry from the Six Dynasties (222–589) onward through the Golden Age of Chinese poetry in the Tang and Song Dynasties are the scholar-official whose contributions to society go unrecognized and who has decided to leave the world of constraining official duties behind and to return to the country; the poet who is tearfully welcoming an old friend or, more often, regretfully bidding farewell to a boon companion; and the poet living far from the capital and thinking of his wife, children, and comrades back home. Sōseki and Shiki, well aware of the personae that populated traditional Chinese poetry and familiar with the cadences of their voices, channeled those voices in their own poetic exchange.

For Shiki and Sōseki, there was no inherent contradiction in adopting voices drawn from the tradition of poetic exchange in China and Japan and writing “authentically” from experience in accord with the modus operandi of shasei. The goal of shasei implied a focus on a genuine emotional response to phenomena rather than contrived emotions. The development of personae in kanshi drawn from tradition allowed for the expression of an authentic emotional response and thus did not stand in odds with the goals of shasei. These personae and the resultant poems were the repositories of timeless truths that transcended the simple act of literati mimesis.

In mimicking the voices of the speakers in traditional kanshi, Sōseki and Shiki drew upon their familiarity with rakugo performance. The coincidental interest in the insouciant wit and word play associated with rakugo would influence the performative quality of their poetic exchange. In their correspondence and the kanshi that they invariably included as part of that exchange, Shiki and Sōseki often chose self-deprecating names that echoed the humility that is a common feature of Tang Dynasty poetic exchange. For example, Sōseki referred to himself in one of these exchanges as ganfu 頑夫 “stubborn fellow” while Shiki signed off as dassai gyofu 獭灾漁夫 “otter fisherman” (Soseki zenshū 1993, vol. 18, p. 124). In another instance of this word play, Sōseki referred to himself as Kikui-machi no namakemono 喜久井町の怠け者 “lazy bones of the Kikui village” while referring to Shiki as Saikikei 才気兄 “genius older brother” (Kang 2005, p. 35; Wada 2002, p. 427).

Similarly, the two young writers sometimes adopted gendered personae in these exchanges. In a kanbun letter that Shiki sent to Sōseki in August 1889, which the latter received at an inn while traveling during the journey that would result in the kanshi cycle Bokusetsuroku that autumn, Shiki playfully referred to Sōseki as “my beloved husband” and to himself as “wife,” adopting marital roles that reflected a trope among primarily male poets in the Northern and Southern dynasties (420–589 CE) (Satō 2020, p. 194). Poem number two in the cycle is a coquettish response to a letter written in kanbun that Sōseki had received from Shiki, in which Shiki had made that marital reference, and the poem ends with the couplet:

馬齡今日廿三歲 始被佳人呼我郎

Finally at age twenty-three/ for the first time to be called “husband” by a beautiful woman (Natsume 1993b, p. 123).

Shiki's poetic rejoinder to Sōseki's poem takes the form of a 次韻 ciyun (Japanese ji'in) in which a poet would compose a poem in the same form as the original received poem, here a seven-character quatrain, ending selected lines with the same characters as the original poem. In Shiki's response poem, the character kō 黄 that ends the first line, the character shō 傷 that ends the second line, and the character rō 郎 that ends the final line are all identical to the characters that end the corresponding lines in Sōseki's poem. Shiki, moreover, refers to the healing properties of saltwater bathing and further attempts to demonstrate his erudition by referring to an amber-colored liquor referred to in a poem by Du Fu. The poem's opening couplet 首聯 shuren (Chinese shoulian), in which that liquor is referenced, reads as follows:

羨君房海酔鵝黄鹹水医痾若薬傷

I envy your drunken revelry in Bōsō, the salt water that heals wounds like medicine.⁴

Shiki's poem ends with the character rō 郎, the same character that delineates ones husband with which Sōseki's poem ends.

The use of gendered personae in these poetic correspondences was on one level a means for Shiki and Sōseki to demonstrate to one another familiarity with this particular convention of the kanshi tradition. There is also a homoerotic quality to this exchange that suggests the level of intellectual and emotional intimacy that the two young writers had achieved by that juncture in their relationship. The relationship between Sōseki and Shiki developed within the largely male world of Meiji intelligentsia forged in their experience as elite students in Tokyo (Kang 2005, p. 12). The scholar Takahashi Hideo described the relationship of the two writers at that stage as “hermetic,” a closed relationship that revolved around the exchange of kanshi and which admitted no outsiders (Takahashi 1996, p. 40).

It is likely that the expressions of romantic intimacy included in the kanshi exchanged between Natsume Sōseki and Masaoka Shiki were a kind of gamesmanship, an ever-deepening competition to demonstrate one's facility with the traditions of the Kanbunmyaku. The personae and voices that they cultivated in this exchange were beholden to the Chinese dynastic tradition of changhe poetic exchange, and this exchange of kanshi, signed off with different stock figures derived from the Chinese poetic tradition, became a kind of role-playing game for the two. In the exchange of kanshi poems, and in the correspondence that accompanied those poems, Shiki and Sōseki inhabited these adopted personae and addressed one another using dramatic figurative expressions designed to show off their respective knowledge of Chinese literature through the use of allusion but also to frame their relationship in terms of Sinitic culture. The terms were often mischievous, reflecting the youthfulness of the two writers that seemed at odds with the sophistication of the word play being conducted in a foreign language.

In fact, many of the kanshi poems exchanged between Natsume Sōseki and Masaoka Shiki were about the experience of meeting or parting and conveyed the heightened yet subtle emotions that are the hallmark of the great tradition of changhe treating this theme. Those emotions are clearly on display in the following poem by Shiki entitled “Seeing Off Natsume Sōseki to Iyō.” Iyō is the former name of Ehime Prefecture, and Sōseki was returning to his teaching post in Matsuyama on January 7, 1896 after spending the New Year's holiday in Tokyo. The poem echoes the sentiments expressed in similarly themed poems from the latter part of the Tang Dynasty, which treat the parting of comrades, and the opening couplet provides a conventional expression of parting:

去矣三千里送君生暮寒

Go, cover three thousand miles/ I will send you off, feeling the evening's chill.⁵

In the second couplet 頷聯 ganren (Chinese hanlian), Shiki describes the setting of Matsuyama, while the third couplet 頸聯 (keiren, Chinese jinglian) treats the loneliness that Sōseki will likely experience in the provinces, having no sympathetic companions with whom to discuss literature, and the problems he will inevitably confront when he has to deal with his unruly students 悪童 akudo famously lampooned, of course, in the novel Botchan. The poem's final couplet, less conventional and more personal than the opening couplet, expresses Shiki's desire to reunite with Sōseki at around the time of the Qingming Festival, or Tomb-Sweeping Day, traditionally celebrated fifteen days after the Spring Equinox:

清明期再開莫後晩花残

We promise to meet again at the Qingming Festival/ before the late spring's blossoms have scattered.⁶

In Natsume Sōseki's 次韻 ji'in (Chinese ciyun) verse response to Shiki's poem after arriving in Matsuyama and resuming his teaching post, lines two, four, six, and eight end with the same characters that Shiki had used in his poem. Moreover, like Shiki, Sōseki too begins his poem with a conventional shuren opening couplet that repeats the “three thousand miles” referenced in Shiki's opening couplet. Sōseki continues in the lines that follow to accentuate the distance that now separates them by means of the very modern image of a steam locomotive belching smoke into the air as it heads south. The third couplet compares the service of the public official with the very private act of returning home, and the poem's final couplet completes that self-reflective turn in which Sōseki, in apparent response to Shiki's concerns that Sōseki's talents are being wasted teaching at that provincial high school, says the following:

為君憂国易作客到家難。

三十巽還坎 功名夢半残

To serve lord and nation is easy/ but perilous is the return home for the traveler./ At age thirty, I struggle to live up to other's expectations/ my dream of attaining fame too seems likely to collapse halfway there.⁷

In this way, Sōseki uses the occasion of this very conventional exchange of poems to share with Shiki the frustration that he was feeling at his inability to pursue the literary career and the public recognition for which both writers felt destined, even as each struggled with a serious illness.

Masaoka Shiki sought to serve as a war correspondent in the Sino-Japanese War but arrived after the signing of the Treaty of Shimonoseki in April 1895. Shiki's tuberculosis was exacerbated by the severe conditions that he experienced in China, and he spent months convalescing in a hospital in Kobe upon his repatriation. In August 1896, having recovered sufficiently to travel, Masaoka Shiki returned to Matsuyama where he took up residence with Sōseki in the latter's rented home. Shiki wrote about his return in a kanshi entitled “Recuperating in My Home Village Matsuyama While Living with Natsume Sōseki.” The poem opens with the image in the first couplet of the official returning to his home village bearing both the “sword and books” 書剣 that serve as symbols of his samurai lineage. The third and fourth couplets reference the landscape of Shiki's hometown Matsuyama, a former castle town, and depict the autumn landscape. The last lines seem to be addressed to Sōseki, the companion with whom Shiki has at last been reunited:

不堪共分手夜静露華深

I cannot bear the thought of parting/ In the depths of night blossoms fill with dew.⁸

In the end, Shiki lived with Sōseki in Matsuyama for a relatively short period, and during his time there, Shiki was engrossed in haiku production with the circle of young poets that comprised the coterie known as the Shōfūkai 松風会. Several months later, Shiki, having returned to his residence in Tokyo, penned an eight-line poem based on a visit from Sōseki. In the opening couplet, Shiki alludes to the constant pain that would come to define his final years, pain that was exacerbated by the chill of winter that he describes in lines 3 and 4 of the poem. In the final pair of couplets, Shiki alludes both to the despair to which he is in danger of succumbing to as a result of his illness and to the excitement and hope occasioned by Sōseki's visit:

盧與山相接吾将世互憎

柴門聞剥啄到屐迓良朋

In the shadows where the village meets the mountain/ I struggle in mutual enmity with the world./ Hearing a clamor at the gate/ I throw on some sandals to meet a dear friend.⁹

Opportunities for Shiki and Sōseki to meet had, in fact, become increasingly rare by this point as a result of both Sōseki's marriage to Nakane Kyōko in December of the previous year and of his teaching appointment in Kumamoto in 1896. Consequently, when Sōseki was directed to study abroad in Great Britain by the Ministry of Education in 1900, he realized that Shiki, whose condition was rapidly deteriorating, might not survive until Sōseki's return. On August 26, 1900, Sōseki and a student of his named Terada Torahiko visited Shiki on his sickbed. Shiki wrote about that meeting in an essay that appeared in Hototogisu (“Sōseki's Life” 2015). Sōseki, for his part, produced a kanshi to Shiki that turned out to be the ultimate poem of parting. The poem begins with the couplet:

君病風流謝俗紛、吾愚牢落失鴻群

In your illness, you have withdrawn from the world's entanglements./ Irresponsibly, I prepare to fly off like a lone goose.¹⁰

In the second and third couplets that follow, Sōseki's poem alludes to Shiki's diminishing health and to the most damning symptom, his vomiting blood, just as, the speaker notes ironically, Shiki is creating some of his most important work as a poet. The poem acknowledges the mixed emotions that Sōseki harbors as he prepares to depart for his study abroad experience, which will entail leaving friends such as Shiki behind. The poem ends with the final couplet 尾聯 oren (Chinese weilian), which picks up the metaphor of the migrating lone goose established in the opening couplet:

詩成投筆蹣跚起 此去西天多白雲

Having completed a poem, I toss aside my writing brush and rise unsteadily./ Preparing to head West, I can but gaze at the billowing white clouds.¹¹

This was to be the final kanshi poem shared between the two writers as part of the long exchange of poems conducted in conjunction with their correspondence. In a real sense, the relationship between these two writers had long been diffused through the filter of kanshi and kanbun. The feelings expressed in this body of poems, alternately playful and melancholy, personal and empathetic, were articulated in a language that was for both writers a foreign language and yet even better suited as the optimal means of communicating true feelings than their mother tongue. If Japanese was unarguably Masaoka Shiki's and Natsume Sōseki's mother tongue, then classical Chinese was indeed a “sacred silent language,” suited to the gravity of this exchange that was at once personal and elegantly formal, and it was through this sacred language that the two writers' friendship was sanctified.

The meaning of Kanshi for Shiki and Sōseki

For both Masaoka Shiki and Natsumei Sōseki, at this critical early stage of their respective careers as writers, the exchange of kanshi served an important function and contributed to their development as writers. Just as each writer would go in different literary directions that would profoundly affect the course of modern Japanese literature, so too kanshi production bore a different meaning for each writer. For both writers, the exchange of kanshi poetry served as a vehicle to demonstrate a reverence for tradition and as a means of establishing their credentials as full participants in the final generation in which the Sinosphere still held sway in Japan. Moreover, this return to the Kanbunmyaku constituted a kind of counternarrative to the dominant Meiji narrative of Bunmei kaika.

Nevertheless, the meaning of kanshi for Sōseki differed from that of Shiki in some subtle but important ways. In the sixty-seven kanshi poems that Sōseki penned during his eleven-year correspondence with Shiki, Sōseki treated a variety of themes including returning to the country, the longing for departed friends, struggles with illness, and the parting from loved ones. Xu Qian has suggested that it is no exaggeration to say that Sōseki had one reader in mind when he composed his kanshi during this period – Masaoka Shiki (Xu 2005, p. 31). Shiki, on the other hand, was a much more prolific writer of kanshi than was Sōseki and wrote hundreds of kanshi during this same period for a wider audience and often with publication in mind.

Masaoka Shiki and Natsume Sōseki were motivated to compose kanshi poetry in large measure for the same reason that motivated kanshi poets from time immemorial, the ethical cultivation of the scholar-official who sought to find a balance between the private pursuit of aesthetic satisfaction and the public pursuit of service to society and to the state. In Dynastic China, particularly in times of turmoil or change, the literati class recognized that a return to the pursuit of the literary and calligraphic arts and to the teachings of the ancients had the power to transform society, and poetry functioned as a tool for personal cultivation and social amelioration. By the Northern Song Dynasty (960–1127), scholar-officials had turned increasingly to the arts, including poetry, painting, and music, as the proper path for moral self-cultivation (Department of Asian Art 2004). Masaoka Shiki and Natsume Sōseki were conscious heirs to that tradition. In fact, the poems by both Shiki and Sōseki in that eleven-year period of exchange betray the tension between the Confucian life of service and the Daoist retreat from the tyranny of duty to the meditative life in a rural setting that constitutes a key thematic concern of the literature in the Sinosphere. The recognition that the production and exchange of poetry is not merely the elegant and affected pastime of the cultured but is part and parcel of the moral cultivation of the Confucian official, constituted a tacit belief for Shiki and Sōseki.

For the Confucian scholar, as Anderer (2002, p. 40) points out in his recent study of Sōseki, the past itself is a teacher, and ethical self-cultivation is a critical dimension of the examined life. As a writer, Sōseki continued to embrace Neo-Confucian values throughout his life and saw his role as writer in largely didactic terms, serving as the social conscience of his generation. Due to his youthful engagement with the Kanbunmyaku and his later mastery of English and engagement with Western cultural values, Sōseki ended up “shuttling between cultures, neither of which could offer him a stable, welcoming home to weather the storms of change of modernity” (Ibid., p. 41). DeBary (1993, p. 10), in the introduction to his translation of Karatani Kōjin's The Origins of Modern Japanese Literature, contends that despite having received a classical Chinese education, by the time that Sōseki was a mature writer, he was “already engulfed by modernity in a way that made it impossible for him to retrieve any pristine and direct knowledge of the past.” Sōseki approached both Chinese and Japanese literature as something innate; Western literatures existed outside of that affinity.

Conversely, Masaoka Shiki's kanshi poems, even the ones penned to Sōseki and others in personal correspondence, were meant for public consumption and were based on the same close observation of phenomenal reality and the same aesthetic approach that he utilized in producing his haiku, waka, and sketches. For Shiki, all genuine sentiment in poetry was based on observing phenomena from the perspective of shasei 写生 or “sketching from life” (Houwen 2020, p. 479). Shiki was influenced by the Japanese Western-style painter Nakamura Fusetsu 中村不折 (1866–1943) in his attempt to capture minute elements of nature “as is” in his poems based on direct observation without the intercession of extraneous aesthetic or cultural influences (Watson 1997, p. 7). Nevertheless, his declared goals notwithstanding, Shiki inevitably allowed some room for the poetic imagination and of maintaining ties with poetic tradition in the act of capturing reality. The adoption of personae from the kanshi tradition represents one such accommodation.

In other words, Masaoka Shiki's kanshi served the same overarching goals as his other forms of artistic expression in focusing on genuine emotional responses to external stimuli. Shiki's goal was nothing less than haiku reform, and although this same sense of reformist mission did not extend to kanshi, he brought to kanshi composition the same approach of representing observed reality “as is.” The following poem illustrates Masaoka Shiki's approach to poetic composition. Ironically, even ostensibly when penning poems that treated historical events as in the following kanshi from 1892, which treated the legacy of the general Kisō Yoshinaka 木曾義仲 (1153–1184) of the Minamoto Clan, Shiki remained true to the goal of the shasei approach of close observation:

群峰如劍刺蒼空、路入歧蘇形勝雄。

古寺鍾傳層樹外、絕崖路斷亂雲中。

百年豪傑荒苔紫、万里河山落日紅。

欲問虎拏龍闘跡、蕭蕭驛馬獨嘶風。

Mountain peaks stab the heavens like swords,/ where the road enters Kiso/ the bell of the old temple tolls beyond the layers of forest/ and the cliff road rises to the clouds./ The old stele is covered by the moss/ and the long rivers and mountains are made red by the sun./ If you want to ask about the results of the duel between tiger and dragon/ Simply listen to the neighing of the abandoned horse (as quoted in Zhou 2006, p. 76.)

Although Sōseki did not share that same sense of destiny to reform either Japanese or Chinese poetry, he did share with Shiki a capacity, unusual among Japanese writers of any period, to see kanshi as a living, flexible form rather than as an ossified mode of expression. In Sōseki's case, this led him to treat kanshi as a lens with which to scrutinize the intellectual life of the Meiji period and to reflect on his own attitudes toward modernity.

Ironically, this function of kanshi also can be seen in Sōseki's fiction. The protagonist in Natsume Sōseki's¹² novel Kusamakura 草枕 “Grass Pillow” is a painter of landscapes who spends much of his time composing kanshi while staying at a country inn where he has come in order to pursue plein air paintings. Rather than using these encounters with bucolic scenes as the inspiration to paint that which he witnesses with his own eyes, the artist paints versions of nature that he encounters in the poems of Tao Yuanming 陶淵明 (365–427) and Wang Wei 王維 (699–759). The background to many of Sōseki's kanshi is the desire to return to the country, a propensity that reflects the thematic concerns of the long poetic tradition of dynastic China in the longing to overcome the attachments to the world (Xu 2005, p. 195). Subsequently, the artist's observations are filtered through the poetic imagination and through cultural tradition rather than being based on direct observation. Unlike Shiki, Sōseki's approach to kanshi, like that of the protagonist of Kusamakura, was to respond to tradition in a mediated dialogue with his cultural predecessors, the “strong poets” of Chinese poetic history, to appropriate a term from Harold Bloom, who served as the final measure of literary achievement in the Sinosphere (Bloom 1997, pp. 149–150).

Natsume Sōseki was characteristically candid in his assessment of what he saw as the constrictiveness of the shasei approach, which eschewed the cultural context in its attempt to produce art directly from nature free from the broad shadow of cultural predecessors. Ueda Makoto, in an essay about Natsume Sōseki's literary approach, includes a quote from Sōseki in which the writer says, “The artist finds himself depicting neither nature nor the mind that projects nature, but rather his own interpretation of it. By ‘his own interpretation’ I mean the way in which he gives meaning to nature, the way in which he views nature” (as quoted in Ueda 1976, p. 235). Ultimately, neither Shiki nor Sōseki, in their kanshi exchange, elicited any discomfort at the apparent tension between the utilization of personae drawn from the tradition of kanshi exchange and the expectations of shasei “authenticity.”

Like the protagonist in Kusamakura, Soseki's literary approach involved channeling memory into an aesthetic re-creation of direct experience. In fact, in Kusamakura, Soseki incorporated two of his poems, “Shunkyo”春興 “Enjoying Spring” and “Shunjitsu Seiza” 春日靜坐 “Spring Meditation”, that had appeared in the Meiji Kanshibunshū 明治漢詩文集 anthology. “Shunkyo” appears with an introductory paragraph:

Once again I felt the inspiration to write, and lying on the grass, I began to arrange my ideas. I wrote down every line in my sketchbook as it come to me, and eventually when I felt I had done all I could, I read them through from the beginning. (Natsume 1990, p. 167)¹³

Moreover, Sōseki's own approach to the shasei sketch-from-life method, an approach that acknowledges the cultural background that gave rise to the poem, can be inferred from the following exclamation from the novel's protagonist:

“Done it! I've done it!” The words escaped from me with contented sigh. This is what I had been waiting to write. These lines exactly expressed my oblivion to the world while I had been lying gazing at the quince blossoms. It did not matter that there was no mention of the blossoms themselves, or of the sea: it was enough that the poem expressed what I felt. (Ibid., p. 167)

In Sōseki's kanshi as in the protagonist's plein-air paintings, the natural world, the unmediated encounter with nature, provides the necessary context for a psychological response to the scene and a dialogue with his predecessors in the poetic tradition.

Natsume Sōseki's approach to kanshi composition reflected a similar commitment to careful observation and a close attention to sensory details as did Masaoka Shiki's but simultaneously demonstrated a propensity to both situate that observation in the cultural context of the literary tradition and to reveal the psychological response of the poet to the observed phenomenon. In the following poem, for example, called “Yamaji kanpū” 山路観楓 “Beholding Maples on a Mountain Road” the poet engages with both the immediate setting but also with the traditional associations of the scene:

石苔沐雨滑難攀 渡水穿林往又還

處處鹿聲尋不得 白雲紅葉滿千山

The moss that coats the stone becomes slick with rain/ Making it difficult to navigate.

I ford the river, pass the forest and come back again./ Here and there and I hear deer but can't see them./ White clouds and bright red maples suffuse the mountains. (Natsume 1993b, p. 142)

The poem both celebrates a particular moment and a specific encounter but also situates itself in the tradition of such poetic encounters through its reference to Wang Wei's celebrated “Lu zhai” 鹿柴 “Deer Park” poem which evokes a similar aural image.

At the same time, in the context of Kusamakura and independently of his fiction, kanshi served as a means for Sōseki of criticizing contemporary society. As the very embodiment of traditional expression in East Asia, kanshi constituted an ironic but ultimately effective tool for scrutinizing contemporary Meiji society. Moreover, kanshi not only provided a mechanism for social critique, it also complemented the critical voice cultivated in Sōseki's fiction. Saitō Mareshi notes, for example, that not only does kanshi provide a vehicle for social commentary in Kusamakura, but in the 1908 novel Gubijinsō 虞美人草 “The Poppy” Sōseki utilizes kanbun rhetorical structures in the narrative (Saitō 2017, p. 8). However, after 1908's Sanshirō 三四郎, Sōseki's novels no longer would reflect the rhetorical influence of Kanbunkyaku though they would retain the critical stance cultivated in those earlier novels.

The kanshi written in Natsume Sōseki's final years, when he was working on Meian 明暗 “Light and Darkness,” adopt a similarly critical posture viz-a-viz contemporary Japanese society. By this juncture, kanshi is no longer the rhetorical device that it was in Kusamakura, but its composition continues to provide Sōseki with a way to problematize the life of the intellectual in early twentieth-century Japan in conjunction with parallel critique conducted in the novels (Ibid., p. 9). Unlike the social criticism carried out in the novels though, the critique of Meiji and Taishō Japan in the kanshi was primarily for self-consumption and was shared only with a small circle of intimates. In this manner, the social commentary articulated in Sōseki's kanshi served as a testing ground for the critique in his widely consumed fiction.

For Masaoka Shiki, and to an even greater degree for Natsume Sōseki, the production and exchange of kanshi also became a way to contest Meiji Japan's wholesale and often impetuous pursuit of Western culture under the guise of Bunmei kaika. Toward the end of his life, Shiki turned away from kanshi production in order to concentrate on tanka and haiku production and theory. This turn may have been spurred in part by Shiki's decision to spend his remaining days consolidating his legacy in the field of Japanese poetry reform, but it may also have been inspired by the decline in popularity and influence of kanshi as evidenced by the disappearance of the kanshi column in some of the major daily newspapers. Natsume Sōseki, on the other hand, continued to write kanshi until the end of his life, maintaining a disciplined regimen of daily kanshi composition in his final years.

Natsume Sōseki maintained via kanshi a running dialogue with Shiki that lasted eleven years, but kanshi also simultaneously constituted for Sōseki a nearly lifelong monologue. Throughout his life, Sōseki used both haiku and kanshi as means of self-reflection and as a mechanism for self-cultivation (Katō 2007, p. 20). Ironically, while leveraging the tendency toward self-examination in his kanshi, Sōseki simultaneously debated the meaning of modern institutions and values. Sōseki, whose experience of studying abroad in London was formative to his mature thought and literary themes, was sensitive to the dangers of what he perceived as the Meiji period's simplistic equation of modernity with the West's experience with modernization. Sōseki, of course, was not alone among Japanese thinkers of his generation in recognizing the complexities of modernization and in cautioning a more measured approach to embracing Western institutions and practices, and in a lecture entitled “Gendai Nihon no kaika” 現代の本の開花 “The Blossoming of Contemporary Japan” delivered in August 1911 in Wakayama Prefecture as part of a lecture tour of the Kansai region and published in 1913 in essay form, Sōseki condemned the vacuity which, in his view, characterized Japan's experience of modernization (Ibid., p. 11). Sōseki portrayed Meiji Japan as a society losing its cultural bearings due to its reckless adoption of everything that smelled of the West.

The sense of melancholy or 愁 (urei) occasioned by that recognition colored Sōseki's attitude toward life and is reflected in both his fiction and in the tone that defines his kanshi as demonstrated by the frequency of the terms 愁 and the variant 憂 in his kanshi throughout his life. On the one hand, this melancholy reflected Sōseki's disposition and hinted at his lifelong struggles with depression, while on the other hand, this melancholy was a symptom of Sōseki's revulsion at the emptiness of Meiji modernity (Ibid., p. 21). For Sōseki, kanshi offered a poignant way to convey this sense of loss resulting from the cultural associations of poetry with this ineffable sense of melancholy in the Chinese poetic tradition (Ibid., p. 25).¹⁴

In an untitled poem composed in 1899, just prior to his study abroad experience in London, Sōseki contemplates this tension between Eastern and Western culture and between tradition and modernity in the opening couplet:

眼識東西字 心抱今古憂

My eyes comprehend the characters for East and West,/ My heart embraces the melancholy of both past and present.¹⁵

The same poem ends with the final pair of couplets:

鳥入雲無迹魚行水自流

人間固無事白雲自悠悠

The bird enters the cloud without a trace./ After the fish passes through, the water flows on unperturbed./ Humankind too was originally untroubled.

White clouds float away unobstructed.¹⁶

Scholars of Natsume Sōseki's poetry have noted that in Sōseki's kanshi the image of the white cloud 白雲 is often a metaphor for this sense of isolation, as in the final couplet from the above poem (Katō 2012, p. 291). As Watanabe Hideki (2013, p. 291) has commented, the white cloud is an image associated in Chinese literature with Zhuangzi and the Daoist escape from the trials of the world. In this case, perhaps the image alludes to Sōseki's frustration with the increasingly ponderous duties that occupied him as an educator in Kumamoto. The poem strikes a tone of inevitability concerning the tension between East and West and between tradition and modernity that seems to reflect the speaker's struggle to reconcile these differences just as Sōseki prepared to confront the unknown in London. The metaphors of the bird being swallowed by the cloud and of the fish leaving no disturbance in its wake convey Sōseki's own lack of confidence in his capacity to contribute to any substantive cultural change, and this, in turn, results in the melancholy mood embodied in the lone, drifting cloud.

In Sōseki's 1905 essay “Danpen” 断片 “Fragments” from 1905, complete capitulation to the West and its values constitutes an irrevocable loss or abandonment of tradition. In that essay and elsewhere, Sōseki often equated modernity with 亡び horobi “destruction,” a term that he uses in the same essay in relation to the decline of culture in the West (Katō 2007, p. 12). Many commentators on the style cultivated in Sōseki's kanshi speak of his command of the aesthetic style embodied in the term fūryū, evincing a refined elegance in expression and in deportment (Makizumi 2013, p. 117). Along with managing the tropes implicit in this aesthetic approach, a command of fūryū also anticipates the poet's ability to express these conventions in fresh and innovative ways. Sōseki is regarded as one of the few modern Japanese poets of kanshi to meet the lofty expectations of fūryū composition.

When Sōseki was writing his final novel Meian, he would work on the novel in the morning and compose kanshi in the afternoon, a process documented in a manuscript of Sōseki's from his final years called Kanshi nōto 漢詩ノート “Kanshi Notes.” Writing fiction was enervating for Sōseki, and he struggled until the end to meet the strict editorial deadlines imposed on him; writing kanshi was for Sōseki therapeutic and meditative, and he was able to find respite in kanshi composition each afternoon in his final years (Watanabe 2013, p. 86). Kanshi Notes describes Sōseki's process for composing kanshi, which would start with an initial composition and then lead to revisions on subsequent days. Only when he had finished his revisions would Sōseki mark “X” over individual kanshi indicating that they were complete (Ikkai 1996, p. 35).

The poems from this period are often inspired, even to a greater extent than his earlier poems, by Zen Buddhism, and they tend to treat themes and to employ imagery and vocabulary items that evoke both Zen satori and Daoist mindfulness. Some of the vocabulary items in these poems were drawn from such celebrated Zen poets as the Tang Dynasty poet Han Shan寒山 (Known in the West as Cold Mountain, the English equivalent of his name) for whom Sōseki had a special affinity; others seem to be drawn from the Five Mountain poetic tradition in Japan including such terms as kūhen 空偏 “side of the sky,” which appear to be unique to the Japanese kanshi tradition (Yoshikawa 1985, p. 133). In a letter from the final year of his life written to the Zen monk Tomizawa Keidō 富澤敬道, Sōseki speaks of the Zen-inspired kanshi that he is writing, “Strange to say, but dullard that I am, it is only now, in my fiftieth year, that I have found a path that I truly wish to blaze” (Satō 2020, p. 206).

Moreover, on the interpersonal level, kanshi served as the means to establish authenticity in Shiki's relationship with Sōseki. When Shiki first met Sōseki, he assumed that Sōseki, clearly more familiar with Western literature than he was, would be enthralled with Western culture like so many of their generation to the exclusion of any real affinity with Chinese literature. When Shiki read the poems included in Bokusetsuroku, he was astonished to learn of Sōseki's keen familiarity with kanshi conventions and of his facility with kanshi composition (Morioka 2015, p. 168). In sharing his collection Nankusashū with Sōseki for critique later that year, Shiki was inviting this fellow literary aspirant and likeminded Sinophile to a greater level of intimacy. This affinity would serve as the cornerstone of the two men's friendship thereafter.

Conclusion: the last generation of Kangaku Seinen

Masaoka Shiki at last succumbed to the tuberculosis with which he had long battled, dying on September 19, 1902, at the age of 34. Natsume Sōseki, who was in the midst of his studies in London when Shiki died, continued to compose kanshi until his own death in 1916, though with Shiki's passing an important element of Sōseki's kanshi output – the changhe exchange with a kindred companion – was inexorably lost. Literary luminaries such as Shiki, Sōseki, and Mori Ogai 森鴎外 (1862–1922), along with Meiji era politicians such as Itō Hirobumi 伊藤博文 (1841–1909) and military leaders including Nogi Maresuki 乃木保典 (1881–1904), and Kangakusha scholars such as Kubo Tenzui 久保天随 (1875–1934), are often cited as the last generation of Japanese writers who were the product of a traditional kangaku education (Katō 2006, p. 211). Hedberg and other commentators warn against a facile interpretation of the status of kangaku in the Meiji period as merely an “episteme under siege.” Although it is undeniable that the status of kanbun composition was dramatically altered as a result of the genbunitchi movement and of the recasting of Japanese as a modern language, there were some Meiji intellectuals who saw kangaku as an integral dimension of modernization (Hedberg 2020, p. 4).

For those born prior to the Meiji Restoration, a facility with kanbun and kanshi still retained a significant cultural cache. In that sense, these writers and others of their generation can be thought of as the last generation of kangaku seinen 漢学青年, youth who were the product of a traditional Chinese education, for whom classical Chinese was a “sacred written language” and a literary touchstone to which they could always return for grounding. According to most commentators, Sōseki and Shiki were by disposition quite different; kanshi served as a means to smooth over those differences and allowed both writers to express views on tradition and on contemporary society and even on events from their own lives in a way that was sympatico.

Sōseki and Shiki were born at the close of the Edo Period, which is generally recognized as the second Golden Age of kanbungaku, second only to the efflorescence of kanbun in the Nara and Heian periods, and these two writers were of that generation that carried the torch of that tradition into the Meiji period. One significant development in Meiji period kanbun was the introduction of futsūbun 普通文, an easier mode of composing kanbun based on vernacular Japanese that widened the appeal of kanbun (Ibid., p. 211). That development notwithstanding, the fall in popularity of kanbun in the Taishō period was precipitous. The Taishō Emperor is said to have been the last Emperor to possess a true facility with kanshi, and he left behind a body of nearly 1000 kanshi poems (Ibid., p. 218). Morioka (2015, p. 168), in her study of modern Japanese kanshi, posits that whereas traditionally in Japan kangaku encompassed all of learning, by the Taishō period it had come to encompass one sliver of very specialized learning. Kanbun had become “detached” from the everyday life of Japanese writers and scholars, and the appellation shijin 詩人 “poet” came to denote only those who wrote poetry in Japanese, including new style shintaishi 新体詩poetry. Although there was a slight revival of kanbun and kanshi exchange following the annexation of Taiwan and Korea that resulted in cultural exchange between writers in East Asia, that exchange proved to be short-lived.

Consequently, by the Shōwa period, kangakusha no longer functioned as cultural guides, opinion leaders, or political advisors as they had in the Meiji period. Although some skilled kanshi poets remained even into the Shōwa period, their poems were little more than “accessories” in the words of one commentator and did not serve the same intellectual and cultural functions as had the productions by kanshi poets in earlier ages (Katō 2006, p. 221). The level of kanbun and kanshi proficiency in Japan only further deteriorated in the postwar era but continued to survive among a small population of dedicated enthusiasts, including popular shigin 詩吟 recitation groups and in nearly symbolic form as an element of both compulsory kokugo 国語 “national language studies” courses and of the high school and university entrance examinations. There have been several resurgences of interest in kanbun in the postwar period triggered by such events as the normalization of relations between Japan and the People's Republic of China in 1972 (Ibid., p. 224). Finally, there remain, of course, a very small number of contemporary Japanese who continue to take pleasure and perhaps find comfort in reading and composing kanshi and who dutifully tune in each week to NHK's long running kanshi radio series.

Wright (2008, p. 38), in her study of exophonic literature, argues that writers who choose to express themselves in a language other than their native language should not be seen as oddities or somehow deficient users of a second language, but should be recognized for the innovations that they often bring to written expression and for the very real risks that they bear in choosing to communicate in a borrowed tongue. Similarly, the novelist Murakami Haruki has said of writing in English: “the limitations of writing in a foreign language remove the obstacles one faces in trying to write in one's own language.”¹⁷

What Murakami said of English could be applied to Sōseki's and Shiki's use of classical Chinese as well. That Natsume Sōseki and Masaoka Shiki were able to mimic so deftly the voices of traditional Chinese poetic expression and to inhabit the personae of changhe exchange is a tribute to the two writers' intimate understanding of the concerns and conventions of the Chinese poetic tradition but also to their facility with the performative features of rakugo that they witnessed as students in the yose theater. For Natsume Sōseki and Masaoka Shiki, this immersion in the traditional personae of kanshi poetic exchange served several purposes: it at once allowed these two young intellectuals to showcase their familiarity with the traditions of the Kanbunmyaku while also encouraging them to experiment as writers. Writing in Classical Chinese allowed Shiki and Sōseki to express themselves in a literary form that complemented their writing in Japanese. It also ironically allowed Sōseki and Shiki to express their emotional lives to one another with a candor that may have been more difficult in Japanese.

Fusao contends that Natsume Sōseki saw the writing of fiction as the highest and truest form of literary production. Kanshi, on the other hand, provided Sōseki with a means of commenting on life's events, and it offered him therapeutic value by serving as a tool to interrogate his own stance toward modernity (Shimizu 1996, p. 4). Both Sōseki and Shiki, like other intellectuals of their age, had one foot each in two worlds: they were at once both Edo-born children of the Kanbunmyaku, who clung to the age's intellectual preoccupations, but they were also the offspring of Meiji “Civilization and Enlightenment” entrusted with nothing short of the wholesale reformation of the intellectual and cultural life of Japan (Watanabe 2013, p. 86). Kanshi offered for Sōseki and Shiki, as it did for Mori Ōgai and others of their generation, a way both to celebrate indissoluble ties with the past and with an imagined community in the Sinosphere, and to communicate with associates in a very public argot that was simultaneously a sacred language and a language paradoxically more familiar, more comforting, than their mother tongue.