TOKUGAWA APPROACHES TO THE RITUALS OF ZHOU: THE LATE MITO SCHOOL AND “FEUDALISM”
Kate Wildman Nakai, Sophia University
From early times government officials and scholars in Japan looked to the Rituals of Zhou (Zhouli 周禮) as a source of information about Zhou political institutions. The early-eighth-century governmental codes, modeled after those of the Tang dynasty (618–907), designated it as one of the seven Classics to be studied at the court university (daigakuryō 大學寮) and as one of the possible subjects for the ex
amination in classical learning (myōgyōka 明經科).1 Analysis of ref erences to Chinese works in writings from the eighth and ninth centu ries—the Nara period (710–794) and the first century of the Heian period (794–1185)—has shown it to be one of the most widely cited of the Classics in this era when the Chinese-inspired codes continued largely to define the governmental framework.2
By and large, however, people both then and later seem to have re garded the Rituals of Zhou primarily as a compendium of institutional information; unlike in China, it did not invite repeated debates over its authenticity, interpretation, and relevance. As Uno Seiichi notes in his definitive study of the Chinese debates, this was true even in the To kugawa period (1603–1868), the high point of Confucian commitment and studies in Japan.3 Compared with the intervening late Heian and medieval periods, when the Rituals of Zhou seems to have received
1 Ritsuryō 律令 (Nihon shisō taikei 3.263–65, 300–301); Wajima 1965, 1–4. 2 Kojima Noriyuki indicates that an early Heian compilation of then-extant com mentaries on the codes includes upward of seventy references to the Rituals of Zhou, in contrast to fifty or fewer for each of the other Classics. See Kojima 1968, 254–333. I am indebted to David Lurie for this reference. Given the institutional orientation of both the codes and the Rituals of Zhou, the relative weight of the latter in this context is perhaps only to be expected. In other Nara and early Heian works, such as imperial edicts or histories, references to the Rituals of Zhou are proportionally fewer. See Uchino 1991, 125–42. 3 Uno 1949, 17.
280
relatively little attention, even by specialists in Confucian learning,4 Tokugawa scholars cited it quite frequently, but such citations tended to be of an incidental nature. The Rituals of Zhou functioned for them largely as an authoritative reference work, so to speak, and for the most part they did not engage in critical analysis of the text or inter pret it prescriptively.5
We can see a good example of this attitude in A Survey of Institu tions (Seidotsū 制度通), a comparative topical survey of the history of governmental and other institutions in China and Japan compiled by the mid-Edo scholar Itō Tōgai 伊藤東涯 (1670–1736). Tōgai’s father, Itō Jinsai 仁斎 (1627–1705), is known for having undertaken a textual critique of several classical works and for having reached the bold conclusion that some sections of the Doctrine of the Mean (Zhongyong 中庸), revered by Song scholars, were not authentic and that the Great Learning (Daxue 大学) diverged substantially from the teachings of Confucius. Coming from this scholarly tradition, Tōgai noted that “it is difficult to be sure” that the Rituals of Zhou was, as it was asserted to be, the work of the Duke of Zhou. He was content to conclude, nevertheless, that one could surmise from it the general out lines of Zhou institutions, and he cited it repeatedly regarding the Zhou situation.6
4 The presumption of less attention to the Rituals of Zhou in the late Heian and me dieval periods is necessarily highly speculative. Wajima Yoshio’s 和島芳男 discus sion of references to Confucian texts appearing in courtier diaries and other records indicates greater interest in works such as the Zuo Commentary on the Springs and Autumns Annals (Zuozhuan 左傳) and the Records of Ritual (Liji 禮記); see Wajima 1965, but any firm conclusion would require close consideration of materials like law codes. 5 The comprehensive catalogue of premodern works Kokusho sōmokuroku 國書總 目錄 and its several digital offshoots list various titles dating from the Tokugawa period (primarily the latter half) that indicate some mode of commentary on or discus sion of the Rituals of Zhou. Almost none of these are published works, and, more telling, for only a few are the editors able to indicate the location of a known extant manuscript version. The vast majority of the Rituals of Zhou–related entries are based simply on the mention of such titles in other bibliographical compilations. To be sure, the Rituals of Zhou is not an isolated instance in this regard. Tokugawa scholars pro duced commentaries on the Analects and a few other core texts, but by and large they tended to rely on the established Chinese commentaries on the Classics, many of which were reprinted in Japan. 6 Seidotsū 1.134. We see a similar situation in the writings of Yamaga Sokō 山鹿 素行 (1622–1685), who in discussing various historical approaches to matters of gov ernment and administration quotes extensively from the Rituals of Zhou as a source of information about Zhou practices on topics ranging from administration of the female quarters of the ruler’s palace to rewards and punishments and the regulation of agri-
TOKUGAWA APPROACHES TO THE ZHOULI 281
As far as wider circulation and broader knowledge of the text was concerned, it is perhaps safe to say that the Rituals of Zhou was held to be a text for advanced students rather than basic to a general educa tion. By the end of the Tokugawa period, a large number of daimyo domains had established academies for the education of domain re tainers (and in some cases commoners), and a number of these acade mies printed inexpensive editions of key texts for use by their students. Various domains issued editions of the Four Books (Sishu 四書) and Five Classics (Wujing 五經) and of texts such as Zhu Xi’s 朱熹 (1130–1200) explication of the Ceremonial Rituals (Yili 儀禮). Ac cording to Kasai Sukeharu’s research on domain editions, however, only one such edition of the Rituals of Zhou was published. Although the evidence regarding curriculum is less comprehensive, it, too, sug gests that domain academies tended to reserve systematic study of the Rituals of Zhou for upper levels of students.7
Within this general environment, some thinkers nevertheless en gaged more intensively and dynamically with the Rituals of Zhou. They were drawn to it in considerable measure because they saw the text, held to explicate the Zhou feudal (fengjian 封建; Japanese: hōken) system, to be directly pertinent to the Tokugawa polity. Al though in the seventh century Japanese rulers, following the model of the post-Qin–Han Chinese imperial bureaucratic state, had adopted a centralized “commanderies and prefectures” 郡縣 (Japanese: gunken) governmental structure, the Tokugawa bakufu-daimyo domain system that eventually evolved out of the decay of the imperial state likewise could be taken to rest on basically feudal principles. A number of thinkers took the feudal model of the Rituals of Zhou to be relevant to the situation of the samurai class in particular. Embedded in these premises, however, was more than one level of irony. First, although one might assume that the Tokugawa thinkers would hold up the samurai tradition as realizing the ideals of the Zhou system, such was not the case. Some asserted that firsthand experience of a feudal order gave them special insight into the nature of the system described in the Rituals of Zhou, but they also were critical of many aspects of Japa
nese samurai society, both past and present, and they looked to the Rituals of Zhou as offering a corrective to just those ills. Second,
cultural and commercial life. See Yamaga gorui 山鹿語類 4.118–119, 273–277, 289– 466. I thank James McMullen for drawing my attention to Sokō’s citation of the Ritu als of Zhou. 7 Kasai 1962, 135; Kasai 1960, 248.
282
while they hailed the text as exemplifying a “true” feudal system, the schematic apparatus it set out in actuality pointed in significant ways toward centripetal mechanisms of government, particularly as regards the integration of lower levels of administration with higher. Third, it was precisely this dimension of the text that seems to have attracted such thinkers. Their engagement with the text as a source of informa
tion about “feudal” methods of rule both veiled and encouraged what might be seen more accurately as an inclination in a quite different direction.
The thinkers to show the strongest affinity of this sort for the Ritu als of Zhou can be linked more or less to the school of Ogyū Sorai 荻 生徂徠 (1666–1728), although Sorai himself notably showed consid erable ambivalence about the text. Below, after briefly addressing
Sorai’s perspective, we will take up the more positive views of his disciple Dazai Shundai 太宰春臺 (1680–1747) and then turn to the group for whom the text assumed the largest significance, the late Mito 水戸 school active in the last decades of the Tokugawa regime.
The Views of Ogyū Sorai and Dazai Shundai
Sorai is known for his trenchant critiques of contemporary Tokugawa society, on the one hand, and of the introspective outlook, metaphys ics, and textual approaches of Song-Ming Confucianism, on the other. In his challenge to the premises of Song-Ming thought, he developed a distinctive interpretation of the Way as not immanent in either the universe or human nature; the Way, he asserted, was no more and no less than the sum of the institutions and rites created by the early kings of ancient China. As for Tokugawa society, he saw some of its most fundamental problems as arising from the removal of the samurai class from the countryside and its concentration in urban centers.
Few measures had left as definitive an imprint on the shape of To kugawa society as the demarcation of the warrior class from the agri cultural population and the removal of samurai from the countryside to castle towns that sprung up throughout the nation. Carried out from the end of the sixteenth to the early seventeenth century as part of the process of national unification following an extended period of civil war, these measures had fundamentally transformed the samurai from landed proprietors with their own holdings, drawing their income from the agricultural population who tilled those holdings, into urban resi-
TOKUGAWA APPROACHES TO THE ZHOULI 283
dents living off stipends paid out of their lord’s storehouses. With the exception of the daimyo and a few high-ranking vassals of the daimyo and shogun, the majority of Tokugawa samurai were thus a feudal class in name only. Objectively speaking, this transformation of the samurai was one of the factors that ensured the Tokugawa peace. In the eyes of various thinkers of samurai origin, however—among them Sorai—it had brought a host of ills in its wake. Sorai saw the removal of the samurai class from the land as having undermined its initiative and martial readiness. Castle town residence, which he likened as be ing akin to “living as in an inn,” was, he believed, a source of the per
ennial inability of samurai to live within their means, despite the fact that a substantial portion of the nation’s agricultural production was devoted to their support, placing a heavy burden on the agricultural populace. The absence of the samurai from the countryside had also left a vacuum that was an invitation to lawlessness and disorder as well as a source of distrust between rulers and ruled. Samurai adminis
trators, who no longer had direct knowledge of conditions in the vil lages, came to look upon the peasantry simply as a source from which to extract taxes. Peasants, for their part, thought only to find ways to evade such requisitions. Sorai saw these problems as resembling those that had arisen in China from the Qin (221–206 BCE) on as a result of abandoning feudal methods of rule in favor of the centralized bureau cratic “commanderies and prefectures” system. His proposed solution was to restore a more thoroughgoing feudal structure by returning the samurai to the land, a step that, he believed, would also have the posi tive effect of encouraging peasants and other elements of society to
remain in place, thereby leading to greater social stability.8 Given Sorai’s rejection of the Song-Ming emphasis on introspec tion and self-cultivation, his “institutional” conception of the Way, and his favoring of feudal (fengjian 封建) approaches to governance, one would expect him to take an interest in the Rituals of Zhou. Rec ognizing the text as a record of “the system of offices of the Zhou dy nasty,”9 he cited it in various contexts as a source of information about Zhou practices and terminology. In his discursive commentary on the Analects, Clarifications of the Analects (Rongochō 論語徴), he re ferred to it, for instance, to counter what he held to be the faulty un-
8 Tōmonsho 答問書 (Ogyū Sorai zenshū[a] 1.447); Seidan 政談 (Nihon shisō tai kei 36.299–300); Taiheisaku 太平策 (Nihon shisō taikei 36.480–481); Yamashita 1994, 63–64; Lidin 1999, 129–131. 9 Keishishi yōran 經子史要覽 (Ogyū Sorai zenshū[a] 1.511).
284
derstanding of Zhou institutions and rites shown by Zhu Xi and other commentators.10 Elsewhere, he adduced it as evidence that the sages regarded “military affairs as one crucial aspect of governing and bringing order to the country.” The Rituals of Zhou describes the or ganization of military activities as coming under the jurisdiction of the minister of war, he observed; it also indicates that at times of war, the six ministers in charge of supervising the overall administration of the state would become the generals in charge of the six armies. Such measures showed that military matters were a core element of the Way of the sages.11
But Sorai did not elaborate upon the nature of the governmental system laid out in the Rituals of Zhou or point to it as a model for his own time. One reason for this was probably his basic stance regarding study of the Classics. He saw a well-founded understanding of the methods of the early kings as essential to making appropriate propos
als regarding the present, but he also believed that, because people were “living entities” (katsubutsu 活物), human society and its prod ucts inevitably were subject to change. The institutions formulated in ancient China, admirable as they were in their own context, were not necessarily directly applicable to the different circumstances of later times. People of his own time should study the classics to grasp the approach to ordering society formulated by the early kings, but it was senseless to try to adopt mechanically the specific details of their insti tutions. This premise presumably figured in his criticism of the Song thinkers’ explication of the texts on rites as “fussy and nitpicking, like an old woman trying to train a young girl.”12 It helps to explain as well his reluctance to present specific modes of rule from the Rituals of Zhou as a concrete model for the present.
Apart from having reservations about the direct applicability to his own times of the concrete institutions of antiquity, Sorai does not seem to have been attracted to the highly schematic, pyramidal pro-
10 See, for example, his use of the Rituals of Zhou to explicate from a fresh per spective a puzzling passage from chapter 10 of the Analects. Rongochō (Ogyū Sorai zenshū[a] 4.86–88, 439–440). 11 Kenroku 鈐錄 (Ogyū Sorai zenshū[b] 6.217); Tōmonsho (Ogyū Sorai zenshū[a] 1.446); Yamashita 1994, 64–65. It is possible that Sorai referred in particular to the Rituals of Zhou as evidence of the weight the sages put on military matters because, as James Legge notes, the Wangzhi 王制 section of the Records of Ritual, the other cen tral source of information on the Zhou governmental system, does not say anything about the minister of war’s management of military forces. Legge 1885, vol. 1, 19. 12 Keishishi yōran (Ogyū Sorai zenshū[a] 1.512).
TOKUGAWA APPROACHES TO THE ZHOULI 285
gram set out in the Rituals of Zhou. To be sure, Sorai is known for his insistence on the need for a well-worked-out plan of governance. As he put it in the memorable opening sentences of Discussions of Gov ernment (Seidan 政談), the set of propositions for reform that he sub mitted to the eighth shogun, governing a state is akin to measuring out and marking the lines on a go board. Just as one cannot play go with out a properly marked go board, it is impossible to bring order to soci ety without an underlying plan.13 But when it came to the specifics of such a plan, Sorai in many ways favored soft structures that would knit rulers and ruled together in an organic fashion.
Given this qualification, the area where Sorai appears to have found the Rituals of Zhou most relevant was in regard to military af fairs and the situation of the samurai class. Certain aspects of his pro posals for returning the samurai to the land might possibly be linked to the Rituals of Zhou. Specifically, his suggestion that samurai should be settled on the land in units under the leadership of samurai of higher rank, who would all live on holdings in the same area, might be held to resonate with the “platoon and company” (zuwu 卒伍) local military system described in the section on the offices under the min ister of war.14 Other thinkers in the Sorai school lineage, including the scholars of the late Mito school, would draw that connection and build upon it. But Sorai himself did not explicitly refer to the Rituals of Zhou as a model for how to settle the samurai on the land. Rather, in some ways at least, he showed a nostalgia for an earlier stage of Japa nese history when, in contrast to their present degenerate state, samu rai had been like “grasses and trees rooted in the earth.” It had been that circumstance that had enabled valiant warrior houses of the four teenth century like the Kusunoki 楠 to survive and their descendants to rally and resume the fight “no matter how many times their for tresses were brought down.”15 Of the thinkers considered here, Sorai was perhaps the one most drawn to a genuinely “feudal” system.
13 Seidan (Nihon shisō taikei 36.263); Lidin 1999, 71. 14 Seidan (Nihon shisō taikei 36.299–300); Lidin 1999, 129–131. See Zhouli zhen gyi 2473–2485. 15 Kenroku (Ogyū Sorai zenshū[b] 6.229). The Kusunoki were among the leading supporters of Emperor Go-Daigo 後醍醐 (1288–1339) and the Southern Court in the turbulent early decades of the Muromachi period (1336–1573). Sorai, who wrote sev eral treatises on military affairs, did not, however, assert the superiority of Japanese military tradition. He consistently affirmed the need for methods of disciplined orga
nization that would enable commanders to deploy troops smoothly in the eventuality of war, and he generally held China to be more advanced in this regard than Japan. A
286
Dazai Shundai, who studied with Sorai and referred to him as his master, but also diverged from him in many ways, showed a consid erably stronger attraction to the Rituals of Zhou. Acclaiming the mer its of the feudal over the centralized commanderies and prefectures system of rule, he observed that although, in China, rulers had aban doned the former in favor of the latter, in Japan, despite the adoption of centralized structures in the seventh century under the influence of Tang institutions, a feudal system had eventually evolved.16 Like Sorai, however, Shundai saw the Tokugawa version of feudal rule as marred and imperfect and formulated a variety of recommendations for over coming its deficiencies. These recommendations show a more regi mented and schematic approach to institutional reform than that of Sorai, and thus it perhaps is not strange that the Rituals of Zhou held a greater appeal for Shundai.
Shundai set out his proposals for reform of various areas of institu tional life in a treatise called A Report on Issues of Governance (Kei zairoku 經濟錄). There, under the heading of “military readiness” (bubi 武備), he referred explicitly to the “platoon and company” sys tem as the appropriate model for settling the samurai on the land. At
present in Japan, he noted, lower-ranking retainers were organized in units, but this, which did not involve settling them on the land, bore only a superficial resemblance to the Zhou zuwu system. Further, to establish an ongoing structure for effective military training, it was crucial to expand the platoon and company arrangement to incorporate samurai of higher rank as well. As an example of the merits of such a structure, Shundai pointed to the description in the Rituals of Zhou of the organization of seasonal hunts in local areas under the supervision of the minister of war. These, he observed, served simultaneously as occasions for training the platoon and company units in coordinated military maneuvers and for checking on their equipment and perform
ance.17
Shundai’s praise for the merits of the platoon and company system shows an orientation toward clearly defined structure and a govern-
Record of Essentials (Kenroku 鈐錄), the treatise cited here, which dates from a late stage in the evolution of his thought, elaborates on these points in considerable detail. An earlier work, a commentary on Wuzi 吳子, provides one example of Sorai explic itly praising the Rituals of Zhou, in this case as exemplifying an “authentic mode” of military organization. See Kojima 1989, 74. On the stages in the development of Sorai’s military thought, see Kataoka 1998. 16 Nihon keizai taiten 9.398–400. See Zhouli zhengyi 2307–2328. 17 Nihon keizai taiten 9.581–587, 635–636.
TOKUGAWA APPROACHES TO THE ZHOULI 287
mental organization well coordinated from top to bottom. The same is true of his recommendations for reform of the bakufu system of of fices, which he prefaced by an extended summary of the administra tive structure described in the Rituals of Zhou. As one particularly problematic feature of the current arrangement he noted the pattern of collegial rule, in which, for instance, a group of bakufu vassals, the senior councilors (rōjū 老中), jointly headed the bakufu administra tion and collectively shared responsibility for overseeing the various affairs of government. Another problem was the custom of appointing multiple figures as the heads of a particular office and having them assume responsibility for overseeing the duties of that office on a monthly rotation. These practices, Shundai argued, encouraged indeci siveness and a diffusion of responsibility, weakening the authority (ken 權) essential to the effective conduct of official affairs. To correct this situation, he called for following the Zhou model, whereby the six ministers each focused on the duties of his own office and did not in terfere with those of his fellow ministers; if there were matters about which one minister felt uncertain, he might consult with his fellows, but “the ultimate decision was his alone.” Further, each ministry should be supported by a well-demarcated hierarchy of suboffices, each with its specific responsibilities. As matters of particular urgency Shundai argued that offices concerned with military matters should be distinguished from those handling civil affairs, which would improve the professional competence of each; the office handling civil judicial suits should be separated from that responsible for criminal cases; and offices should be established to supervise the daimyo and coordinate matters concerning them. The office charged with hearing and dispos ing of criminal cases should have jurisdiction over all sectors of the population, from samurai to the different categories of commoner, rather than the jurisdiction over different status categories being di vided among various offices, as at present.18
Sorai, too, criticized the current bakufu practice of appointing mul tiple figures of the same rank to the same office and having them serve on a monthly rotation, and he called instead for the establish ment of a better-defined structure of suboffices. He also advocated a clearer demarcation of functions between offices and agreed that of fices involving military responsibilities should be better differentiated from those focused on general governance. Yet, compared with Shun-
18 Keizairoku (Nihon keizai taiten 9.452–469, 585, 594).288
dai, he tended to emphasize incremental reforms within the existing administrative structure; the systematic bureaucratic mechanisms held to have been developed in the Zhou and continued by later dynasties do not seem to have held so strong an appeal for him.19 With Shundai, by contrast, there appears to be an undeniable, if unadmitted, impulse toward a centripetal mode of government, and we may assume that he was encouraged in such a tendency by his reading of the Rituals of Zhou, which presented such a mode as an authentic form of “feudal ism.”
Mito Approaches to the Rituals of Zhou
The Tokugawa thinkers for whom the Rituals of Zhou had the greatest resonance were perhaps the scholars of the late Mito school: Fujita Yūkoku 藤田幽谷 (1774–1826), Aizawa Seishisai 會澤正志齋 (1782–1863), and Yūkoku’s son Tōko 東湖 (1806–1855). In the last decade of the eighteenth century and first half of the nineteenth, these figures associated with the Mito domain, one of the three main To kugawa collateral houses, developed a distinctive variety of “Japa nese” Confucianism that combined a highly nationalistic orientation (including an acclamation of Shinto) with an intense, if somewhat idiosyncratic, commitment to the Way of the Sages. The late Mito scholars rejected Ogyū Sorai’s interpretation of the Way as something created, the sum total of the rites and institutions established by the sages. To the contrary, they asserted, the Way was grounded in a natu ral and universal moral order. Despite this significant difference, though, the Mito scholars shared much with Sorai and may be located in the lineage of thought deriving from him. Like Sorai, they were highly dubious about the inward-looking, meditative dimensions of Song and Ming Confucianism, which they saw as encouraging an overly subjective outlook and a debilitating taste for abstract theory on the part of its practitioners. Such an approach, they were convinced, did not connect with the practical needs of society. In its place, they, again like Sorai, looked to rites and institutions as the key means for ordering society.
Establishing appropriate rites and institutions was for the Mito scholars an urgent matter. They saw their society as standing on the brink of disaster, eroded from within by a lack of social cohesion and
19 See Seidan (Nihon shisō taikei 36.350–393); Lidin 1999, 195–249.
TOKUGAWA APPROACHES TO THE ZHOULI 289
threatened from without by forces of subversion. To counter these dangers they sought to carry out a variety of reforms at the domain level and called as well for changes at the national level. With the backing of the energetic daimyo Tokugawa Nariaki 德川齊昭 (1800–
1860), they succeeded in implementing some of their policy recom mendations for the domain. Factional strife within the domain that became entangled with national politics, and the resulting ups and downs in the careers of Nariaki and those associated with him, meant that most of the reform program was never realized, but the reverbera tions of what the Mito scholars attempted had a considerable impact on late Tokugawa political and intellectual life.
In formulating their program the late Mito scholars both drew in spiration from the Rituals of Zhou and made use of it for their own purposes.20 Fujita Yūkoku, the founder of the late Mito school, wrote relatively little in the way of treatises or explications of Confucian doctrines and texts. In a memoir of Yūkoku’s method of instruction and views on various texts and issues, Aizawa Seishisai, his leading disciple, noted, however, that Yūkoku “was particularly fond of read ing the Offices of Zhou (Zhouguan 周官, i.e., the Zhouli) and declared that the sages’ methods for regulating the realm and ordering the state are all contained within this work.” According to Aizawa, Yūkoku’s exposition of the Rituals of Zhou to his students showed “many in sights not yet discovered by earlier scholars”; hearing his explanation of the institutions of feudalism described in the text and their perti nence to the Tokugawa present, Aizawa himself “suddenly realized how relevant this work was to actual issues of governance.”21 Acting upon this perception, Aizawa cited the Rituals of Zhou in various of the works in which he called for a response to the needs of the day and set forth his vision of society and government, from the early, polemi cal (and eventually widely read) New Theses (Shinron 新論, 1825) to the late, comprehensive Simple Words about Matters Close at Hand (Kagaku jigen 下學邇言).22 In his seventies, as part of a series of
20 To date, this issue does not seem to have been much studied. The major excep tion is two articles by Imai Usaburō 今井宇三郎 (1973, 1974) that provide brief but valuable analyses of Mito readings of the Rituals of Zhou and other key classical texts. There is considerable overlap between the two articles; for the sections specifically on the Rituals of Zhou, see respectively 543–552 and 525–532. The following discussion owes much to Imai’s analysis of Mito approaches to the Rituals of Zhou. 21 Kyūmon ihan 級門遺範 789. See also Imai 1973, 528. 22 Aizawa evidently compiled an initial draft of Kagaku jigen in 1847 but contin ued to revise it until his last years. Imai 1973, 527.
290
notes and reflections on various of the Classics, Aizawa also wrote a study of the Rituals of Zhou entitled Reflections on the Offices of Zhou (Doku Shūkan 讀周官, 1854) that stands as one of the few Tokugawa explorations of the implications of the text. Although not the most vo luminous of Aizawa’s studies of classical works, it shows, in Imai Usaburō’s description, the greatest investment of energy and effort in terms of content.23
Whether citing the Rituals of Zhou in his treatises or discussing the text, Aizawa consistently focused on the aspects of its content that he found most relevant to his own concerns. He was aware of the various debates over the authenticity of the text but did not take a particular interest in this issue. Remarking at the beginning of Reflections on the Offices of Zhou that various later scholars had raised doubts about whether the Rituals of Zhou was truly a work of the sages, he attrib
uted their suspicions to the fact that, living in an age of centralized rule, they could not correctly grasp the nature of the feudal institutions it described. Those institutions were “so perspicacious and far reaching in intent that only a sage could have conceived them.” In short, he concluded, “while this work may not have been written by Zhou gong himself, that it is from the hand of a Zhou official scribe, there can be absolutely no doubt.”24 Aizawa’s approach to the text was selective rather than comprehensive. He did not comment on it line by line; instead, he extracted from it and from various commentaries a synthetic overview of what he saw as the main features of Zhou gov
ernance. After presenting this overview at the outset of his reflections, he proceeded to develop and supplement it through comments on the function, scope, and key characteristics of the top offices of each of
23 Imai 1973, 534. The other Chinese works on which Aizawa compiled such notes include the Classic of Poetry (Shjingi 詩經), Doctrine of the Mean (Zhongyong 中庸), Classic of Documents (Shangshu 尚書; three compilations), Analects (Lunyu 論語), Classic of Filial Piety (Xiaojing 孝經), and Classic of Changes (Yijing 易經). None of these reflections are available in published form. 24 Doku Shūkan 1.1a–b. See also Imai 1974, 532. The manuscript copy of Doku Shūkan cited here, made in 1925 by Terakado Makoto 寺門誠, a relative of Aizawa’s, is held today by the Mukyūkai 無窮會 library in Machida-shi, Tokyo. It is unpagi nated, and the page numbers cited are provisional. I am indebted to Sawai Keiichi 澤
井啓一 for generously providing me with a copy of the Mukyūkai text. Aizawa took a somewhat cavalier attitude to questions of textual authenticity. In his writings on the Classic of Documents he generally followed the contemporary script (jinwen 近文) version, and he criticized various sections of the ancient script (guwen 古文) version as false interpolations, but he also cited passages from those sections when they were of use to him. See Imai 1974, 520–523.
TOKUGAWA APPROACHES TO THE ZHOULI 291
the five major branches of government surveyed in the Rituals of Zhou. Of the suboffices he singled out for consideration only those that were of particular interest to him.25
For the Department of the Royal Household, headed by the premier, for instance, Aizawa devoted most of his attention to the “six general principles of governance” (liudian 六典), the “eight aspects of manag ing the central administrative offices” (bafa 八法), the “eight dimen sions of governing the fiefs held by ministers of high rank” (baze 八 則), the “eight mechanisms for regulating the king’s vassals” (babing 八柄), and the “eight means for regulating the populace” (batong 八 統). He focused as well on the several “nines” having to do with the management of economic resources, income, and disbursements. He then summarized the functions of the four major suboffices under the premier and, among the lesser offices, took up as particularly pertinent the great treasurers (dafu 大府) and accountants (sikuai 司會), who were responsible for managing the reception and auditing of various tax and tributary goods, and the chiefs of the harem (neizai 内宰), who were responsible for overseeing the management of the women’s quarters of the king’s palace.26 As this selection shows, Aizawa’s pri mary interest lay in the dynamics of the system of rule described in the Rituals of Zhou rather than in the details of its elaborate structure of offices.
That Aizawa found such aspects of the Rituals of Zhou well tai lored to his needs can be seen from the fact that in other works, such as Simple Words about Matters Close at Hand, he often cited or para phrased the same passages to illustrate and provide supporting evi dence for institutional arrangements or ritual practices he was recom mending. As with the other sources he utilized, however, he also fre quently took points out of context, cited fragments, and wove them together with other elements. Such an approach was a characteristic feature of “Mito learning” (Mitogaku 水戸学) and can be seen in the late Mito scholars’ use of Japanese as well as Chinese sources and in the very language of their works in Chinese (kanbun 漢文), which are
25 Imai provides an outline of the table of contents of Doku Shūkan in Imai 1974, 526–529. See also his description of the work on 529–531 and the several passages that he quotes from it in the text and notes of the article. Aizawa discussed only the first five branches of government described in the Rituals of Zhou and did not take up the “lost” section on the Offices of Winter, which was replaced by the “Artificer’s Record.” 26 Doku Shūkan 1.27a–50a.
292
laden with snippets of passages from the Classics. The ultimate result was in many ways a montage juxtaposing and conjoining originally disparate elements.27 The same may be said of their overall interpreta tion of the dynamics of the Zhou system that they extracted from the Rituals of Zhou.
The Rituals of Zhou as a Model of Feudal Governance
The Rituals of Zhou seems to have appealed to the Mito scholars on several different levels. One aspect of the text’s attraction was likely its comprehensive nature, the way in which its schematic structures seemed to cover every eventuality. The Mito scholars had an abhor
rence of divisions, of cracks in society and potential conflicts in values and norms. Seeing such cracks and conflicts as the point of entry for the forces of subversion, they struggled to contain and forestall divi sion in all forms. This dimension of their outlook is apparent in the emphasis on fusing dyadic elements into one that permeated late Mito thought, finding expression in slogans of unity such as “the unification of governance and instruction,” that is, conveying moral instruction through the process and mechanism of government administration (seikyō itchi 政教一致 or jikyō itchi 治教一致); “the unification of ritual and governance” (saisei itchi 祭政一致); “no division between civil and military” (bunbu fuki 文武不岐); and “no separation be
tween military and agricultural functions,” that is, grounding the mili tary class in local society (heinō buigyō 兵農無異業). In their reading of the utopian schematics of the Rituals of Zhou, the Mito scholars found instantiated the means of pursuing such visions of unity. As Ai zawa put it regarding the unification of governance and instruction, the Rituals of Zhou attested that this was fundamental to the method of the sages: “they incorporated governance within instruction and in struction within governance.” It was not like the situation in later ages,
27 Imai points out, for instance, that Aizawa took references to Heaven in the Clas sic of Documents out of context to provide supporting evidence for his vision of the unification of ritual and governance, one of the central pillars of the Mito program. Imai (1974, 522–525) suggests that this approach led to a “japanization” of Confu cianism, in which “Shinto” elements were given precedence over Confucian princi ples. The argument is valid, but one might also hold that the opposite occurs as well— that elements of Japanese tradition are likewise taken out of context and reoriented by being placed in a Confucian framework. It is the difficulty of untangling the pieces that makes the result a montage.
TOKUGAWA APPROACHES TO THE ZHOULI 293
when instruction was held to be something outside the affairs of gov ernment and treated as a secondary activity devoted to the explication of texts.28
A second attraction of the Rituals of Zhou for the Mito scholars was undoubtedly its institutional focus. As noted above, they shared with Sorai and Shundai a deep suspicion of Song and Ming thought. If anything, they felt even more strongly than these earlier figures that the Song and Ming thinkers had failed to grasp the mind-set and emo
tional needs of the populace. As a method for keeping the common people from going astray and for mobilizing their energies in service to society and the ruler, the Song penchant for wordy, didactic ser monizing was, the Mito scholars held, merely counterproductive. The Rituals of Zhou offered a far more reliable model as to how to accom plish the essential task of inculcating virtue through institutional and ritual structures that took human instincts into account while simulta
neously organizing economic and military resources effectively. Third, the Mito scholars found in the Rituals of Zhou specific guidelines for realizing the integration of different levels of govern ance that they saw as essential to forestalling the emergence of social divisions and fostering unity in aims and activities. As we can surmise from Aizawa’s description of the insights he gained from Fujita Yūkoku’s exposition of the text, the Rituals of Zhou held a particular appeal for the late Mito scholars in this regard because it purported to describe a feudal system of rule; it thus could be taken to be directly pertinent to the circumstances of their own society, which was like wise based on feudal principles. At the same time, as with Dazai Shundai, for the Mito scholars, too, the Rituals of Zhou seems to have performed the ideological function of disguising, or even encouraging, a slippage between formal commitment to a decentralized feudal sys tem and an actual inclination toward more centripetal mechanisms of rule. Whereas for Shundai this slippage occurred largely in the sphere of theory, in the case of the Mito scholars it was applicable as well to their concrete efforts at reform.
This circumstance raises the question of the ultimate focus of the Mito scholars’ belief in the relevance of the Rituals of Zhou. After all, the Rituals of Zhou described a national structure, centered on the government of the king but also encompassing the enfeoffed regional states (bangguo 邦國) of the feudal lords. The Mito scholars had
28 Doku Shūkan 1.6a.
294
strong national concerns, as is evident from their fear of incursions by foreign powers and from the weight they put on the place of the em peror within the Japanese polity as the apex of a dynamic of ritual, on the one hand, and loyalty, on the other.29 Coming from one of the three main Tokugawa collateral houses, they also took for granted the central administrative role of the bakufu and shogun. Yet Mito’s posi tion relative to the bakufu or the nation as a whole was that of an en feoffed regional state. How did this affect their perspective on the Rituals of Zhou? Did they envision it as something applicable to the Tokugawa state as a totality, or did they approach it from a more local orientation? The answer remains ambiguous. Unlike many of the Chi nese commentators on the text discussed in other contributions to this volume,30 the Mito scholars took no great interest in its bearing on the relationship between the central government and the enfeoffed re gional states.31 They likewise did not share Dazai Shundai’s concern with bureaucratic rationalization and clarifying lines of jurisdiction. Their interest lay above all in methods for organizing and utilizing human resources, and the methods they singled out could be applied at either a macrocosmic or a more microcosmic level. We see a key in stance of this in one of the centerpieces of the Mito interpretation of the Rituals of Zhou: the argument that it described a telescoping sys tem of offices that linked different dimensions and levels of govern ance.
As mentioned above, as evidence of the importance the sages ac corded military affairs, Ogyū Sorai observed that, according to the Rituals of Zhou, at times of war the six ministers responsible for the overall administration of the state would become the commanding officers of the six armies. He likely adopted this assumption from an observation made by the later Han commentator Zheng Xuan 鄭玄 (127–200) on a passage concerning the functions of the Ministry of War: “The commanders of the armies were all appointed from among the ministers” (軍將皆命卿。). Zheng interpreted this to mean that the commanding generals of the six armies were not appointed sepa-
29 On these points, see Nakai 2002, 279–291; Nakai 2006. See also Wakabayashi 1986; Koschmann 1987. 30 See, e.g., chapter 8 (Jaeyoon Song). 31 Related to this point, but also reflecting a latent ambiguity in Mito thought, is the fact that Mito scholars do not address at all the issue of whether the equivalent to the “king” of the Rituals of Zhou was the emperor or the shogun. Dazai Shundai, by con trast, makes quite clear that it was the shogun.
TOKUGAWA APPROACHES TO THE ZHOULI 295
rately on a standing basis; instead, officials of the six departments and six districts (xiang 郷), beginning with the minister, were selected to serve in military capacities, which they combined with their regular duties.32
Unlike Shundai, who had stressed the separation of responsibility for military and civil affairs as one of the strong points of the govern mental system depicted in the Rituals of Zhou, Yūkoku, taking much further Sorai’s passing comment, made the doubling up of military and civil offices a major feature of his interpretation of the Zhou feu dal system. In Reflections on the Offices of Zhou, Aizawa gathered evidence in support of this view and elaborated upon its implications. Collating data from different sections of the Rituals of Zhou about the ranks of the heads of the different levels of suboffices and the number of men under their authority, extrapolating from the information con cerning one office to a more general picture, fitting the description of different types of stipendiary fields together with his image of how the system must have operated as a whole, and combining selectively elements of the interpretations advanced by various commentators, he built up a picture of the Zhou system as combining key central and local offices. It thereby, he held, not only integrated the different di mensions of government, including military and civil functions, but also knit the countryside and the seat of government together in such a manner as to ensure appropriate guidance of the rural population and their effective mobilization when necessary.
The heart of this interpretation, as Imai Usaburō points out, was the premise that the six ministers heading the departments that constituted the king’s government served simultaneously as the district adminis trators (xiang dafu 郷大夫) of the six central rural districts (xiang), which were located near the capital. Each of the ministers was in vested with one of the rural districts, which thus also provided him with his source of sustenance. Although the six ministers themselves normally resided in the seat of government so as to attend to the func tions of their department, the subofficials of each district, arranged in a pyramid of units of descending size, were based in the district, with the lowest-level officials responsible for immediate supervision of the general populace living in their neighborhoods. Living in their neigh borhoods, these lowest-level officials were able to provide for their own sustenance, but at the same time, because the districts were not
32 Zhouli zhengyi 2237–2238.
296
too far from the government seat, some of them would commute to the center, where they served as the miscellaneous staff of the various government offices based there. In time of war, the same pyramidal arrangement was converted into a military organization. The six min
isters, who were also the six rural district administrators, became the commanding generals of the six armies. The bottom-level officials of the district, leading the men of their respective neighborhoods, formed the basic fighting units, which were grouped and coordinated with other units in successively larger assemblages headed by the interme
diary officials of the district.
Supporting this integrated military and civil administration cen tered in the capital region was a parallel organization of the resources of the more remote rural areas beyond the central region; these formed six outlying districts (liusui 六遂). In peacetime, the subunits of these outlying districts, headed by officials who were ranked lower than those of the six central rural districts, concentrated solely on agricul ture. Unlike the different levels of officials of the central rural districts, those of the outlying districts did not have the dual functions of serv ing in the staff of the central government departments or of seeing to the military training of the men of their neighborhoods. Rather, in time of war, the subunits of the outlying districts served as support auxiliaries to the six armies, providing them with horses, military equipment, and porters.33
As regards the level at which such a scheme might be applied, in principle it could be implemented at either the level of the government of the king (or, in the Tokugawa instance, the bakufu) or that of the enfeoffed feudal states (or daimyo domain). For the Mito scholars, the immediate target for trying to institute an arrangement with compara
ble dynamics was the domain.
In postulating that the feudal system encapsulated in the Rituals of Zhou rested on a dovetailing of military and civil functions, personnel, and their means of sustenance, Yūkoku and Aizawa were likely in spired by a certain resonance with the Tokugawa setting. Describing the Zhou system in terms that would be familiar to a Tokugawa audi ence, Aizawa remarked on one occasion that “the six armies of the Rituals of Zhou were the king’s bannermen troops (hatamoto sonae 旗
33 Imai Usaburō provides an outline of Aizawa’s reasoning and argument regarding these issues (and points out some of the problems in his equations) (1973, 545–552). Aizawa himself gives a summation in Kōko fuken 江湖負暄 498–500; Kagaku jigen 79b–81a; with a full account in Doku Shūkan 1.1b–27a.
TOKUGAWA APPROACHES TO THE ZHOULI 297
本備).”34 In the Tokugawa context “bannermen” indicated the direct retainers of the shogun below the rank of daimyo; by origin, these made up his personal army. Although they had long since become permanently resident in Edo, at the beginning of the Tokugawa period the bannermen had been allocated fiefs in the surrounding area, from which they were supposed to commute to the shogunal seat to take up their duties there.
Aizawa and Yūkoku believed that their firsthand knowledge of the principles of a feudal system had provided them with a special insight into the nature of what was described in the Rituals of Zhou: it en abled them to grasp the essential dynamics of the Zhou mechanism of rule in a way that had escaped Chinese commentators of the Han on. “In debating feudal institutions while living in an age of commander ies and prefectures,” such commentators, Aizawa quoted Yūkoku as saying, “had not been able to detect those institutions’ true form. Their interpretations often miss the point, like one trying to scratch an itch through his shoes.”35 For all their confidence in the insight into the text afforded by their Tokugawa experience, however, the late Mito scholars did not look to the Rituals of Zhou to validate current To kugawa institutional arrangements. To the contrary, they sought in it guidelines for rectifying the problems of the day by establishing a more authentic form of “feudalism.” Such an approach colored many aspects of their reform program. It is evident above all in their propos als for returning domain retainers to the land (dochaku 土着), an issue closely related to their conviction that the Zhou system, by integrating the civil and military dimensions of governance in the manner de scribed above, offered a means for effective guidance and mobiliza tion of the rural population.
The Mito scholars inherited from Sorai and Shundai the assump tion that the removal of the samurai from the land had caused multiple social and economic problems and had undermined military fitness, which should have been one of the samurai class’s major raisons d’être. Echoing Sorai’s description of the stalwart samurai of earlier ages as like “grasses and trees rooted in the earth,” Fujita Tōko wrote that the removal of the samurai from the countryside and their relega tion to a castle town existence had turned them into “potted plants” or,
34 See Kōko fuken 498. 35 Doku Shūkan 1.1a. For similar observations, see also Kyūmon ihan, 789; Kōko fuken 498.298
even worse, had left them no more than “flowers in a vase,” weak and ineffectual. Having grown accustomed to living beyond their means, samurai no longer strove to maintain the number of retainers neces sary to fulfill their military service, an urgent matter given the loom ing danger presented by the increasing Western presence in the vicin ity of Japan.36
Compared to Sorai, however, the Mito scholars were far less san guine about the potential consequences of returning the samurai to the land without putting in place a system of institutional safeguards. Sorai had not seen the return of the samurai to the land as entailing any particular dangers; the sense of mutual solidarity that would result from settling the samurai on the land in units under the leadership of one of higher rank would suffice, he suggested, to restrain any poten tial troublemaker among their number.37 In their recommendations concerning this issue, Tōko and Aizawa, by contrast, pointed out bluntly that, left to their own devices without any controls, landed samurai might become a new source of disorder, as in the Kamakura (1185–1336) and Muromachi periods (1336–1573). It was not only loyal vassals like the Kusunoki, Tōko noted, who had proved difficult to eradicate and control.38 To avoid the twin evils of concentrating the samurai in castle towns, on the one hand, and simply allowing them to live in an unregulated fashion in the countryside, on the other, the op
timal course, he and Aizawa argued, was to follow the Zhou system and adopt a “truly” feudal style of settling the samurai on the land.39 In a set of policy recommendations compiled in 1848, Aizawa indi cated how this might be done. Collating the pyramidal organization of units based in the rural and outlying districts that he reconstructed from the Rituals of Zhou with the ranks and sections of the Mito re tainer band, he proposed a similar geographical dispersal of retainers and a restructuring of the stipendiary system. These measures, he held, would make it possible to improve military readiness and reduce ex penditures. At the same time, they would forestall the dangers of al lowing retainers too high a degree of independence or too much indi vidual authority over the agricultural population.40
36 Fujita 1940, 773–774. See also Kōko fuken 497–498. 37 Seidan (Nihon shisō taikei 36.343); Lidin 1999, 186. 38 Fujita 1940, 777. 39 Fujita 1940, 776–777; Kōko fuken 497–498. See also Kagaku jigen 81a–82a. 40 Kōko fuken 498–502. Aizawa had made similar recommendations earlier, in 1838. See Kōko fuken 161–179, esp. 172–175. The Mito domain had already taken
TOKUGAWA APPROACHES TO THE ZHOULI 299
Unification of Governance and Instruction;
Fusion of Ritual and Governance
Aizawa and his fellows thus identified in the Rituals of Zhou a map for integrating the countryside with the seat of government in such a way as to avoid both the problems that had arisen in earlier periods of Japanese history, when the samurai resided on the land, and those re
sulting from the fact that they no longer did so. The text spoke to other of the Mito scholars’ concerns as well: in particular, its depiction of how the Zhou had “incorporated governance within instruction and instruction within governance” suggested ways to not only uphold hierarchy but also allow promotion of talent. Simultaneously, in their view, it attested to the efficacy of the fusion of ritual and governance as closely related means of mobilizing and guiding the people.
Their fear that social cracks and divisions would create openings for subversive forces led the Mito scholars to emphasize the impor tance of reinforcing hierarchical structures. Assuming hierarchy to be built into the nature of things, they held that the maintenance of social order depended on inferiors loyally serving their superiors by per forming the functions appropriate to their hereditary status. Such a graded status order would ensure the security and stability, and thus the welfare, of society as a whole. At the same time, the Mito scholars were fully aware of the need to keep channels open for the promotion of talent. Automatic hereditary succession to office, they noted, en couraged laziness and a taking for granted of privileges on the part of high-ranking officeholders. Further, without input from people with different backgrounds and perspectives, policy would become narrow and one-sided, and governors would be unable to respond effectively to the needs of the realm as a whole. The Mito scholars also had a quite immediate personal interest in the issue of “raising up the wor thy.” Within the context of the status- and rank-based Tokugawa so cial order, they themselves were in many ways upstarts; their call for reforms challenged the status quo and consequently was opposed by many within the upper echelons of the domain and bakufu ruling class.
steps toward revising the stipendiary system and reallocating fiefs in the 1830s but had not been able to implement the full system of settling retainers on the land envi sioned by the reformers. For the reforms, see Koschmann 1987, 101–114. As noted above, factional politics and the vicissitudes in the situation of both the Fujita camp and their key backer, the daimyo Tokugawa Nariaki, meant that the reform program followed an erratic course.
300
They knew firsthand that, as Aizawa put it, to leave a man of ability in the provinces, without utilizing his talents at higher levels of govern ment, was like leaving a tiger to roam free in the wild.41 The Zhou system, in the Mito scholars’ eyes, had developed mechanisms for maintaining a balance between the different but equally critical goals of preserving the social hierarchy and developing ways of promoting talent.
In Simple Words about Matters Close at Hand, his comprehensive summation of his overall intellectual and political perspective, Aizawa took up “learning” (gaku 學) as the second major topic (following the Way, michi/dō 道). There he described at some length the Zhou ap
proach to education as he reconstructed it from the Rituals of Zhou, pointing out that it integrated various forms of instruction with the feudal administrative arrangement that he acclaimed as the appropriate foundation for returning the samurai to the land. Citing various pas sages describing the functions of the central rural district administra tors, he noted that one of their major responsibilities was to see to the education of those under their jurisdiction. The district administrators received directions as to the proper methods of instruction from the minister of the multitude, who had overall responsibility for educa tional matters, and in turn appointed subordinate district officials (xiangli 郷吏) to oversee the provision of instruction in the district with which they were charged. Such officials were in particular to “foster virtue and keep track of talents and skills” (攷其德行、察其 道藝) and were to utilize various occasions to single out and recom mend for recognition those of particular accomplishment in these ar eas. According to Aizawa’s interpretation, the officials assigned this task were, of course, the people of varying rank who had been in vested with land and responsibility for overseeing the local agricul tural populace but who also served on a rotating basis as the lower level officials in the central government seat and, in time of war, would make up the army under the command of the same central rural district administrator. Such educational activities, he emphasized, were one of their most important functions.42
Aizawa identified various other officials as supporting the role of the central rural district administrators and their subordinates in “fos tering virtue and keeping track of talents and skills” and in enabling
41 Shinron 110; Wakabayashi 1986, 217–218. 42 Kagaku jigen 15a; see Zhouli zhengyi 839–850.
TOKUGAWA APPROACHES TO THE ZHOULI 301
those commended for their accomplishments to be properly acknowl edged and rewarded. He noted, in particular, among the offices under the Department of the Multitude, the palace master (shishi 師氏) and palace protector (baoshi 保氏), who were responsible for offering re monstrance to the ruler and for overseeing the education of the heir and scions of high-ranking vassals, and the remonstrator (sijian 司諫), who was responsible for checking on the conduct of local administra tion and whom Aizawa held to be a subordinate of the palace master and palace protector. In the course of their travels of surveillance through the districts, the remonstrators would record the especially worthy. These then would be employed in government service or would be given the opportunity to obtain further tutelage at the capital, under the direction of the palace master and palace protector, again with the assumption that this would lead ultimately to government service. The students who gathered to develop further their virtue and skills through schooling in the capital would benefit from intercourse with their fellows from other areas, and the positive effects of this contact would eventually extend to the entire realm. Superior to ab stract attempts to inculcate morality through didactic sermonizing, such tangible demonstrations that virtue and worth would be rewarded served as positive incentives for people to devote themselves to their assigned functions and behave appropriately. Aizawa observed that the ancients did not speak of “teaching” virtues and skills; rather, by granting office and rank on the basis of accomplishment in these things, they “taught people not through argument and debate but through concrete facts.”43
The modes of instruction directed at the districts, in Aizawa’s view, served in effect as a careful and measured means, within the frame work of a hereditary status order, to promote the occasional remark able figure of ability to a level of office higher than what his heredi tary status would normally qualify him to hold. The central educa tional institutions supervised by the palace master and palace protector and meant for the instruction of the scions of high-ranking vassals (guozi 國子) had a somewhat different purpose. This was, on the one hand, to nurture the greater degree of virtue and knowledge incumbent on those destined by birth to hold important offices and, on the other, to act as a testing ground for selecting from the pool of candidates de-
43 Kagaku jigen 14b, 15b. Aizawa similarly singled out the offices of palace master, palace protector, and remonstrator for attention in kan 3 of Doku Shūkan.
302
termined by rank those who were in fact best qualified to hold the leading offices. Those who did not prove worthy might be relegated to a category of retirement, and the house stipend even reduced in the next generation.44 In this way, again, the hereditary status order could be upheld without inviting the pernicious consequences of hereditary office.
The development of educational institutions along these lines was central to the Mito reform program. One of its centerpieces was the founding in 1841 of the domain school, the Kōdōkan 弘道館. By far the largest such academy in Japan at the time, the Kōdōkan combined the forms of schooling for high-ranking retainers and meritorious re
tainers of lower rank that Aizawa held had taken place under the su pervision of the palace master and palace protector. Although he ad mitted it might be difficult to accomplish immediately, Aizawa also stressed the importance of developing locally based schools compara ble to those he envisioned as having existed under the supervision of the rural district administrators. These would provide for the education of the body of retainers of lower rank and serve as the foundation for selecting the most able and motivated among them to continue at the domain school.45 Fujita Tōko included a similar proposal (and similar acknowledgment of the difficulties involved) in his recommendations
for implementing a policy of settling domain retainers on the land.46 Although samurai retainers (including those to be settled on the land in the vicinity of the domain seat), who might be expected to serve in government, needed to cultivate high levels of virtues and skills, the Mito scholars did not think such training was essential for the rest of the population. Aizawa quoted Yūkoku as having said that, in line with the difference in function between the rural and outlying districts, the education appropriate to the inhabitants of the two areas
44 Kagaku jigen 16a–17a. See also Kōko fuken 江湖負暄 180–190, where Aizawa connected these issues to the specific circumstances of Mito and plans for establishing a domain school. 45 Kōko fuken 180–182. It seems likely that Aizawa equated the function of the palace master and palace protector to his own role within the domain as advisor to Nariaki, instructor of the daimyo’s sons (including the later shogun Keiki 慶喜, 1873– 1913), and eventual head of the Kōdōkan; presumably this, too, was one reason for the weight he assigned to these offices. Aizawa saw the proposed local schools, which he termed shōgakkō 小學校 (in the sense not of primary education but small-scale, rather than large-scale, institutions), as an addition to the private academies that were already a major feature of domain education. On Mito educational reforms, see Koschmann 1987, 114–125. 46 Fujita 1940, 783.
TOKUGAWA APPROACHES TO THE ZHOULI 303
naturally differed. With the rare exception of a person of extraordinary merit, inhabitants of the outlying districts would not be expected to serve in office. Their days were spent in the fields, and in time of war they would function as porters and suppliers of goods. “It was not that the sages did not wish for the people of the fields (yajin 野人) to de velop virtues and skills; it was simply that it would suffice if they worked hard, devoting themselves to tilling the fields throughout the year.” There thus had been no need to provide them with detailed and complicated knowledge; the outlying district administrators posted to supervise these areas had instead focused on providing simple instruc tion of practical use.47 Despite this blunt image of what was appropri ate to the lower orders as opposed to the leadership stratum of society, under Nariaki the Mito domain in fact encouraged the creation of local schools targeted at the village leadership and rural samurai (gōshi 郷 士) class. The emphasis on the provision of training in subjects such as medicine that would be of immediate benefit to the populace of the area may be said, nevertheless, to echo Aizawa’s view of the differ ence in the function of and thus the education appropriate to the cen tral rural and outlying districts.48
If the Mito scholars found in the Rituals of Zhou’s depiction of the unification of governance and instruction methods for recognizing, nurturing, and promoting talent while upholding the premises of social hierarchy, they also drew evidence from the text on how the fusion of ritual and governance would further enhance the ideological efficacy of the unification of governance and instruction. In Simple Words about Matters Close at Hand, Aizawa organized the sections on ritual, which occupy close to a third of the entire work, according to the “five categories of rites” (wuli 五禮) mentioned in the Rituals of Zhou un der the responsibilities of the minister of the multitude and described as “that whereby he prevents falsity among the populace and instructs them how to achieve the mean.”49
47 Kagaku jigen 13b–14a. In Doku Shūkan Aizawa substantiated this argument by comparing evidence regarding the difference in duties of the central rural and outlying district administrators. See Doku Shūkan 3.30a–33a. 48 See Koschmann 1987, 123–125; Mito-shi shi 1971–1976, vol. 2, part 3, 199–204. While these local schools are known as kyōkō 郷校, the term kyō here indicates a rural district comparable to the outlying sui, not the Zhou xiang, situated in the vicinity of the government seat. 49 Technically, Aizawa used the definition of the “five rites” given by Zheng Zhong 鄭眾 (d. 83; quoted as “Zheng Sinong 鄭司農” by Zheng Xuan) and Zheng Xuan in their commentaries on the text. See Zhouli zhengyi 761.
304
As with his interpretation of the Zhou feudal administrative struc ture, Aizawa collated references to various forms of rites in different sections of the Rituals of Zhou to support his presumption that the fu sion of ritual and governance was a fundamental principle of Zhou rule. Whereas the minister of cult and rites was responsible for over seeing the five rites, “when it came to their implementation,” Aizawa emphasized, “the various offices worked in coordination… It was not like in later ages, when people regarded rites as ‘rites’ and governance as ‘governance,’ demarcating them as matters to be carried out sepa rately.” Thus, for instance, different offices gave the same priority to sacrifices to the spirits. Such sacrifices were listed under the duties of the minister of cult and rites as the first category of rites, “auspicious rites” (jili 吉禮); similarly, of the twelve methods of instruction for which the minister of the multitude was responsible the first was sacri fices to the spirits (sili 祀禮). Not only that, six of those twelve meth ods, Aizawa argued, “pertained directly to the use of rites to extend instruction,” and through the timely and appropriate conduct of rites in this manner, ritual was kept from becoming a “dead object” (shibutsu 死物). By contrast, in later ages, although in form rites may have nominally continued to exist, they were no longer integrated with governance and instruction and so had lost their efficacy and become simply dead objects.50
In his discussions of methods of returning the samurai to the land and forms of instruction, Aizawa presented specific institutions men tioned in the Rituals of Zhou as models for the present. For rites, while acclaiming the Zhou approach to the implementation of ritual as seen in the text, he tended to put more emphasis on showing that in antiq uity the same ritual categories had existed in Japan, as attested by early Japanese records, such as the Chronicles of Japan (Nihon shoki 日本書紀). This aspect of his argument and the call for the restoration of these ancient native ritual practices constitute the most overtly “Shinto” aspects of the Mito program. Even in the area of “auspicious rites,” however, where this tendency is most apparent, one may also readily detect a framing and reinterpretation of “native” rituals by aligning them with rites described in the Chinese Classics.51 A number of Aizawa’s concrete proposals for realizing the integration of ritual
50 Kagaku jigen 27b–28b. See Zhouli zhengyi 1297, 705–710; Doku Shūkan 2.3a– 5a, 4.1a–b, 4.3b. 51 See Doku Shūkan 4.2b–3b. I have examined some aspects of this phenomenon in Nakai 2002, 279–291; Nakai 2006.
TOKUGAWA APPROACHES TO THE ZHOULI 305
with governance and instruction show still more clearly his attraction and debt to the Rituals of Zhou in this regard.
A calendar of annual rites that Aizawa sought to distribute to the rural populace of the Mito domain provides one instance of this influ ence from the Rituals of Zhou. Compiled in the 1840s as part of a pro gram to reform village life, the calendar was intended both to encour age the elimination of heterodox religious customs (including the long-standing influence of Buddhism and its syncretic combination with Shinto) and to coordinate village spiritual practices with the round of court ceremonies, similarly to be reformed through the re vival of ancient, authentic rituals. The anticipation that such coordina tion would serve to instruct the populace in correct forms of behavior and to inculcate an instinctive readiness to serve their superiors itself bespeaks the Mito scholars’ affinity for the modes of instruction radi ating from the center that they found in the Rituals of Zhou.52 But they also adopted a more specific aspect of those modes.
In his discussion in Reflections on the Offices of Zhou of the func tions of the Department of the Multitude, Aizawa included a section entitled “Transmission of Instruction and Distribution of Laws” (“Fukyō hinpō” 布教頒法), in which he took up the implications of the “reading of laws” (dufa 讀法) that figures among the duties of several levels of the local officials under the rural district administra tor. The minister of the multitude, he noted, publicly posted laws as one dimension of the promulgation of instruction, and the rural district administrators, having received directions concerning methods of in struction and laws from the minister of the multitude at the beginning of the year, distributed them among the officials of their districts.53 Neither the minister nor the district administrators engaged in the “reading of laws.” Lower-level officials, however, all incorporated the “reading of laws” in various ritual occasions, with this happening more frequently the lower the level and the closer the contact with the populace. Aizawa praised this method of conveying guidance from above as far more efficacious than “noisy sermonizing” in encourag ing the populace to “work hard and be reverent, to uphold the instruc tion [received from above], and not to dare to be lazy and indolent.”54 Presumably not coincidentally, the ritual calendar that the reformers
52 Regarding this calendar, see Nakai 2006, 369–372. 53 See Zhouli zhengyi 751, 839–840, 861–868, 876–878, 884–886. 54 Doku Shūkan 2.29b–31a.
306
planned to distribute to the Mito domain villages included “reading of laws” as one of its features. The occasions when this was to be done were the anniversaries of the deaths of Tokugawa Ieyasu 德川家康 (1542–1616) and the first two domain lords.55 Although not strictly parallel to the occasions for the “reading of laws” listed in the Rituals of Zhou, the choice of these particular days for introducing such a practice may be said to have served a double purpose. It would en hance the authority of the laws by associating them with reverence for the deceased founders of the bakufu and the domain, and it would also provide a substitute for the Buddhist ceremonies normally performed as memorial rites in the Tokugawa period but regarded by the Mito reformers as ideologically deleterious.
While the plan to institute a reformulated round of village rites was not realized, the reformers did succeed in implementing another mode of ritual practice delineated in the Rituals of Zhou: organized hunts (tian 田). Like Shundai, the late Mito scholars assigned a special value to organized hunts. Whereas Shundai had treated hunts primarily as a form of military training (as is evident from the fact that he discussed them under the heading of “Military Readiness”), the late Mito schol
ars invested hunts with a larger meaning. In Simple Words about Mat ters Close at Hand, Aizawa elaborated on the significance of hunts as an aspect of ritual as well as martial activity, discussing as key forms of “military ritual” (junli 軍禮), the fourth of the “five rites,” the dif ferent types of hunts described under the Department of War in the Rituals of Zhou. These rites, too, were a means of instruction. People assembled from scattered locations were fused into one in spirit as well as organization through the structured drill of the hunt and the offering of its fruits at the ancestral shrine of the ruler.56 Aizawa con trasted these positive effects of hunts described in the Rituals of Zhou to the far less appropriate nature of hunts that had figured in earlier periods of Japanese history. A famous hunt organized at Fujino 富士 野 in the late twelfth century by Minamoto no Yoritomo 源頼朝 (1147–1199), the founder of the Kamakura bakufu, was often held to exemplify “a great hunt,” Aizawa noted. Yet it had involved little more than letting the warriors of the day maneuver as they wished and shoot freely at game. It did not incorporate the organized discipline of
55 Mito-han shiryō 5.50–54. 56 Kagaku jigen 51a–52a; Doku Shūkan 5.9a–11a.
TOKUGAWA APPROACHES TO THE ZHOULI 307
the hunts delineated in the Rituals of Zhou.57 We see here yet another dimension of the Mito view that the feudal system recorded in the Rituals of Zhou was more suitable as a model for the present than pre vious Japanese warrior regimes.
From 1840 on, the Mito domain did in fact organize a series of large-scale bird hunts that echoed various features of the comprehen sive “military ritual” that Aizawa identified in the Rituals of Zhou. The hunts involved the entire retainer band, on horseback or foot, de pending on rank, and dressed in armor. Manifesting the integration of the civil and military dimensions of governance, civil and educational officials, including Tōko and Aizawa, rode among the mounted warri ors. And even if the domain had not succeeded in carrying out the set tling of the samurai on the land that the Mito scholars saw as the es sential foundation of both military readiness and the unification of governance and instruction, the hunts gave concrete expression to the emphasis they placed on the need to coordinate the resources and ac tivities of the peripheral and central regions. The hunt incorporated as representatives of the outlying districts a unit of rural samurai (gōshi), led by the district magistrate responsible for their area, and a unit made up of Shinto priests and Shugen 修驗 mountain priests.58
As with the implications of resettling the samurai on the land, the forms of the unification of governance and instruction and of the fu sion of ritual and governance that the late Mito scholars found in the Rituals of Zhou may have had a “feudal” garb, but they in fact pointed toward highly integrated mechanisms of rule. Rudolf Wagner notes that late Qing commentators also emphasized the value of the unifica tion of governance and instruction depicted in the Rituals of Zhou, which they saw as facilitating the flow of information and opinion back and forth between local society and the center.59 The Mito schol ars did not totally ignore the function of such “channels of communi cation” as a vehicle for conveying the views of those below to those above. Yet clearly their major concern lay in fostering a reverse dy namic: the effective transmission of instruction from above so as to mobilize and coordinate the human resources of people of all classes, living in all areas of the domain. The Mito scholars were not prepared to step outside the Tokugawa governmental mode, and the name “feu-
57 Kagaku jigen 52a. 58 Mito-shi shi 1971–1976, vol. 2, part 3, 224–229. The hunts were said to involve three thousand mounted warriors and a support force of twenty thousand men. 59 See chapter 12 in this volume (Rudolf Wagner).308
dal” served to reassure them that what they sought was nothing more than an improved form of that mode. Ultimately, however, the inspira tion they drew from the Rituals of Zhou seems to have had quite dif ferent consequences. The collapse of the bakufu and the Tokugawa order would soon make the “feudal” aspects of their program, includ ing the notion of settling retainers on the land and the emphasis on preserving a hereditary social hierarchy and class structure, a thing of the past. But the vision of utilizing the unification of governance and instruction and the fusion of ritual and governance to mobilize and guide the populace from above was another matter. It had a momen tum of its own and soon would acquire a new potency under the changed political circumstances of the post-Restoration period. Even if in an indirect form, the Mito reading of the Rituals of Zhou would thus continue to have a significant afterlife in Japan’s modern history.
Rethinking Confucianism
Past and Present in China, Japan, Korea, and Vietnam
Edited by
Benjamin A. Elman
John B. Duncan
and
Herman Ooms
UCLA Asian Pacific Monograph Series
University of California, Los Angeles
2002
Chapter 8
Chinese Ritual and Native Identity in Tokugawa Confucianism
Kate Wildman Nakai
One of the central issues in the Japanese experience of Confucianism is the interaction between a “Chinese” system of thought and “native” Japanese cultural, social, and political practice. This interaction was in many regards a tug-of-war with no permanent victor. While Confu
cianism spread from its Chinese homeland to neighboring regions in considerable measure because of its plausible claims to universality, many of its core principles were rooted in the historical specificity of Chinese experience. The Chinese dimensions of Confucianism in combination with its relatively weak transcendental character made its reception by other societies an occasion of some tension. In the Japa
nese case various particularities of the native environment impeded the penetration of Confucianism as a cohesive structure, and the dynamics of a number of its key notions underwent substantial alteration at the hands of Japanese thinkers. But the modifications were not all in one direction. Under the impact of Confucian influences, the perception of
what was “native” also changed. The ramifications of Tokugawa atti tudes toward Confucian ritual (/i, rei ¥L.) well illustrate this two-sided process.
This chapter was prepared over the course of the various workshops and con ferences organized by the editors of this volume. Portions were presented as well at the conference “Women in Confucian Cultures in Premodern China, Korea, and Japan,” held at the University of California, San Diego, 28 June-1
July 1996. I would like to thank also my colleagues in the Faculty of Com parative Culture, Sophia University, for their cogent and helpful criticisms of the penultimate version.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 259
Tokugawa Confucian scholars expressed reservations about a number of aspects of li that were basic to Chinese Confucian thought but at odds with contemporaneous Japanese social practice. Yet the potential of ritual as an instrument of governance fascinated major thinkers of the time, and by the end of the period, visions of tapping that potential had spurred the formulation of a notion of rule through ritual as fundamental to Japanese tradition. Rites highlighted in this process acquired a Japanese coloration through linkage to events and practices described in early histories such as Kojiki ¥7 ¥ it. (Record of ancient events; 712) and Nihon shoki H A #&7FC (Chronicle of Japan; 720). Conceptualizations of the specific features of these rituals and the grounds for their efficacy rested ultimately, however, on accounts of li found in Chinese Confucian sources.
To explore some dimensions of these phenomena, this chapter will first examine the views of several early and mid-Tokugawa think ers regarding adoption of Confucian family ritual. “Family ritual” is understood here to encompass both ceremonial practices, such as fu nerals and ancestral sacrifices, and social norms closely associated with these practices in the Chinese context, including the observance of exogamy, agnatic adoption, and generational distinctions. While intellectual notions about practice cannot be totally separated from the tendentious question of actual custom, my concern in the opening sec tions (and throughout the chapter) is primarily with the former. Re view of the perspectives on family-related li expressed by a range of Tokugawa thinkers will lead to a brief consideration of the concept of ritual as an instrument of governance propounded by Ogyii Sorai 3X 4 1H (1666-1728). As the foremost Tokugawa critic of Song Confu cianism, Sorai set out an interpretation of the nature and efficacy of li quite different from earlier notions, which owed much to the emphases of Song thinkers. Sorai’s arguments aroused an extensive debate within the Tokugawa intellectual world regarding the relevance of li and Confucianism to Japanese society. The final sections of the chap ter will take up one outcome of this debate, the formulation in the first half of the nineteenth century of a “native” Confucian ritual by the late Mito 7K = school.
260 KATE WILDMAN NAKAI
Social Setting and Practice
Among the various categories of li, those pertaining to the central events of family life such as marriage and ancestral rites occupied a particularly important place in both the theory and practice of Chinese Confucianism. By situating individuals properly within the framework of the most basic social relationships, family-based li were held to provide an initiation into an underlying patterned structure that linked human society to the way of Heaven. It was thus with good reason that the issue of family-related li drew the attention of Song Confucians such as Cheng Yi #2 BA (1033-1107) and Zhu Xi % 2 (1130-1200). These thinkers’ concern to reemphasize the existence of a natural moral order and develop mechanisms for inculcating identification with it led them to focus as well on the formulation of appropriate family rituals.
In that upholding the proper moral norms of family life was taken to be a defining characteristic of the gentleman, observance of the li set forth in the classics and recodified in works such as Jiali %& tL (Family rituals), attributed to Zhu Xi, came to be an important part of
the self-image of the scholar-official class from the Song on. Gentry families did not always, of course, adhere to the letter of the li de scribed in such works. In funerary rites and ancestral sacrifices, for instance, Timothy Brook has pointed out a gap between Confucian ideal and actual social practice in late imperial China. Not only com moners but gentry families as well often relied on Buddhist priests to conduct funerals instead of following the prescriptions of Jiali or the early ritual classics. Both Brook and Kai-wing Chow agree, neverthe less, that the ideal of adherence to those prescriptions was an impor tant aspect of gentry status identity and was utilized to foster lineage solidarity.! In Yi-dynasty Korea, too, Confucian family rites and the social norms associated with them exerted a strong influence on elite culture. Korean Confucian scholars affirmed the universal validity of principles fundamental to Chinese Confucian li such as surname exog-
! Timothy Brook, “Funerary Ritual and the Building of Lineages™; Kai wing Chow, The Rise of Confucian Ritualism.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 261
amy and adherence to a mourning code expressing finely articulated degrees of relationship. Observance of these norms consequently be came a central element of yangban i Ht identity.
Circumstances were quite different in Tokugawa Japan. The ideal of observance of Confucian family ritual did not become a defining attribute of the elite class, the samurai, or buke iE ZF . While ceremo nial figured significantly in the lives of Tokugawa buke, it did not have an identifiably Confucian background, but was rather an eclectic mixture of elements of various origin. The lukewarm attitude of the Tokugawa elite toward Confucian li was grounded in part in the char acter of the basic unit of buke society, the house (ie &). The ie was not a lineage or household, but a corporate structure encompassing nonrelated retainers as well as those linked by blood. The organiza tional dynamics resulting from this circumstance acted as a barrier to the adoption of Chinese Confucian family li, which assumed a consan guineous lineage organized around principles of exogamy, agnatic de scent, and strict differentiation of generations. The nature of buke ceremonial testified to other features of the ie. Confucian family li were centered on the individual, whose place within the family, and thus ritual role, changed in accordance with the different stages of life and the passage of generations. The primary focus of buke ceremonial, by contrast, was the house as a component of the larger social order. By specifying particular activities, styles of dress, and such for each house or status category, buke ceremonial affirmed and reinforced the ie’s assignment to a fixed location within the hierarchy of social and political status.’
The position of Confucian scholars within Tokugawa society also bore on the degree to which observance of Confucian li was held (or not held) to be a marker of elite status. Whereas in China and Korea Confucian scholars were often themselves members of the elite, such was rarely the case in Japan. By and large confined to a professional niche on the fringes of buke society, Tokugawa Confucians could not easily dismiss surname endogamy and nonagnatic adoption, both of
2 Martina Deuchler, The Confucian Transformation of Korea. 3 On the issue of Tokugawa buke ceremonial, see Watanabe Hiroshi, “*Goikd’ to shocha.”
262 KATE WILDMAN NAKAI
which were practiced widely within the buke class, as simply the ir regular practices of an uninformed populace. While they might per sonally attempt to adopt Confucian funerary ritual, they similarly could not expect their example to exert a wide social influence. Re flecting such circumstances, Tokugawa Confucians exhibited substan tial ambiguity and diversity in their thinking about family-related li. A glimpse of this situation is offered by some of their responses to the insistence on surname exogamy, agnatic adoption, and generational distinction closely associated with Chinese Confucian family rites.
Tokugawa Views of Confucian Family Ritual
Some Tokugawa Confucians, such as Yamazaki Ansai (| #5 fd 2% (1618-82), took a rigorous position on the issue of surname exogamy. A committed adherent of Zhu Xi, Ansai argued that strict observance of the prohibition of endogamy was essential. “While the principle of
differentiation between husband and wife (fiifu no betsu F< td @ Hl) is rooted in the moral nature with which all people are innately en dowed,” Ansai declared, “men and women are easily led astray by lustful impulses.” Insistence on exogamy would help to check this tendency by reinforcing the principle of differentiation.’ Being of uni versal validity, the natural moral order, Ansai asserted, had been ex emplified in the practices of Japanese antiquity as well as the works of the Chinese sages. He thus prefaced his discussion of the principle of differentiation with examples of marriage found in the age of the gods (kamiyo 1X; ) sections of Nihon shoki. He also noted that the Chinese reports on ancient Japan found in History of the Northern Dynasties
4 Watanabe Hiroshi has constructed a useful diagram comparing a range of practices in East Asian countries bearing on Confucianism, such as the observance of exogamy and prohibition of nonagnatic adoption. The chart, published originally in Shisé 792 (June 1990), has been reproduced in his Higashi Ajia no oken to shisa.
5 Yamazaki Ansai, Yamato shogaku, p. 56.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 263
(Beishi it. #1) praised the Japanese for upholding the principle of ex ogamy. Despite such claims, both the kamiyo sections and the later human emperor sections of Nihon shoki and Kojiki were rife with in stances of brother-sister, aunt-nephew, and uncle-niece marriage. Given this, Ansai understandably did not try to further specify the place of the principle of exogamy in early Japanese history.
In insisting on strict observance of exogamy, Ansai was clearly in the minority. His contemporary Kumazawa Banzan fit {R # (Ll (1619- 91) offers in this instance a more representative stance. Having been instituted in China only from the Zhou, the prohibition of endogamy
“even unto a hundred generations,” was not, Banzan claimed, a uni versal aspect of the Way. To try to force such an extreme principle on Japanese society in the absence of any historical foundation or social consensus for it would be counterproductive. People should first be educated gradually in the broad principles of the Way; once such knowledge was widely established, those in positions of authority could take steps to institute the more moderate restriction of endogamy within the five mourning relations customary in China prior to the Zhou.®
Tokugawa Confucians were more sensitive to nonagnatic adop tion. Tokugawa society at large showed little concern about adopting a daughter’s son or someone of an entirely different lineage to continue a house that lacked an heir in the agnatic line of descent. According to Miwa Shissai = # #4 77 (1669-1744), in his day no more than “three out of ten” daimyo did not resort to such arrangements, while among the general samurai and commoner population the practice was even more prevalent.” Although adoption of a daughter’s son and other forms of nonagnatic adoption were not unknown in China, too, from the perspective of Confucian theory, such practices presented signifi cant problems. Since an ancestral spirit was assumed to respond only to sacrifices offered by its agnatic descendant, nonagnatic adoption was held to interfere with the proper ritual treatment of such spirits. It thus was tantamount to an act of filial impiety. Given such assump tions, Chinese works on li presumed agnatic adoption as the norm.
6 Kumazawa Banzan, Shiigi washo, pp. 92-100.
7 Miwa Shissai, “Y&shi ben o benzu,” p. 459.
264 KATE WILDMAN NAKAI
Tokugawa Confucians likewise, as James McMullen has shown in an illuminating discussion, came down largely on the side of adherence to Confucian principle.?
There were those, however, who expressed doubts. One was Ku mazawa Banzan, who as with the issue of endogamy, called for a gradualist, low-key approach. Another was Miwa Shissai. Among the samurai class, Shissai pointed out, where a house was in fact a socio political unit comprising the lord and his vassals, the welfare of the latter depended upon the continuation of the house in name, not bio logical substance. To allow a house to be terminated on purist grounds because there was no appropriate agnatic heir would unjustly set adrift the vassals who had served it faithfully since the time of their ances tors. To justify the priority he gave continuation of the house as a cor porate entity over maintenance of the blood link between ancestor and descendant, Shissai offered a twist on the Chinese Confucian notion of the “exhaustion of kinship” figuring in mourning regulations. “Blood ties” (kechimyaku ifi ik), he noted, “no matter how important . . . come to an end anyway in five generations.” A similar reordering of Chi nese Confucian views of the relationship between house and kinship was invoked to support the deliberate choice of an unrelated heir. Given the importance of preservation of “the house inherited from one’s ancestors,” Shissai argued, it would be justifiable to pass over a child who was not up to that task and select instead a worthy outsider as heir to the house. !?
Miwa Shissai may represent an extreme position among Toku gawa Confucians on the issue of nonagnatic adoption. Yet when it
8 I. J. McMullen, “Non-Agnatic Adoption™; see also Watanabe Hiroshi, Kinsei Nihon shakai to Scgaku, pp. 125-34.
9 The original Chinese idea was that individual rites were no longer per formed for an ancestor more than five generations distant, not that the blood tie as such came to an end.
10 Miwa Shissai, “Yoshi ben o benzu,” pp. 459-60. Shissai proposed that this was what the ancient Chinese sage Yao had done in choosing the virtuous Shun in place of his own son as his successor. In that this event involved po litical succession rather than the continuation of a lineage, the argument fur ther illustrates Shissai’s blurring of the concept of “house” as a sociopolitical unit and as a kinship group.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 265
came to observance of the generational distinctions fundamental to the Chinese Confucian notion of the lineage as a ritual unit, even those who held out for the prohibition of nonagnatic adoption and endogamy were prepared to compromise in ways that would have startled their Chinese and Korean confreres. According to Confucian theory, an adopted heir should not only be of the same surname as the missing descendant whose place he filled, in principle he should be of the same generational relationship to the ancestral spirits he was expected to serve. The orientation toward house rather than lineage in Tokugawa society blurred such considerations. Ogyu Sorai, for instance, criti
cized the prevalence of nonagnatic adoption in the samurai society of his day. But, to avoid it, he was willing to countenance a reversal of generational relationships that from the perspective of Chinese Confu cian norms was equally irregular. There was nothing wrong, he ar gued, with designating an uncle as successor to a nephew without an heir. To the contrary, it was a recommendable means of preserving a house without resorting to nonagnatic adoption.!!
Asami Keisai #& 5 #1 3 (1652-1711), a disciple of Yamazaki Ansai who counted otherwise among the strictest Tokugawa upholders of Chinese Confucian family norms, took a similarly cavalier attitude toward the observance of generational proprieties within the context of
agnatic adoption. In answer to a query about the management of an cestral rites in a family where a younger brother had succeeded a de ceased elder brother and had then been succeeded in turn by a second elder brother, he gave precedence to the implications of succession to the position of househead (katoku ZF ). “Whatever [the actual rela tionship],” he declared, “if one is heir to the house, the order of gen erations [of one’s predecessors] becomes that of ‘great-great grandfather, great-grandfather, grandfather, and father’ (kdsdsoko & #1 #&, Chinese gaozeng zukao). The arrangement of the spirit tablets should follow that order.” Even if the object of the sacrifices was in
1 Ogyii Sorai, Seidan, pp. 405-6. For the Chinese prohibition of such ar rangements as contrary to the principle that heirs should be chosen from the “range of agnatic kin of appropriate generational relationship” (dosé shoboku sot5 [7] 77% Bl #3 Hl &, Chinese tongzong zhaomu xiangdang), see Shiga Shiizd, Chigoku kazoku hé no genri, pp. 313-18.266 KATE WILDMAN NAKAI
fact the younger brother of the sacrificer, since the current head was his brother’s successor as head of the house, “there is no problem in [the head/elder brother] sacrificing to him as if he were his father.”!2
Similar ambiguities figured in Tokugawa Confucian approaches to funerary ritual and ancestral sacrifices. A key feature of such rites in the Chinese context was the wearing of different styles of mourning garb to indicate the degree of relationship with the deceased. Even those who sought to promote the practice of Confucian funerals in Ja
pan in place of Buddhist rites tended to downplay observance of such distinctions. This was the case, for instance, with Wakabayashi Kydsai FE AK 58 7 (1679-1732), another follower of Ansai. Kyosai’s Karei kunméso 7 FL J % Hi (An introductory commentary on Family Ritu
als), a Japanese-language commentary on Jiali, was probably the most important vehicle for dissemination of knowledge of that text in Toku gawa society. In this work, commenting on Jiali’s detailed prescrip tions regarding mourning wear, Kydsai acknowledged that it was a universal principle to show grief by donning a more somber or rougher style of dress than usual. But, he asserted, each country should hold to its own distinctive customs in that regard. For Japanese to follow in stead the specifications of Jiali regarding mourning garb would in fact “be highly disrespectful (furei 4 #L, lit., ‘contrary to li’) toward our own country.” Ogyu Sorai, who produced a radically abbreviated ver sion of Jiali, likewise did not recommend adoption of differentiated styles of mourning dress comparable to those found in China.!3
12 Asami Keisai, Jowa sakki, p. 547. See also Tajiri Yaichird, “Keisai, Kyosai to Bunké karei,” pp. 19-20.
13 Wakabayashi Kyosai, Karei kunméso, kan 2: 12a; Ogyi Sorai, Sorei ryaku, p. 384. The lack of concern for differentiated mourning dress was true as well of one of the most overtly “Confucian” institutions of the Tokugawa period, the mourning code (bukkiryé fit =: 45) promulgated by the bakufu at the end of the seventeenth century. Following earlier codes established under Chinese influence in the eighth century, the Tokugawa code made the period of mourning substantially shorter than it was in China and Korea and did not stipulate a specific style of mourning dress. It similarly did not make nearly so finely graded distinctions between different degrees of relationship as did the original Chinese mourning system. For a recent, excellent account of the history of Japanese mourning codes, see Hayashi Yukiko, Kinsei bukkiryé no
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 267
Tokugawa Confucian approaches to the central object of ritual at tention in Confucian ancestral rites, the spirit tablet (shinshu f# ¥, Chinese shenzhu), show other repercussions of a lack of attunement to degrees of relationship. According to Jiali, the spirit tablet was to be inscribed with the name of the deceased at the time of interment of the coffin. Installed in the family offering hall (shido fia &, Chinese ci tang), it served as the repository for the soul of the deceased and for the next several generations was the object of individual sacrifices by the deceased’s descendants. With each change in generation the tablet was reinscribed (to indicate, e.g., that what had been the tablet of a father was now that of a grandfather) and moved to a new place in the ancestral hall appropriate to its changed relationship with the sacrifi cer. Eventually, after “kinship was exhausted,” the tablet was removed from the offering hall to the grave site.
These practices reflected the Chinese concept of the soul as gradually dispersing as its link to the current generation of family members became more distant. They also indicated the ways in which ritual rested upon and actualized norms of generational hierarchy. Through the reinscription and relocation of the tablets of his ancestors, the sacrificer, one might argue, acquired a heightened awareness of his precise position relative to his ancestors and kin. He likewise experi
enced in a tangible form the underlying pattern of Heaven and Earth of which the graded relationships within the family were a fundamental part. We see here a prime example of premises regarding the efficacy of li as a means of aligning the individual with the Way. It is precisely these aspects that fade from view in Tokugawa discussions of spirit tablets.
Wakabayashi Kyosai, for example, while nominally adhering to Jiali’s premise of the gradual “exhaustion of kinship,” tended, in fact, to regard the spirit invested in the tablet as imperishable. This is evi dent in his reference to the tablet as shintai # {4 , the term used in the native Japanese tradition to indicate an object embodying a deity, and in his identification of the spirit as “the divinity (shinmei #8f) that exists eternally.” Underlining this view of the shinshu as shintai, the
kenkyii. See also Okada Jusei, “Kibukukd”; and Koji ruien, Reishiki bu 2: 577-926.
268 KATE WILDMAN NAKAI
sacred locus of an eternal spirit, Kyosai exhibited discomfort about the Chinese custom of reinscribing the tablet each generation. To write the name of the sacrificer “on the shintai,” as was stipulated by Jiali, “is distasteful,” he demurred, and by inscribing only the name and title of the deceased on the tablet and not his generational relationship to the current head of the house, one could avoid the “troublesome” practice of reinscribing the tablet each generation. !4
Sorai and Kyosai proposed other modifications as well of ele ments of the ritual prescribed by Jiali that they regarded as potential obstacles to the adoption of Confucian funerary nites in Japan. One was the extravagant expression of grief. Repeated “wailing to the full extent of one’s grief” figured centrally in Jiali’s description of the se quence of actions involved in preparing the corpse for burial and the burial itself. Kyosai suggested that this, too, was an area where it would be better to follow national custom. “While the feeling of grief is no different in China and Japan, each country has its own appropri ate customary manner of expressing it.” Sorai agreed that it would be difficult in Japan to implement wailing and calling for the soul of the deceased to return to the body, “particularly in urban quarters.”!® Sorai
14 Wakabayashi Kyosai, Karei kunmaso. kan 2: 10a-b: kan 3: 16b-17a. Cf. Zhu Xi. Chu Hsi’s Family Rituals. p. 123.
15 Wakabayashi Kydsai, Karei kunméso. kan 2: 2b. Kydsai made similar comments on the passage stipulating the actions of the mourners at the laying out of the body: “The presiding male and female mourners embrace the body, wail, and beat their breasts” (Zhu Xi. Chu Hsi’s Family Rituals. p. 83). “Al
though this is an ancient rite in China. it is not something that should be done here. It is better to preserve an attitude of reverence and not lay one’s hands [on the corpse] in an overly familiar manner.” Kan 2: 14b. Tajiri Yaichird incisively discusses approaches to Confucian funerary practices among fol
lowers of Yamazaki Ansai in his article on Asami Keisai and Wakabayashi Ky®dsai cited above. See also his “Raisai. Tekisai 10 Bunké karei,” and the overview that he presents in his chapter in vol. 13 of Nihon no kinsei. On Confucian funerary rites in the Tokugawa period and their modification, see also Kondé Keigo. Jusé to shinsé. On the views of Tokugawa Confucians on the issue, see Watanabe Hiroshi, Kinsei Nikon shakai to Sogaku, pp. 164-70.
16 Ogyii Sorai, Séreirvaku, p. 384. Sorai allowed. however, that should there be “gentlemen fond of li living in the countryside.” it would be appropriate for them to adopt this practice. Sosai benron, a critique of con temporaneous Buddhist-influenced funerary customs held to be an early work
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 269
and Ky®osai also shared the view that it would not be practical for their countrymen, who by and large did not live in extended families and lacked the retinue of servants taken for granted by Jiali, to adhere to the stipulation that as a sign of their inconsolable grief the sons and daughters of the deceased should “refrain from eating for three days.” As Kydsai put it, while what was “fundamental to li is the same past and present, near and far,” when it came to enacting that fundamental essence, it was necessary “to take time and circumstance into consid
eration and make adjustments in accordance with social station and the degree of wealth or poverty.”!”
Kyosai and Sorai recommended adjustments of this sort in Con fucian funerary ritual with the aim of facilitating its acceptance by their countrymen. Despite their efforts, practice of such rites remained confined to a small circle of the informed and committed; it never be came part of either the actuality or image of elite culture, let alone general custom. We can suggest several reasons why this was so. One was that by the latter part of the seventeenth century, people were ex pected to be registered as parishioners (danka 1 3%) of a recognized Buddhist temple as evidence that they did not harbor subversive reli gious inclinations (meaning primarily, but not exclusively, Christian ity). A household’s affiliation with a temple was, in turn, largely con solidated by having that temple perform funerary and memorial ser vices for deceased family members. In such circumstances, utilization of alternative funerary ritual came to be regarded with suspicion.
by Kumazawa Banzan, takes a similar position. Introducing in a simplified form some of the practices set forth in Jiali, this work offers an explanation of the rationale behind the Chinese practice of calling for the soul of the de ceased to return to the body. It also notes, however, that as Japanese would likely find such a practice “startling,” it would be difficult to implement in Japan. Kumazawa Banzan, Sosai benron, pp. 97-98. In later years Banzan expressed stronger reservations about the practicality and cost of trying to implement Chinese Confucian funerary practices in Japan. “Even should a sage ruler appear,” he declared, “it would not be appropriate to extend the Confucian practices stipulated in Jiali down to the level of the ordinary popu lace.” Shiigi gaisho, p. 182. Okayama, where Banzan had earlier served, was one of the few domains to attempt to promulgate Confucian funerary ritual.
17 Wakabayashi Kyosai, Karei kunmaso, kan 2: 4a-b. See also Ogyu Sorai, Sereiryaku, pp. 384-85.
270 KATE WILDMAN NAKAI
One might speculate that the deep-rooted association of death with pollution was a further impediment to acceptance of Confucian funerary ritual wherein the officiator was not a priest but the family of the deceased. In that situation the burden of dealing with death pollu
tion and safely stabilizing the spirit fell upon the chief mourner and his relatives rather than a ritual specialist. Doubtless many regarded the professional Buddhist priesthood, with its rich panoply of rituals for pacifying spirits and long association with native forms of kami ## worship, as better equipped to manage such demanding tasks.’® The consequences of efforts to naturalize Confucian funerary rites by play
ing down elements such as wailing perhaps indicate yet another reason why such rites did not become established in Japan. While de emphasis on wailing served to soften an aspect that Tokugawa Confucians evidently found alien, it also eliminated one of the few emotionally cathartic elements from what was otherwise (at least to modern eyes) a quite flat and monotonous liturgy. Given that even in China many showed a preference for the emotionally richer Buddhist
rites, it is not surprising that the expurgated version of Confucian fu nerary ritual produced by Tokugawa Confucians did not succeed in gaining a following. Whatever the reasons, the failure of such ritual to take root in Japan also meant the loss of perhaps the most significant arena for the practice of family-related li.
In pointing to these ambiguities and gaps in the attitudes of To kugawa Confucians regarding family-related Confucian li, we need, as noted above, to be careful not to measure the actual practice of one society against the idealized norm of another. We should also keep in mind that within the Confucian tradition as a whole, li, while regarded as manifesting an underlying natural order, were held to be human constructions. It was thus taken for granted that changes in time and setting might necessitate modification of various of their particular features. As the ancient ritual classic Liji #L 3C (Records of ritual) put
18 Anxiety about such matters perhaps was a factor in Kydsai’s strong concern for stabilizing the spirit of the deceased immediately after death. This was another point where his recommendations diverged from those in Jiali. See Wakabayashi Kyosai, Karei kunmadso, kan 2: 3a; kan 3: 28b, 29b-30a. See also Tajiri Yaichird, “Keisai, Kyosai,” pp. 26-27.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 271
it, “Li are rooted in heaven; activated, they extend to earth . . . chang ing, they accord with the times. . . . Li embody what is right (yi & ). If it fits what is right, even a li not known to the kings of old may be used to achieve what is right.”!?
Following this principle, the official codes of different dynasties in China had made changes in the degree of mourning stipulated for one relative or another. Cheng Yi and Zhu Xi justified quite major departures from the li set forth in the classics on the grounds that the forms of li might properly be altered to make them more efficacious in achieving “what is right.” Notably, they did away with status distinc
tions in the number of generations of ancestors to which one might offer sacrifices, and specified that all families should hold rites for four generations of ancestors.2? The effect was to downplay the mani festation of status distinctions that were no longer of immediate social relevance and to emphasize instead the absolute nature of the impera tive of filial piety, the actualization of which was one of the primary purposes of ancestral ritual. The alteration was consistent with the universalizing thrust of the Song Confucian view of human nature and approach to the inculcation of morality.
Tokugawa Confucians who criticized li of Chinese origin as overly rigid or who tried to find some accommodation between Chi nese li and contemporaneous Japanese social practice often (as with Wakabayashi Kydsai above) alluded to the notion that li should be adjusted in accordance with the exigencies of time and place. When it came to li intended to express and confirm fundamental family rela tionships, however, it was not always easy to distinguish between le gitimate modifications to “accord with the times” and changes that would result in a deviation from “what 1s right.” The degree of blur ring of norms of lineage relationship that resulted from the adaptation of family-related li to Tokugawa circumstances arguably would have been seen by Chinese and Korean Confucians as falling in the latter category. It also unquestionably weakened the basis for perceiving li
19 Liji 9: Liyun $L 38.
20 See Zhu Xi, Chu Hsi’s Family Rituals, pp. xx-xxi: and Patricia Ebrey, Confucianism and Family Rituals, pp. 61-64, 106-7.
272 KATE WILDMAN NAKAI
as an interlocking, coherent structure through which the individual might experience the way of Heaven.
Reconceptualizations of Ritual:
Ogyu Sorai and His Successors
Deviation from the ideas of li advocated by Song thinkers such as Zhu Xi may be found not only in Tokugawa discourse about family li but also in notions of the basis for the efficacy of ritual. It is most apparent in the writings of Ogyu Sorai, whose systematic recasting of the premises of Confucianism produced a sea change in its conceptual dynamics. Sorai directly challenged the assumption that through the enactment of li the individual realized an innate capacity for alignment with a natural moral order. In his view the very notion of an innate order encompassing both the human realm and Heaven was a misin
terpretation that had resulted from ambiguities in the writings of suc cessors of Confucius such as Zisi + fB and Mencius. The Way, Sorai held, was not something “natural.” It was nothing other than the total ity of li and other institutions created by the ancient kings with the aim of bringing order to human society. Like those entities it thus was it self a human construction and necessarily external to the individual. From this perspective Sorai took issue with Zhu Xi’s observation that “li represent the principles of differentiation innate to Heaven and provide the standards for the regulation of human affairs.” The asso ciation of li with the notion of a “heavenly principle” perpetuated, he said, the mistaken perception that li were somehow part of human na ture (sei t%, Chinese xing).2!
21 Ogyii Sorai, Benmei, p. 72. Zhu Xi’s remark is from his commentary on Analects 1:12. For a recent discussion of Sorai’s understanding of li relative not only to Song conceptions but also those of earlier thinkers such as Xunzi, see Olivier Ansart, L ‘Empire du rite.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 273
Sorai’s condemnation of the Song approach to li was not a rejection of the importance of li per se. To the contrary, he strongly af firmed the value of li as he believed them to have been employed by the ancient kings. The ancient kings created li, Sorai argued, because they recognized that “words” (i.e., the kinds of rationalizing, didactic explanations favored by Confucians of later ages) were insufficient to teach people how to live together peacefully and harmoniously. The virtue of li as opposed to words was their transforming effect. Carried along by the performative power of li, people were induced without realizing it to act within the normative structures established by the ruler.?2 This premise led Sorai to particularly emphasize li in the sense of state rites. “The way of the ancient kings was rooted in none other than reverence for Heaven and the spirits.”2> Whenever the rulers of the three ancient Chinese dynasties (the Xia, Shang, and Zhou) took action on even minor governmental matters, they thus endowed their measures with divine authority by “conducting rites to their ancestors in conjunction with those to Heaven, issuing measures as at the com mand of Heaven and their ancestors, and carrying them out in accor dance with the indications gained through divination.”?4
In Sorai’s view, the postulation of an innate principle that linked human beings to a heavenly order had led the Song thinkers to take an inflated view of their own self-importance. This was reflected in their disregard of the awe toward Heaven and the spirits emphasized by the ancient kings; instead they had stressed a program of introspective reflection intended to lead to identification with heavenly principle. Similarly they had presumptuously attempted to articulate a “fixed, immutable system of 1i.”%> In so doing, Sorai argued, Cheng Yi and
2 Ogyu Sorai, Benmei, p. 70.
23 Ogyii Sorai, Bends, p. 29.
24 Ogyi Sorai, Benmei, p. 73. Whether Sorai believed in the objective ex istence of human spirits is open to question. Hiraishi Naoaki argues that Sorai, who speaks of the sages as “establishing” (37) spirits, in fact regarded the spirit—the object of rites—as well as the rites themselves as a construc tion of the sages. The sages established both as a means of instructing people in the proper norms of family life. Hiraishi Naoaki, “Soraigaku no saikdsei,” pp. 95-99. See also Benmei, pp. 132-33.
25 Ogyii Sorai, Benmei, pp. 73-74.
274 KATE WILDMAN NAKAI
Zhu Xi had failed to grasp one of the most important features of the ritual structures devised by the rulers of antiquity: li were specific and concrete, designed to fit the needs of a particular age, not universal and general 2%
Sorai’s critique of Song Confucianism and his theory of the power of ritual had a wide-ranging impact on Tokugawa ideas about li. On the one hand his deconstruction of Song ideas further undermined the already shaky place in Tokugawa thought of li as activating an alignment between the individual and an abstract natural moral order. On the other, his acclamation of the ancient kings’ use of awe for Heaven and spirits as a strategy for governing suggested alternative ways of thinking about the social and political functions of ritual. It also raised the question of the meaning for Japan of the specific li they had created.
Sorai’s position on that question was open to varying interpreta tions. He professed absolute faith in the sages of ancient China and their achievements and held that Japan had not produced any figure of comparable stature or accomplishment. From that perspective he took a critical view of Japanese society, both historical and that of his own day. Relative to the civilization of China, Japan remained in the posi tion of one of the four barbarians. This did not mean, however, that Japanese rulers should attempt to rectify the situation by adopting as is the 1i and institutions created by the ancient kings. Having been de vised to suit particular circumstances, those li and institutions could not be applicable in their original form to all times and places. The three dynasties, Sorai pointed out, had each employed different li; fur ther, as recorded in Liji, the ancient kings had recognized that as a matter of course the customs of the people of different regions varied, and had not tried to impose a single uniform standard on them.?” Even if the current situation called for reform, it thus would not be appropri ate, nor in accord with the stance of the ancient kings, for Japanese rulers to adopt the particularities of ancient Chinese li. The most that could and should be done was to examine the records of the institu
26 Ogyit Sorai, Rongocha, 3: 399; 4: 548.
27 Ibid., 3: 399; Sorai makes these observations in his commentary on Analects 1:12, with reference to Liji 2: Qulixia #1 tL T and 5: Wangzhi E i].
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 275
tions of the ancient kings as exemplary models—in contemporary par lance, as case studies—of how to approach the challenge of bringing peace and order to society.
Sorai himself did this to a certain extent with his recommenda tions for administrative and policy changes in Seidan B #¢ (Remarks on government) and Taiheisaku KX F # (Provisions for an age of peace). Those works, however, focused primarily on governmental institutions. They did not really address the issue of a system of ritual or the question how rulers of his own time might utilize the awe for Heaven and the spirits that had been the foundation of the governance of the ancient kings. That matter was eventually to receive its fullest and most notable exploration in the first several decades of the nine teenth century at the hands of scholars of the late Mito school. Their pursuit of this issue owed much to Sorai’s ideas about the power of ritual.2® Their approach to implementation of that power was likewise shaped by the intellectual repercussions of Sorai’s critique of Song thought and affirmation of China as the sole example of true “civiliza tion.”
Sorai’s acclamation of Chinese civilization evoked a strong reac tion. Many saw it as implying a denigration of Japanese tradition. The writings of his senior disciple Dazai Shundai K 3 & © (1680-1747) served to draw the lines between the camps even more sharply. Carry ing further Sorai’s rejection of the notion of an innate moral nature, Shundai argued that human beings “in their natural state are no differ ent from beasts.” While this had been true in China as well as Japan, in the case of China, at a certain point sages had appeared and had created the li and the norms of behavior that had made it possible for society to rise above the brute state of nature. The Japanese, too, even tually had been able to become more civilized, but this was due en tirely to the introduction of the teachings of the Chinese sages. That no Japanese ruler had achieved anything comparable was evident from the lack of native Japanese readings for the characters “benevolence
28 On the links between Sorai and the late Mito school, see Bitd Masa hide, “Kokkashugi no sokei toshite no Sorai,” and idem, “Mitogaku no toku shitsu.”276 KATE WILDMAN NAKAI
and propriety, ritual and music, filial piety and deference” (jingi rei gaku kotei {— 3% #L 28 3 #8, Chinese renyi liyue xiaoti).2® The most sophisticated and influential response to this argument came from the Kokugaku [# *¢ scholar Motoori Norinaga # J& H & (1730-1801). Norinaga turned around the implications of the assump tion that the sages had created li. If li were a human construction, they could not carry an absolute value, he argued. To the contrary, many were problematic because they required excessively strict norms of behavior that ran counter to human feeling. Family-related li stood as an example. In a degenerate society such as China it might have been necessary to impose rigid standards of behavior such as the prohibition of endogamy, but there was no reason to extend such principles to Ja pan.’ The same applied to Chinese mourning codes. The whole idea of a fixed schedule of mourning was, he held, a typical instance of Chinese artifice and formalism that “only served to encourage hypoc risy.” Rather than keep up a pretense of piety, a son with no deep feel ings for his parents would do better to abandon mourning before three years had elapsed. A truly devoted son, on the other hand, should not be expected to contain his mourning within the limits set by a formal schedule.’
Norinaga did not stop at a rejection of Chinese Confucian li as ar tificial and formalistic. He extended his challenge to the notion of the Way. Sorai and Shundai had taken the Way as well as li to be the crea tion of the sages and thus a human construct, not something innate or natural. They had condemned the efforts of Cheng Yi and Zhu Xi to substantiate through metaphysical exegesis the existence of a natural moral order. The Song thinkers’ positing of such an order was no more, they held, than the projection of their own imaginings. Taking this argument a step further, Norinaga posed a question: If the formu lations of the Song thinkers were false, in what regard were the actions of the sages in fabricating the so-called Way any different? There was no reason for the Japanese to accept their invention as valid. They did
29 Dazai Shundai, Bendé sho. pp. 56-66.
30 Motoori Norinaga, Naobi no mitama, pp. 305-6. See also idem, “The Way of the Gods,” p. 38.
31 Motoori Norinaga, Tamakatsuma, pp. 114-15.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 277
so simply because long exposure to Chinese writings and customs had left them infected with a “Chinese mentality” (karagokoro 1 &).32 Despite his radical deconstruction of Confucian norms, Norinaga was circumspect about proposing alternatives. He noted, for instance, that foreign systems of thought such as Confucianism or Buddhism had developed “useless empty theories” about the principles of Heaven and Earth and what happened after death with the aim of providing people with “a degree of peace of mind” (kotemae no anshin I~ F Hii MZ >). “Shinto” scholars of later days (whom Norinaga regarded as thoroughly contaminated by “Chinese mentality”), had fabricated their own version of “peace of mind” by borrowing from these Confucian and Buddhist ideas. Such solace, however, could never be other than illusory. True peace of mind could come only from recognizing the falsity of such constructions and accepting the stark reality set forth in the accounts of the age of the gods in Kojiki and Nihon shoki: what awaited all, high and low, good or bad, at death was simply the putrifi cation of Yomi, the land of the dead described in the kamiyo accounts. To be sure, Norinaga acknowledged, it was the nature of human sen timent to worry about death. Without some explanation of why things were that way, few people could be expected to find sufficient the true peace of mind offered by the ancient Japanese texts. Yet to try to pro vide such an explanation would be to repeat the error of the Confuci ans and Buddhists, for the exigencies of life and death were in fact beyond human understanding.3?
For Norinaga personally the cool existential stance of refraining from offering false comfort may have been the only appropriate solu tion to the human dilemma, but few of his followers were able to adhere to that position. Hirata Atsutane = Ml & fil (1776-1843), for instance, while similarly starting from criticism of “Chinese mental ity,” arrived at a quite different conclusion. Atsutane took indignant exception to Dazai Shundai’s assertion that Japan had attained a de
32 The classic analysis of the intellectual relationship between Sorai and Norinaga is Maruyama Masao, Nihon seiji shisoshi kenkyi, which has been translated into English as Studies in the Intellectual History of Tokugawa Ja pan.
33 Motoori Norinaga, Témonroku, pp. 525-27.
278 KATE WILDMAN NAKAI
gree of civilization solely through exposure to the teachings of the Chinese sages. The fundamental principles of human morality were universal, Atsutane argued. To say that the li and teachings of the sages were needed to actualize them was absurd. Insofar as those teachings had something valid to say about the basic human relation
ships what they offered was obvious. To revere them as exceptional was like expressing gratitude for being saved from death from starva tion because an official in formal dress had solemnly announced that people should eat when they were hungry. And to go beyond that basic
level and try to follow the minute, stultifying (ikitaru kokochi mo su mai = % 7-3 > #1 b+ FV) prescriptions of Confucian li was beyond human capacity.
Atsutane also recognized, however, that even if the principles of human morality were universal, adhering to them in the face of the misfortunes and setbacks encountered in the course of ordinary human existence was not always easy. To do so, some encouragement was necessary, some promise of a “peace of mind” beyond the acceptance of the inevitability of death and decay recommended by Norinaga. A tsutane therefore took it as his mission to offer an outlook that by alle viating people’s worries about their fate after death would motivate them to act in this life as obedient and loyal members of Japanese so ciety.’ The intellectual system he devised was in fact a bricolage of elements drawn from Chinese, Buddhist, and even Christian sources, as well as key native works. Concern for li and the observance of ritual norms played little part in it; instead it emphasized the ideas of judg ment and reward after death.
34 Hirata Atsutane, Kamdsho, pp. 128, 119.
35 Hirata Atsutane, Tama no mihashira, p. 12.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 279
The Mito Formulation of a
Native Confucian Ritual
The reworking of Sorai’s ideas about ritual by the late Mito school during the first several decades of the nineteenth century took place against the backdrop of the developments sketched above. The late Mito scholars saw the stability of society being eroded by social and spiritual divisions, which they feared might weaken Japan’s ability to withstand the gathering threat of the Western powers. They agreed with Atsutane that to shore up social order some sort of spiritual rein
forcement was needed. But they were critical of his solution. In their view, the eclectic mixture of ideas Atsutane had put together to pro vide the populace with “peace of mind” would only exacerbate the problem. By pandering to their worst instincts, it would encourage people to wander into heterodox paths and to abandon, not consoli date, a commitment to duty. The crucial task, as the Mito scholars saw it, was to formulate a spiritual and moral structure that would simulta neously reinforce social unity and order, compel the wholehearted al legiance of the populace, and inspire in them an attitude of devoted service to their superiors.
The Mito scholars took for granted that this structure had to be grounded on the basic Confucian social norms. They also recognized that Sorai’s and Shundai’s depiction of the Way as the creation of the sages had provided Kokugaku scholars such as Norinaga and Atsutane with a weapon for deconstructing the coherence of Confucian thought about society. Determined to reassert the universal validity of Confu
cian norms, the Mito scholars insisted that the principles of the Way recorded in the Confucian classics were “natural,” not human con structions, and that they thus applied to Japan as well as China.
While holding the Way to be natural, the Mito scholars shared Sorai’s distaste for the metaphysical theorizing and introspective self cultivation used by the Song thinkers to underwrite that proposition. Such an approach, they agreed, encouraged empty speculation and impeded the kind of instinctive, active social commitment they sought. To substantiate the naturalness of the Way, the Mito scholars instead relied heavily on the intuitive argument that there was an “intrinsic match” (ang ME 4) between the principles of human relations eluci-
280 KATE WILDMAN NAKAI
dated by the sages and the original customs of Japanese society.>® The difference lay in how those principles had been transmitted. In China the li established by the ancient sages to manifest the crucial points of the Way had been explained in some detail in the classics. In Japanese antiquity, by contrast, those principles had been expressed in sym
bolic, nonverbal form.
The Mito scholars’ assertion that in antiquity the import of the fundamental principles of human relations had been conveyed to the populace through nonverbal means dovetailed with another dimension of their outlook: a belief in the efficacy of ritual that they inherited from Sorai together with suspicion of Song metaphysics and introspec
tion. As the leading late Mito thinker Aizawa Seishisai 2 {R IE & (1782-1863) put it in an observation that clearly built on Sorai’s ideas, the people could not be made to understand, but they could be made to follow. To do so, however, a “tool” (gu FH.) was necessary, and the only thing that could function effectively as such a tool was li. If li were used to “carry” the people, there would be no need to rely on strident and ultimately ineffective sermonizing. The people would in stinctively “keep within the current” of what was correct.’
The combination of the Mito scholars’ interest in li as an instru ment of rule with their insistence that there was an “intrinsic match” between the Way of the sages and the original customs of Japanese society had important repercussions for their approach to both native tradition and Confucianism. With respect to the arguments of Koku gaku scholars such as Norinaga and Atsutane, the claim of “intrinsic match” served to “naturalize” the Chinese Confucian Way and li by reference to parallel, “known” elements of native tradition. Equally it became a device for a selective rereading of what was native, a read ing that in fact relied heavily on allusions to the Chinese classics to describe entities and events of Japanese provenance.3® The Mito inter pretation of “Shinto” provides a succinct example.
36 See, for instance, Aizawa Seishisai, Kagaku jigen, pp. 1a-2a, 37a. 37 Aizawa Seishisai, Kagaku jigen, p. 26a.
38 The Mito scholars referred to the coming of Buddhism to justify their dependence on Chinese sources to clarify the native Way. While in antiquity nonverbal forms had sufficed to convey what was proper, Buddhism had con taminated native custom and made people devious. To recover what was truly
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 281
“Shinto,” while long assumed to be fundamental to native tradi tion, had in fact been subject to constant redefinition and embellish ment over the centuries. By the Tokugawa period, what was called Shinto—whether practice or theory—consisted largely of a mixture of elements drawn from esoteric Buddhism, Chinese cosmology, and Song metaphysics. Both Ogyil Sorai and Dazai Shundai pointed this out in dismissing the arguments of many Tokugawa Confucians that Shinto could be identified with the Way of the sages. An oft-cited element in this identification was the passage in Yijing 5 #& (Classic of changes) stating that “the sage uses the divine way to give instruc tion (LA ## jE 3% #0), and the entire world submits to him.”3® Com menting on this passage, Shundai noted that utilizing awe for the various unfathomable spiritual forces at work in the universe was fun damental to the approach to governing the populace devised by the sages. Yijing therefore described the sages as basing the teaching of social norms on shendao #1 3H (the divine way). That did not mean, however, that shendao was equivalent to the Way; it was rather a sub set, one feature, of it. Even less did it mean that the shendao of the sages was the same as Japanese “Shinto,” despite the fact that the lat ter was written with the same characters. What the Japanese priestly lineages and scholars of later generations called Shinto was spurious, concocted by them for their own purposes out of “seven or eight parts Buddhism and two or three parts” Song Confucian metaphysics.4?
Norinaga had accepted as valid Shundai’s dismissal of the “Shinto” of their own time, but he had gone on to argue that a true Shinto had existed in antiquity. This original Japanese Way was dif ferent, he claimed, from both the eclectic, spurious latter-day Japanese “Shinto” and the equally specious, contrived shendao of China. Con fronting this array of definitions of Shinto, the Mito scholars set forth a position distinct from them all. Agreeing with Shundai and Norinaga that what was commonly called Shinto was corrupt, they rejected
native, it thus was necessary to turn to Chinese sources where the same prin ciples could be found recorded in written form.
39 Richard Wilhelm, The I Ching, p. 486 (kan #l, Chinese guan, hexa gram; translation slightly modified).
40 Dazai Shundai, Bends sho, pp. 44-46.
282 KATE WILDMAN NAKAI
Norinaga’s dismissal of the shendao of the sages as contrived and Shundai’s assertion that shendao was unknown in Japan. To the con trary, they declared, despite differences in detail, the shendao of the sages and the Shinto of Japanese antiquity were innately alike: both were founded on the recognition that to convey moral norms through rituals to Heaven and Earth and ancestral rites was the most effica cious means of securing the submission of the populace.! Building on such premises, the Mito scholars drew from Yijing to describe the place of Amaterasu X Bf as the font of Japanese tradition. “In antig uity,” Aizawa wrote, “the Heavenly Ancestor (Tenso X ff, i.e. Amaterasu), using the divine way to give instruction, established the norms of human behavior by clarifying loyalty and filial piety.”?
In equating Shinto with the conveying of moral norms through rituals to Heaven and Earth and ancestral rites, the Mito scholars fol lowed an observation made by Sorai. Sorai, like Shundai, had dis missed the Shinto of his day as a spurious recent concoction, but he had allowed that in antiquity Japanese rulers, too, had made the “con ducting of rites to their ancestors in conjunction with those to Heaven” a fundamental part of their approach to governance.*> Developing this point further, the Mito scholars made it one of the core features of their interpretation of the efficacy of ritual. In so doing they wove to gether elements from a variety of Chinese classical sources.** One was Xiaojing Z #& (Classic of filial piety), which acclaimed the Duke of Zhou J& 4 for establishing the practice of “conducting rites to [the ruler’s] ancestors in conjunction with those to Heaven.” Linking the ancestral rites to Houji fi #8, the ancestor of the Zhou royal lineage, and King Wen 3 FE, the founder of the Zhou dynasty, to the rituals that the ruler offered to Heaven and Shangdi Lt #7, the Duke of Zhou
41 Aizawa Seishisai, Doku Naobi no mitama, p. 36.
42 Aizawa Seishisai, Shinron, p. 140. Shinron has been translated by B. T. Wakabayashi in Anti-Foreignism and Western Learning in Early-Modern Japan. See p. 252,
43 Ogyil Sorai, Taiheisaku, pp. 451-52.
44 For lucid analyses of how Mito scholars combined elements from a range of classical sources, see Imai Usaburd, “Mitogaku ni okeru jukyd no juyd”; and Sawai Keiichi, “Aizawa Seishisai no saisei itchi ron.”
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 283
had set forth the foundation for a supreme form of filial piety.> Be cause the Zhou rulers also used the mingtang Bf i, the hall in which they performed rites to their ancestors, to conduct affairs of govern ment, these acts of filial piety, the Mito scholars noted, were made manifest to the populace. The people were thereby awakened to the need to act on their own filial obligations, a crucial form of which was to continue the intent and deeds of one’s ancestors.*” Inspired by the power of the ruler’s ritual, the people would realize anew that the in tent of their ancestors had been to serve faithfully the ruler’s ancestors. To fulfill their filial obligations they thus should assist the ruler of their own day by carrying out the hereditary functions handed down from their forebears.
The Mito scholars used a number of phrases to describe this fe licitous state of unity and closure. One, descriptive of the impact on the populace of the ruler’s demonstration of filial piety, was “loyalty and filial piety rooted in one source” (chika wa ichi ni izu 8 Z% Hy —); a related phrase was “the multitudes being of a single mind” (okuché isshin {& Jk — .»). Another, evocative of an active commit ment to serve the ruler, was “recompensing the source and holding to the origin” (héhon hanshi # A Z #4). This last term, drawn from Liji, alluded to the spontaneous welling up of a desire to repay the ruler and one’s ancestors for the benefits received from them and to replicate in the present moment the order that had once existed. A fourth phrase, which pointed to the key means for achieving these ide als, was the “fusion of rites and government” (saisei itchi £3 B — 0). A closer look at what the Mito scholars envisioned this last to entail should offer further insight into the implications of their approach to Chinese ritual on the one hand and native tradition on the other.
45 Xiaojing: Shengzhi 52.15.
46 Ajzawa Seishisai, Shinron, p. 57.
47 Zhongyong + f# 19.
284 KATE WILDMAN NAKAI
The Fusion of Rites and Government
As is clear from their interpretation of the rites established by the Duke of Zhou, the Mito scholars held that to have the desired effect, the ruler’s conduct of rites had to engage the populace in some way; court ritual could not be simply the private affair of the ruler and his circle of officials. They found an archetypal model for the ideal of
broad engagement, together with a locus classicus for the phrase Ao hon hanshi, in Liji’s description of the Suburban Sacrifice (ké 3%, Chi nese jiac), the central rite offered by the ruler to Heaven. The solemnity of the Suburban Sacrifice, according to Liji, spurred the people to contribute spontaneously to preparations for the ritual, “ac cording with the wish of the ruler without being explicitly ordered to do s50.”*8 It was thus a prime occasion for the ruler to mobilize the populace and focus their energies in his service.
True to the “intrinsic match” between the Way of the sages and that of the founders of the Japanese state, the rulers of ancient Japan, Aizawa asserted, had also recognized the need to involve the populace in their rituals. As a key example he pointed to the passage in Nihon shoki that relates how the tenth emperor, Sujin 4 ##, in the face of a pestilence plaguing the land, moved the mirror granted by Amaterasu to her grandson Ninigi #8 H§ #f out of the emperor’s residence. To ac
count for Sujin’s move of the mirror, Nihon shoki stated simply that “he was fearful of the power of the deity and uneasy about keeping it in his living quarters.”*® Aizawa, however, alluding to “parallel” events and practices described in works such as Xiaojing, recast this episode into something comparable to the Zhou rulers’ heightening the effect of their ancestral rites by making their piety manifest to the peo
ple. Sujin’s transfer of the mirror, Aizawa declared, was motivated by a similar concern to involve the populace in his rites to Amaterasu, thereby both enlarging his expression of filial piety to her and bringing the people to realize the meaning of such reverence:
48 Liji 11: Jiaotesheng 8 5 4.
49 Nihon shoki, 1: 238.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 285
If he conducted [those rites] within his residence, he would not be able to express fully his own reverence for her; nor would he be able to make clear to the realm the import of such reverence. The emperor therefore offered his rites in public, thereby revering [Amaterasu] to
gether with the people of the realm. The meaning of reverence having been made clear to all, the people understood without being told. 3?
Aizawa extracted implications of much the same sort from the descrip tions of court ritual found in works such as the tenth-century Engishiki #E E R (Formulary of the Engi period). He gave particular attention to the Daijdsai XB £%, the Grand Thanksgiving ceremony held at the beginning of each reign in which the new emperor offered the first fruits to Amaterasu and entered into communion with her. Noting that there were “not a few” similarities in structure and import between the Suburban Sacrifice and the Daijosai,’! Aizawa looked to the latter to generate the same spiritual dynamic that Ljji attributed to the former. To enhance the parallel, out of the welter of specific ceremonial details
contained in Engishiki, he highlighted those comparable to Liji’s ac count of the activities associated with the Suburban Sacrifice, such as the sweeping of the roads by the populace. He used terms and phrases taken from the Chinese classics to describe the actions of the partici pants and the spiritual impact of the ritual. And he substantiated the implications he read into the native ritual by glossing his account with a string of references to Chinese sources where those implications could be found in a more readily recoverable form.5?
Through their rereading of the nonverbal, symbolic events of Japanese antiquity in light of the fuller Chinese record, the Mito schol ars constructed the foundation for a native structure of li. This struc
30 Aizawa Seishisai, Shinron, pp. 140-41; B. T. Wakabayashi, Anti Foreignism and Western Learning, p. 253.
31 Aizawa Seishisai, Kagaku jigen, p. 38a.
32 Aizawa Seishisai, Shinron, pp. 53-59, 147-51. In his otherwise fine translation of Shinron, B. T. Wakabayashi omits the chain of allusions to Chi nese sources contained in the long gloss, pp. 56-59. The omission is under standable, for the gloss could hardly be rendered into readable English. The importance of this passage to Aizawa’s overall argument, however, is evident from the fact that he repeated it verbatim in his discussion of the “intrinsic match” between the Confucian and native Japanese way in Kagaku jigen (pp. 37a-38a), written more than two decades later.286 KATE WILDMAN NAKAI
ture, they asserted, was specific to Japan, but it also was endowed with the power to foster the unity and closure that they identified as a prime feature of the li of the Chinese sages. Working from this interpretation of native tradition, they condemned the loss in later centuries of the public dimension that, they held, originally had been fundamental to the ruler’s ritual. Aizawa, for instance, bewailed the curtailment of nationwide participation in the Daijosai. Whereas in antiquity prov
inces had been chosen through an open-ended divination to provide the sacred rice for the rite, in later centuries, he pointed out, this honor had become the exclusive prerogative of a small number of regions. Aizawa likewise criticized the longstanding abandonment of the prac
tice of distributing imperial offerings to shrines throughout the nation on the occasion of the Toshigoi no matsuri #T ££ £3, the spring ritual to pray for good crops.
It is hardly coincidental that much of what Aizawa singled out as part of a lost native tradition of engaging the populace as a whole in the ruler’s ritual in fact dated to the late-seventh—early-eighth-century attempt to replicate the nationwide scope of the imperial Chinese state.” The legacy of that historical experiment contained in works such as Nihon shoki and Engishiki unquestionably facilitated his ef
forts to bring to light the “intrinsic match” between the li of the sages and those of the ancient Japanese rulers. Yet the Mito call for involv ing the populace in a national structure of ritual centered on the ruler clearly entailed more than simply the recovery of archaic forms. This is evident from the nature of the rites that the Mito scholars empha sized.
As one step toward realizing a new fusion of rites and govemn ment and of arousing in the people the dynamic of héhon hanshi, Aizawa drew up an annual ritual calendar.’ He based this work, in
53 Aizawa Seishisai, Kagaku jigen, pp. 35b-36a.
54 For this late-seventh—early-eighth-century attempt, see Okada Seishi, Kodai ken no saishi to shinwa, pp. 41, 146-52, 165-68; Hayakawa Shoha chi, “Ritsurydsei to tennd,” pp. 78-82.
55 The title of the work in which he set out and explicated this calendar, Séen kagen # {BE Fn Z, clearly indicates its purpose. Séen (“the grass bends™) comes from the famous description of the people bending before the virtue of the gentleman, like grass before the wind, in Analects 12:19, while kagen is
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 287
tended for popular edification, to a considerable extent on a fifteenth century summation of the round of court ritual that drew in turn from earlier compilations such as Engishiki.>® In his use of these earlier works, however, Aizawa again was selective. He omitted ceremonies that in his view were primarily decorative, superstitious, or esoteric (or he downplayed their details if, as in the case of rituals included in Engishiki, their place in tradition was too secure to allow their total omission). The rituals on which, by contrast, he lavished attention were ones such as the Daijdsai or Toshigoi no matsuri. While, he ac knowledged, one should take care not to be disrespectful in listing up the details of court ceremonies, these were precisely the things that “the populace should look up to with reverence.”s” He also hoped to make such reverence more than just a matter of intellectual knowl edge. Ideally, he held, the court should revive the practice of dispatch ing offerings to shrines throughout the land on the occasion of major rituals such as the spring Toshigoi no matsuri and the annual autumn thanksgiving festival.>® As something more immediately feasible, he called for local shrines to coordinate their ritual activities with those of the court. Thereby the people taking part in those rites would be made aware that the blessings they received were owing to the benevolence of Amaterasu and the line of rulers descended from her and would be inspired to “recompense the source and hold to the origin.”*?
Taking the call for alignment to a yet more fundamental level, Aizawa recommended that individual families should conduct their own ancestral rites in conjunction with those performed at court. Re
based on a passage in Shujing (Classic of documents) referring to the need to speak appropriately to the populace. Cf. Shujing iv:7:i:12 (Pangeng shang #&% BL).
36 The work Aizawa used was Kuji kongen 7%: % #8 Ii (Sources of court ceremonial), a summation of the annual round of court ritual compiled by the fifteenth-century court noble and scholar Ichijo Kanera — 4 3% & (1402-81). 57 Aizawa Seishisai, Sden kagen, p. 309.
38 While in SGen kagen Aizawa advocated implicitly rather than directly revival of the sending of offerings from the court, in Kagaku jigen he put the point much more forthrightly. See pp. 35b—36a.
39 Aizawa reiterated this point for each of the central rituals of the year. See Soen kagen, pp. 283, 291, 299, 303-4.
288 KATE WILDMAN NAKAI
gretfully, he noted, the court had abandoned the custom of sending offerings to imperial tombs in the last month of the year. Nevertheless, as with the other major annual ritual events, it would be possible to reproduce some of the effect of ancient rites if the people were to match their own activities with those occurring within the confines of the court. Just as the emperor faithfully fulfilled filial obligation through sacrifices to the imperial ancestors, at the end of the year the populace should visit the graves of their ancestors and perform appro
priate rites to them.
Aizawa looked to such activities to evoke in the populace more than just an awareness of the presence of the court. For people to feel a true commitment to reenact in the present the devoted service offered the ruler in antiquity, they also had to have a sense of continuity with the past. The common people of his day, he noted, lacked such a sense. They knew that “they have a father, but not that they should revere their ancestors.” This was one reason why they were susceptible to heterodox promises of a false peace of mind. To rectify this Aizawa did not attempt to promote orthodox Confucian family ritual with its premise of the generational dispersal of spirits. Concerned to foster an instinctive identity with a social chain of existence stretching into the past, he instead sought to impress upon the people that ancestral spirits remained permanently close to their descendants with whom, “no mat
ter how many generations have passed,” they shared “one identical body.”! Through year-end rites at the graves of their ancestors the people would be able to “preserve eternally” the spirits of their fore bears. In this way they, too, like the emperor, would be true to the command of Amaterasu to uphold loyalty and filial piety unto eter nity. 62
60 Aizawa Seishisai, Saen kagen, pp. 315-16.
61 Aizawa Seishisai, Tekiihen, pp. 270-71. While his emphasis on the im portance of a sense of familial continuity stretching into the distant past led Aizawa to downplay the notion of the generational dispersion of spirits, it also encouraged him to affirm the Chinese Confucian prohibition of nonagnatic adoption. Nonagnatic adoption, he argued, would interrupt the “common vital essence” (ikki — &) linking ancestor and descendant, and thus impede their ritual interaction.
62 Aizawa Seishisai, Sden kagen. pp. 315-16.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 289
This emphasis on the coordination of popular practice with court ritual testifies to a new level of awareness of the necessity for rulers to heed popular emotions and guide them in appropriate directions. It also reflects the Mito obsession with insuring the integration of dis
tinct, and thus potentially divergent, elements (such as loyalty and fil ial piety). These concerns endowed the Mito approach to Confucian ism with another characteristic feature. Even as they reread and redefined native tradition in light of ritual prototypes depicted in the Chinese classics, the Mito scholars significantly altered the thrust of those prototypes. Aizawa’s approach to ancestral rites illustrates this as does his interpretation of the Daijosai.
As noted above, Aizawa found particular import in the passage from Xiaojing about the Duke of Zhou establishing the practice of “conducting rites to [the ruler’s] ancestors in conjunction with those to Heaven.” The Daijosai, he held, took this logic a step further. Amaterasu, as the sun, was not only the emperor’s ancestor, but also equivalent to Heaven. In the Daijosai, consequently, rather than simply conducting ancestral rites “in conjunction” with those to Heaven, the emperor was able to integrate the two.
To emphasize the implications of this situation, Aizawa, while af firming that there were “not a few” similarities between the Daijosai and the Suburban Sacrifice, also took care to preserve a certain differ ence. The first “historically” clear performance of the Daijosai, he held, was the rites to the “ancestral heavenly deities” (mioya no amatsukami 2 tl X 4) performed by the first emperor, Jinmu ## i, at Tomiyama £5 R (Ll following his accession. Nihon shoki referred to these rites as ké 5%, the term used in China for the Suburban Sacrifice. This circumstance undoubtedly facilitated Aizawa’s conclusion that the rites enacted by Jinmu were analogous to the Suburban Sacrifice. He also asserted, however, that the Tomiyama rites were distinct from the ceremony explicitly termed Suburban Sacrifice that the early Heian emperors had incorporated into the ritual of their court. This latter ceremony, he acknowledged, “was modeled completely upon the Chi nese Suburban Sacrifice . . . and practiced for only a short period.” In comparison to this later, direct imitation of Chinese ritual, the special merit of the original native rites conducted at Tomiyama and preserved in the Daijosai was their concision and comprehensiveness. The join ing of recompense to Heaven and ancestors in a single ritual enhanced
290 KATE WILDMAN NAKAI
its efficacy, insuring that all brought within the scope of its transform ing influence would be of one mind and safe from temptation by “alien elements.”53
The collapse into one of the elements that in the original Chinese context remained separate, if related, unquestionably supported the Mito pursuit of a spiritual closure that would unite society from top to bottom. It indicates as well some of the consequences of the changes in the ideas about the function of li wrought by Sorai and Shundai. By building on their notion of li as state ritual and an instrument of gov
ernance, the Mito thinkers succeeded in formulating a powerful vision of a nation united in the simultaneous performance of rituals centered on the ruler. Their intricate cross-referencing of Japanese “event” and Chinese text likewise enabled them to incorporate a range of Confu
cian elements into an image of “original” Japanese tradition. But in the process those elements acquired a new valence. Instead of locating people in a natural order that, extending out through a sequence of familial and social structures, ultimately encompassed the universe, the li of the Mito scholars situated them firmly in a bounded political and social framework. The ultimate consequences of this development became evident only some time later.
Aizawa’s proposals for ensuring that loyalty and filial piety were “rooted in one source” and inspiring in the populace a commitment to “recompensing the source and holding to the origin” were not enacted in his lifetime. They did, however, attract the attention of the officials who formulated the ritual measures implemented by the Meiji state. In
63 Aizawa Seishisai, Kagaku jigen, pp. 30a-b. Aizawa may be seen as building here on an equation already implanted by the compilers of Nihon shoki. That work describes Jinmu as declaring that he will “perform sacrifice to the Heavenly Deities (5 jE X #1), and therewith extend filial duty to the full.” Nikongi, 1: 134 (translation slightly modified); Nihon shoki, 1: 214. Kojiki, by contrast, does not mention the rites at Tomiyama. As with the Chi
nese-influenced aspects of the Daijosai and Toshigoi no matsuri mentioned above, this passage is another instance of the eighth-century incorporation of Chinese elements that the Mito scholars were able to put to good use in their rereading of native tradition. The reference to being safe from temptation by “alien elements” (7~3& i> 4), another phrase that occurs frequently in Mito writings, comes from Guanzi %& F: Xiaokuang /| [E.
Chinese Ritual and Japanese Identity in Tokugawa Confucianism 291
the first year of Meiji, the government made a point of announcing to the populace as a whole the performance at the court of the annual autumn thanksgiving festival. The following year there was a call for the revival of the Toshigoi no matsuri together with the practice of
sending offerings to shrines throughout the nation. The cycle of na tional ceremonies eventually established by the state paralleled in pur pose, and substantially in content as well, Aizawa’s ritual calendar.%
In this and other ways the Mito articulation of the notion of a fusion of rites and government helped shape what eventually emerged as state Shinto. Here we may see one major dimension of the impact of Confu cian ritual on Tokugawa society and of the part it played in the con struction of “native” identity.
6 For early Meiji ritual innovations, see Sakamoto Koremaru, Kokka shinto keisei katei no kenkyii, pp. 48-50, 56-58; Takeda Hideaki, “Shiji saiten teisoku seiritsu katei.”
15 “Esoteric” and “public” in late Mito thought
Kate Wildman Nakai
As the other contributions to this book have detailed, the esotericism prevalent in medieval Japanese Buddhism had a far-reaching impact on various dimensions of religious, cultural, and intellectual life. Central among its legacies was the articulation of “Shinto,” claimed by successive thinkers to be an autonomous body of doctrines and practices, but in fact owing much of its content and orientation to the matrix of esoteric Buddhism from which it emerged. What happened to this combination of elements with the rise of interest in Confucianism and the concomitant disavowal of Buddhism characteristic of early modern intellectual life? Confucian thinkers of all schools put priority on the ordering of society and the establishment of proper social relations; they also took as a given that, as categories, “public” (k2 公) was something positive, in contrast to “private” (shi, watakushi 私), which carried dubious and negative connotations. One might thus expect a commitment to Confucianism to bring with it a querying of esotericism and its attendant assumption that privileged knowledge should be guarded and transmitted only to the initiated.
In many regards this proved to be the case. A story told of Hayashi Razan 林羅山 (1583–1657) is emblematic. According to the biography compiled by his sons, Razan’s career as a scholar began with a challenge to the medieval norm of privileged transmissions. Having become familiar with Zhu Xi’s commentaries on the Four Books, the young Razan, so the story goes, began to offer public lectures in Kyoto on the Analects. The court scholar Kiyohara (Funahashi) Hidekata 清原(船橋)秀賢 (1575–1614), whose house specialized in the reading and recitation of the Confucian classics, protested that only those with an imperial sanction to do so should be allowed to conduct lectures on the classics. The newly appointed Shogun Tokugawa Ieyasu, however, the biography relates, intervened on Razan’s behalf. Whether or not these events occurred precisely as described, the story is indicative of a new value assigned to the “public” over the “secret” or “private.”1
Yet the situation was not so simple and straightforward as the story might be taken to suggest. Just as earlier generations of religious figures had drawn equations between elements of kami worship and Buddhism,
358 Kate Wildman Nakai
many Tokugawa Confucians, Razan included, posited a unity between Confucianism and Shinto. We may assume various factors to underlie this phenomenon. There was doubtless a concern, on the one hand, to infuse Confucian practice with immediacy by linking it to an existing spiritual regimen and ritual forms. The fact that various late medieval and early modern proponents of Shinto were in the process of divesting their formulations in name, if not fully in content, from an affiliation with Buddhism facilitated reconnection with Confucianism. Confucians participating in these developments, on the other hand, also wished to reform and reorient Shinto practices by joining them more explicitly to the Confucian goal of ordering society.
Regardless of the underlying motivation, in their efforts to combine the doctrines and practices of the two traditions, early Tokugawa Confucians often fell under the sway of medieval Shinto’s propensity for casting knowledge as a secret to be conveyed only to the properly initiated. They also inherited specific points of esoteric lore regarding the nature of certain kami and ritual acts and the meaning of passages from the accounts of kamiyo 神代 (the age of the gods) in Nihon shoki and other texts. Razan’s Shint2 denju 神道伝授 (Transmissions on Shinto), written in the 1640s, provides a pertinent example. To be sure, in this work Razan took steps to disassociate his version of Shinto from currently prevalent forms. The “true” form of Shinto that he propounded was, he asserted, identical with the “way of the king” (2d2 王道), in other words, the Confucian Way. Drawing from Song Confucian ideas, he further identified Shinto cum 2d2 with an inner spiritual realm linking the individual to a moral cosmic order, and he demarcated this “true” Shinto from lesser forms associated with shrine priests, whom he characterized as dealing in shinji 神事, technical matters of ritual and such, not Shinto.2 But, as Bernhard Scheid has recently pointed out, Razan also composed Shint2 denju in the form of a collection of separate secret transmissions, and he evidently conveyed, or planned to convey, individual sections of it in the kirikami 切紙 (“separate sheet”) format traditionally used for such transmissions.3 Many of the points to be transmitted show the direct influence of Yoshida Shinto ideas, and Razan’s characterization of the metaphysical inner world of the spirit built on medieval interpretations of the initial generations of kami such as Kuni no Tokotachi 国常立 and Ame no Minakanushi 天御中主, who had been the subject of much allegorical exegesis. And, in formulating what were essentially initiations into the meaning of such matters, Razan followed the patterns for conveying esoteric knowledge by identifying one item or another as “a secret among secrets,” or by stating that “this is an inner secret of Shinto; it should not be disclosed to others” (Shint2 denju, pp. 36, 46).
The somewhat younger Confucian Yamazaki Ansai 山崎闇斎 (1618–82), known for his fervent commitment to Zhu Xi and for his emphasis on a rigorous, introspective self-cultivation, showed an equal debt to the
“Esoteric” and “public” in late Mito thought 359
medieval Shinto fondness for esoteric hermeneutics. His efforts to demonstrate the identity between the kamiyo chapters of Nihon shoki and Zhu Xi metaphysics, for example, depended heavily on the allegorical word play and numerical associations favored by medieval commentators. Like Razan he took Shinto to be something of universal relevance and held that it should not be treated as the privileged possession of one particular lineage. But he also continued to couch central aspects of Shinto doctrine as esoteric knowledge to be conveyed through stages of initiation.4 Consolidated further by his followers, this aspect of Ansai’s approach to Shinto became an established feature of the Suika 垂加 school of Shinto devolving from him.
But not all Tokugawa efforts to establish an identity between Confu cianism and Shinto stayed within the framework of the allegorical, inward-looking perspective with its deep debt to medieval esotericism as well as Song metaphysics that we see in Razan and Ansai. The thorough going critique of Song thought mounted at the end of the seventeenth century and in the first decades of the eighteenth century by the so-called Ancient Learning (kogaku 古学) scholars, most particularly Ogy5 Sorai 荻生徂徠 (1666–1728), inevitably had an impact on such constructions as well. Sorai attacked the Song emphasis on introspective self-cultivation supported by a metaphysical apparatus that reified the “abstract and minute” as a “specious product of private speculation” (shichi m2saku 私智 妄作).5 The same criticism could be readily extended to the medieval and early modern forms of Shinto, described caustically by Sorai’s senior disciple Dazai Shundai 太宰春台 (1680–1747) as no more than something concocted out of “seven or eight parts Buddhism and two or three parts” Song metaphysics (Bend2 sho, pp. 44–46).
Among the heirs to this perspective were the thinkers of the late Mito school, active from the end of the eighteenth to the mid-nineteenth century. The late Mito thinkers – Fujita Y5koku 藤田幽谷 (1774–1826), Aizawa Seishisai 会沢正志斎 (1782–1863), and Y5koku’s son T3ko 東湖 (1806–55) – disagreed strongly with some of Sorai’s propositions. Most notably, they took issue with his assertion that the Way was a human construction. To the contrary, they insisted, the Way was grounded in the natural order of things, and, in line with this position, they reiterated again and again the obligation to uphold the moral norms – the five constants of human relationships – that constituted the foundation of the Way. While they adhered to the traditional Confucian position that the Way was something natural and innate, however, the Mito thinkers had more in common with Sorai than with followers of the Song tradition. They shared Sorai’s suspicion of the Zhu Xi emphasis on metaphysical analysis and reasoning and on introspection and cultivation of the self. Like Sorai they fully recognized the power of the religious impulse in human beings, and they were sensitive to the dynamics of the quest for spiritual reassurance and consolation in the face of death and the360 Kate Wildman Nakai
unknown. But they also saw these impulses as something potentially dangerous. To be directed to positive ends, the instinctive yearning for spiritual reassurance needed to be controlled and properly guided through effective structures of ritual.6
Lending urgency to the Mito thinkers’ focus on these issues was their intense concern about the fractures and divisions that they saw under mining the social order of their time. These, they were convinced, might well provide an opening to foreign incursions, the way for which would be paved by insidious efforts to spread subversive forms of belief and religion. To prevent this from happening, it was essential to do away with all potential sources of divided allegiance and to develop means of fostering a spiritual unity in which “the multitudes would be of one mind” (okuch2 isshin 億兆一心).7
These convictions permeated the Mito approach to Shinto. Like Razan and Ansai, the late Mito thinkers posited an “intrinsic match” (ang2 暗合) between Shinto and the Confucian Way. The essence of both, they held, lay in the clarification and reinforcement of the moral fabric of the social and political order. Beyond that, both Shinto and the Confucian Way were founded on recognition that rituals to the spiritual forces of Heaven and Earth and ancestral rites were the most efficacious means to unify the minds of the populace and convey moral norms.8 As evidence of this intrinsic match, the Mito scholars pointed to the passage from the Yijing 易経 stating that “the sage uses the divine way (shendao; i.e., shint2) to give instruction, and the entire world submits to him.”9 This, they argued, was exactly the method of rule original to Japan. Citing the phrase from the Yijing, Aizawa wrote of Amaterasu that “in antiquity, the Heavenly Ancestor (Tenso 天祖), using the divine way to give instruc tion, established the norms of human behavior by clarifying loyalty and filial piety.”10
Shinto, according to the Mito scholars, was thus by origin an instrument for conveying moral principles and establishing order throughout society. In “using Shinto to give instruction,” Amaterasu had set forth standards essential to the eternal preservation of the imperial line and to ensuring the welfare of the realm over which it ruled. The formulators of the various schools of so-called Shinto that had appeared in Japan in the wake of the introduction of Buddhism had lost sight of this purpose and had distorted Shinto’s true nature. The cosmological and allegorical interpretations of the kamiyo texts and other works that they had elaborated were nothing more than their own “private speculations.” By further framing these private speculations as stages of initiation, the medieval and early modern Shinto thinkers had, as Fujita T3ko put it, “turned the record of the accomplishments of the sagely and divine founders of the realm into something like a riddle or a secret code.”11 In these forms, Shinto could hardly provide appropriate moral guidance or serve to unify the minds of the multitudes. To meet the needs of time, it was
“Esoteric” and “public” in late Mito thought 361
crucial to rescue Shinto from its current state of corruption and restore its original character as a means of extending moral instruction throughout society. The Mito scholars pursued two intersecting routes to this end. One was to reclaim territory over which esoteric notions had extended a far-reaching influence; specifically this meant freeing deities such as Amaterasu of the accretion of “distorting” interpretations and resituating them within the framework of Shinto as the Mito scholars conceived it. The other was to set forth a structure of ritual appropriate to “using Shinto to give instruction.”
Reclaiming esoteric territory
In place of the expansive allegorical interpretations of the kamiyo deities characteristic of medieval and early modern Shinto, the late Mito thinkers took a fundamentally historical approach.12 Although a sacred entity identical with the sun, Amaterasu was also the founder of the imperial line and the polity of Japan. As such, in their view, she was analogous in many regards to the sages of ancient China.13 Other key deities, like Ame no Koyane 天児屋根, the ancestor of the Nakatomi 中臣 lineage and its offshoot, the Fujiwara, the late Mito thinkers likewise saw as historical figures who had served Amaterasu and assisted her in establishing the foundations of the polity. To this premise the late Mito thinkers attached a sequence of interlocking propositions. The founders of the polity had provided later generations with immense benefits for which their descendants owed eternal recompense. Such recompense meant not only honoring the founders through appropriate rituals, but also continuing the tasks they had undertaken. The same charge applied to the general populace as well as the rulers of the day. Concretely, in antiquity, the ancestors of those now living had carried out diverse roles in service to Amaterasu; in the same way, the people of the present should serve the emperor, her descendant, by re-enacting the functions performed by their ancestors. They would thereby fulfill two obligations at once: loyalty to the ruler and filial piety to their ancestors. The two central moral norms, loyalty and filial piety, would be “rooted in one source” (ch4k2 wa ichi ni izu 忠孝出一), while “the present would replicate the antiquity” (ky2 wa sunawachi j2ko; j2ko wa sunawachi ky2 今日即上古、上古即今日).14 Such fusion would ensure the unification of the minds of the populace, guard against any cracks in the social fabric providing an opening to subversive forces, and secure the preservation of order and hierarchy. The late Mito thinkers summed up this sought-for dynamic with a phrase from the Chinese classics: h2hon hanshi 報本反始 (“recompensing the source and holding to the origin”).
To inscribe (or, as they would see it, to bring to light) the mechanism of h2hon hanshi at the core of Japanese tradition, the late Mito thinkers concen trated attention on the “historical” deities – above all Amaterasu – whom
362 Kate Wildman Nakai
they held to be central to its operation. Simultaneously they downplayed the significance of those kamiyo deities that did not fit readily within the framework of h2hon hanshi as they envisioned it. Among these, hardly coincidentally, were shadowy entities such as Kuni no Tokotachi and Ame no Minakanushi that were difficult to link to activities of a “historical” nature and that throughout the medieval and early modern period had been favored subjects of esoteric allegorical interpretations.
Aizawa’s Tekiihen 迪彜編 (Exhortations about Adhering to Moral Norms), a work he wrote in 1833 for popular edification, provides a revealing example of this strategy. In this piece Aizawa did not cite directly from the original kamiyo texts such as Nihon shoki. Instead, to explain the distinctive features of the Japanese polity, he stated that he would quote from the “correct account” of the origins of the imperial line that Kitaba take Chikafusa 北畠親房 (1293–1354) had set out in his Jinn2 sh2t2ki 神皇 正統記 (Chronicle of the Legitimate Line of Deities and Sovereigns). Presumably Aizawa chose this method because Chikafusa’s “correct account” brought out much more fully than the original texts the image of Amaterasu as a source of moral norms that the late Mito scholars wished to emphasize. In describing Amaterasu’s grant to her grandson Ninigi 瓊瓊杵 of the regalia, for instance, Chikafusa had explained that she had not bestowed upon Ninigi simply tangible symbols of authority. More importantly she had provided him with teachings as to “the correct way to govern and preserve the country.”15
But while quoting Chikafusa directly on such points, Aizawa also excised or modified elements that he found inconvenient. Among other things, he omitted almost all of Chikafusa’s version of the opening section of the kamiyo accounts. Over the centuries this section, which described the state of primal chaos prior to the separation of Heaven and Earth and named the initial generations of deities, had been a particularly rich source for allegorical and speculative reinterpretation. Of Chikafusa’s version, Aizawa kept the statement that the first deity at the beginning of Heaven and Earth was known as Kuni no Tokotachi and alternatively as Ame no Minakanushi. He added a cautionary note, however, observing that “although there are various theories about this name, it being a matter of the distant past, the details of the situation are not clear” (Tekiihen, pp. 251, 253–55). And through a discreet sleight-of-hand, he shifted attention away from the role that esoteric interpretations had long assigned Kuni no Tokotachi as an originary force.
Referring to Kuni no Tokotachi as Tenso (which we may translate here as Heavenly Progenitor), Chikafusa began Jinn2 sh2t2ki by declaring that Japan was known as the country of the gods because “the Heavenly Progenitor set forth its foundations and the Sun Deity bequeathed its rule eternally to her descendants.”16 For Chikafusa, the involvement of Kuni no Tokotachi as an originary force was thus a crucial element in the establishment of the Japanese polity.17 Aizawa, on the other hand, as we
“Esoteric” and “public” in late Mito thought 363
saw above, commonly used Tenso in the manner of the laudatory names of founders of Chinese dynasties to refer to Amaterasu. Elsewhere he specified that this term traditionally had been used in contrast to the more generic tenjin 天神 (heavenly deities) to distinguish Amaterasu as a sole actor.18 In quoting the opening passage from Jinn2 sh2t2ki without any reference to Chikafusa’s identification of Tenso as Kuni no Tokotachi, Aizawa elided the role ascribed to the latter deity together with its various esoteric connotations. Instead he combined in Amaterasu the function of both founder and bequeather. Similarly, while reproducing in extenso the passage from Jinn2 sh2t2ki concerning the regalia, through the omission of certain phrases and the modification of others, Aizawa reshaped Chikafusa’s depiction of this event. Although he retained, for instance, the emphasis that Amaterasu conveyed teachings to Ninigi together with the regalia, Aizawa systematically suppressed the allegor ical and inward-looking interpretations of those teachings that Chikafusa had derived from Watarai Shinto texts. Symbolic of these changes is Aizawa’s substitution of the matter-of-fact phrase “[the regalia] are the divine treasure of this country” (kono kuni no shinp2 nite この国の神宝にて) for Chikafusa’s more mystical statement that Amaterasu conveyed them “as the divine spirit of this country” (kono kuni no shinrei toshite 此国ノ神霊 トシテ). Similarly Aizawa excised the phrases in which Chikafusa charac terized the regalia as the “font” (hongen 本源) of uprightness, compassion, and wisdom, the qualities that the emperor needed to cultivate so as to govern the realm properly.19 In place of Chikafusa’s emphasis on the process of self-cultivation incumbent upon the ruler, Aizawa stressed that Amaterasu’s bequest of the regalia and command that her descendants should rule Japan eternally constituted a central pillar of the imperative to “recompense the source and hold to the origin.” For the emperor, fulfillment of this imperative meant above all to uphold his filial duty to Amaterasu through preservation of the unbroken imperial line and ongoing devotion to her.
To reinforce Amaterasu’s centrality within the dynamic of h2hon hanshi, Aizawa did not merely reorient Chikafusa’s version of the events of kamiyo to bring out more sharply that the essence of her teachings resided in the “clarification of loyalty and filial piety.” Moving outside the parameters of Chikafusa’s account, he drew attention to the significance of other blessings received from Amaterasu. Apart from granting Ninigi the regalia, she had, “with her own hands,” bestowed upon him sheaves of grain from her sacred field and had shown the way to cultivate silkworms as well as the use of other fibers. Through these beneficent gifts she had provided the means for the populace to sustain life and protect themselves from the cold.20 Not only the ruler responsible for the welfare of the populace, but the people themselves had thus incurred an inexhaustible debt to Amaterasu as the source of the most fundamental necessities of human existence as well as moral teachings.
364 Kate Wildman Nakai
The late Mito thinkers’ approach to deities like Okuninushi 大国主 illustrates another aspect of the challenge to esoteric understandings embedded in the resituation of the deities of kamiyo as “historical” actors. Described in Nihon shoki as having declared his readiness to yield authority over the land to the descendants of Amaterasu and to devote himself to “divine matters” or “hidden matters,”21 Okuninushi could be assigned a place within the historical process of the founding of the polity. But through imaginative exegesis at the hands of medieval and early modern commentators he had also acquired a variety of other characteristics and associations. As Bernhard Scheid discusses in this book, Shinto thinkers took the passages about “divine matters” and “hidden matters” as a key sanction for esoteric interpretations and practices. In the early Tokugawa period Yamazaki Ansai had interpreted a dialogue between Okuninushi and his “wondrous spirit” (sakimitama kushimitama 幸魂奇魂) as a model for a personal spiritual regimen of introspective moral reflection.22 Nearer in time to the late Mito thinkers, Hirata Atsutane 平田篤胤 (1776–1843) had acclaimed Okuninushi as the ruler of the unseen world of the spirits of the dead.23
For the late Mito thinkers, these interpretations of Okuninushi were prime examples of problematic “private speculation” that, by fostering heterodox beliefs and practices, served only to divide, not unify, the allegiance of the populace. Deflating the aura of expansive associations surrounding Okuninushi, the late Mito thinkers instead presented him simply as the chief of the Izumo region. As such he was one among the “many” early chiefs of this sort who had pacified local territories and done things beneficial for the populace. That the people of the regions where such chiefs had been active should revere them was only natural.24 Okuninushi was further worthy of emulation as an exemplar of a local chief who, having presented his territory to the sovereign, had dispatched emissaries to the court to pay homage. And he deserved praise for his achievements in developing methods of healing.25 But the reverence offered to him should be incorporated within an appropriate overall framework of ritual that would contribute to realization of the unifying dynamic of h2hon hanshi.
The efficacy of public ritual
The locus classicus of the term h2hon hanshi was Liji 礼記 (Book of Rites), and, as this suggests, the major inspiration for the late Mito thinkers’ concept of a proper system of ritual likewise came from the classical Chinese canon, particularly works such as Liji, Zhouli 周礼 (Rites of Zhou), and Shangshu 尚書 (Book of History). In these works the Mito scholars saw a picture of society governed through ritual structures that also conveyed moral norms and that thereby guided the populace in the proper direction while simultaneously providing an outlet for its
“Esoteric” and “public” in late Mito thought 365
religious impulses. Put slightly differently, they found delineated in concrete form further core components of their notion of Shinto, such as the fusion of governance and doctrine (jiky2 itchi 治教一致) and ritual and governance (saisei itchi 祭政一致). More than simply channeling the religious instincts of the populace into an appropriate framework, this model of government through ritual had the capacity, the Mito thinkers were convinced, to tap into the energies of the people and arouse in them the spontaneous commitment to serve the ruler summed up in the term h2hon hanshi. Prime examples of this ideal were the ruler’s annual sacrifices to Earth and Heaven – the she 社 (Jap. sha) and jiao 郊 (Jap. k2; i.e., the Suburban Sacrifice). As described in Liji, the populace did not take a direct part in the rituals. Yet, the solemnity of the occasion aroused in the people an instinctive desire to contribute to their successful perform ance. For the sacrifices to Earth, members of all levels of society went out to secure the items to be offered. For the sacrifices to Heaven, the people undertook to water and sweep the road along which the ruler would pass, “according with the wish of the ruler without being explicitly ordered to do so.” The transforming effect of the ritual inspired in them a spontaneous desire to “recompense the source and hold to the origin.”26
The late Mito thinkers’ conviction that an “intrinsic match” existed between the Way of the founders of the Japanese polity and the Way of the sages carried with it the corollary that something like the types of rituals described in Liji must have existed in Japan. Initially, they admitted, the forms had been quite different. In antiquity, when customs had been simple and the people straightforward, the Japanese sage rulers had not found it necessary to explicate the Way through formal structures; they had merely conveyed it symbolically, as had Amaterasu in granting Ninigi the regalia and sheaves of grain. It was the natural course of things, however, for society to grow more complex over time. As a complement to simplicity (shitsu 質), cultured forms of expression (bun 文), including explicit ritual structures, became correspondingly essential. Recognizing this, the early emperors, such as the first emperor, Jinmu 神武, and the tenth emperor, Sujin 崇神, had established appropriate rituals. But the incursion of Buddhism had corrupted native practices and blinded people’s perception. People no longer could grasp directly the import of what the founders had intended. To clarify the founders’ intent it was thus necessary to turn to Chinese sources where the same fundamental methods and aims were set forth in more explicit written form.27
Armed with this proposition, the late Mito scholars identified in works such as Nihon shoki several key instances of the early emperors instituting forms of ritual comparable to those described in the Confucian classics. To substantiate these cases and to bring out their full implications, the Mito thinkers adopted the same interpretive strategies that we have seen them employ to establish that Amaterasu provided instruction through Shinto and to reshape Chikafusa’s version of kamiyo to fit their own aims.
366 Kate Wildman Nakai
In place of the word play and numerology favored by the practitioners of medieval and early modern esoteric hermeneutics, the Mito scholars built up an argument based on analogy. Carefully selecting the relevant details of the Japanese event, they drew out its meaning (or, we might say, more accurately, infused it with the desired meaning) by enveloping those details in a web of references to the fuller Chinese accounts of the ancient Chinese “parallel.” While we might see this process as entailing veiling and concealment, from their perspective it enabled the recupera tion of elements of Japanese tradition that hitherto had been susceptible to esoteric distortion.
We can find an instructive example of this process in Aizawa’s inter pretation of the Nihon shoki passage that relates how Sujin moved the mirror granted by Amaterasu to Ninigi out of the emperor’s residence. Prior to this, Nihon shoki states, the emperor had worshiped Amaterasu and the deity Yamato no Okunitama 倭大国魂 within his own living quarters. But following a pestilence that swept the land and popular rebellions, in response to which he begged the deities of Heaven and Earth for forgiveness, he became “fearful of the power of these deities and uneasy about living in proximity to them. He thus entrusted Amaterasu [i.e., the mirror] to Toyosukiiribime no Mikoto 豊鍬入姫命 and worshiped her at Kasanui 笠縫 in Yamato” and made arrangements as well for the separate worship of Yamato no Okunitama.28 In Heian and medieval texts the transfer of the mirror to Kasanui assumed importance as the first stage in its eventual conveyance to Ise, a development that received much attention and elaboration at the hands of Watarai Shinto thinkers.
Aizawa, however, alluding to “parallel” events and practices described in Chinese classical texts, recast this episode into something quite different from both the original and the medieval versions of it. The Zhou rulers, he pointed out, had performed ancestral rites to the founder of the dynasty, King Wen 文王, in the hall where they conducted the rituals and affairs of government. By thus associating the populace of the realm with their rites to King Wen, the later Zhou rulers had manifested more fully their reverence for him. Sujin’s transfer of the mirror, declared Aizawa, was motivated by similar concerns. Not only did the involvement of the populace in his rites to Amaterasu serve to enlarge Sujin’s expres sion of filial piety to her, it also had the edifying effect of bringing the people to realize the meaning of his piety. “Conducting rites within his residence,” Aizawa wrote, Sujin “might have been able to express inwardly his own sincere reverence, but he could not make clear to the realm the import of such reverence. The emperor therefore conducted rites outside his residence, revering [Amaterasu] publicly together with the people of the realm. The meaning of such reverence having been manifested to the realm, the people understood without being told.”29
Aizawa used similar interpretive procedures to extract from the accounts of the early emperors evidence of their conducting rites
“Esoteric” and “public” in late Mito thought 367
comparable to the she and jiao recorded in Liji. As the ritual analogous to the jiao, the suburban sacrifices to Heaven, he singled out the daij2sai 大嘗祭, the ceremony of accession in which the new emperor engaged in ritual communion with Amaterasu.30 Its transforming effect made the daij2sai, he held, the ultimate expression of h2hon hanshi. Using the language of Liji, Aizawa wrote evocatively, on the one hand, of the communion that occurred when the emperor, the “bequeathed body” (itai 遺体) of Amaterasu, offered the first fruits of the harvest to her, a devoted act of return for the grain she had granted Ninigi. It was as if “full of reverence, he could faintly see her before him, present today in her original form.” But Aizawa equally stressed the impact of the occasion on others. The assembled officials felt as if they, too, “were in the presence of the Heavenly Ancestor, and looking upon the emperor, it was as if they saw the Heavenly Ancestor.” And, as the descendants of those deities who had served Amaterasu and the early emperors, they contributed to the realization of h2hon hanshi by re-enacting the roles performed by their ancestors.31
Further, as with the jiao, the transforming effect was not limited to those immediately present. Since the “emperor takes the realm as his house and shares all his actions with the realm,” originally he had involved the entire populace in the daij2sai through various preparatory events, such as the selection of the provinces to provide grain for the ceremonies. Lamenting that in later centuries this process had become formalistic and perfunctory, with the provinces designated to perform this function permanently fixed, Aizawa pointed out that in antiquity they had been chosen through a true divination, and thus all the realm had remained involved. “There was no province that might not be charged with providing the offering.” Within the province selected, the people all gathered together to prepare the grain and send it off properly guarded. As other provinces offered tax revenues to pay for miscellaneous costs, “there was no province that did not serve the deity”; through the ceremonies of purifying the roads along which the offerings were sent, “the entire realm knew what was to be revered”; and through the emperor’s distribution of celebratory offerings to the shrines of the land, “the entire realm knew that there was no deity not under the authority of the Heavenly Ancestor and her imperial descendants.” In this way the people were brought to “manifest the utmost in devotion and reverence without any need for insistent explanations. [. . .] The emperor’s devotion and reverence extended from the palace and was felt throughout the realm, and the devotion and reverence of the realm was gathered from the multitudes and focused around the seat of rule.”32
In his presentation of Sujin’s transfer of the mirror and the daij2sai, Aizawa emphasized the beneficial impact of making “public” ceremonies and matters usually understood as things to be carefully veiled from view. To be sure, in noting the late Mito scholars’ affirmation of “public”
368 Kate Wildman Nakai
ritual in place of “esoteric,” we need to be careful to keep our own present-day assumptions from intruding into what they meant, as in the passage about Sujin, by “revering [Amaterasu] publicly (k2zen to 公然と) together with the people of the realm.” The Mito scholars did not see such rituals as “public” in the sense of an open display, a performance directly before the masses. What they anticipated, rather, was that the aura surrounding the ritual would extend beyond the immediate partici pants to encompass society as a whole. Nor did the late Mito thinkers presume that rites recovered from the realm of esoteric practices would be transparent or foster enlightened attitudes. To the contrary, in their eyes, the power of ritual lay in its capacity to “carry” the people along without the latter realizing what was happening, in contrast to direct orders or didactic sermons, which often aroused resistance rather than active cooperation. It was this ability to evoke an instinctive readiness to assist and serve the ruler that made the “public” conduct of rites an effective means of mobilizing the populace to carry out various crucial social tasks.33
The late Mito thinkers further did not assume that public ritual meant open or equal access to the sacred entity that was the object of the ritual. They condemned the graded initiations to different levels of knowledge characteristic of esoteric practices for treating as a personal possession matters that were not “private.” But they took for granted the importance of hierarchy. Seeing their own age as rent by potential divisions, they looked to the institution of a proper structure of ritual to restore and reinforce the hierarchy essential to social stability. In the sacrifices to Earth and Heaven described in Liji, the populace did not worship Heaven and Earth directly; rather, each element in society contributed to the performance of the ritual in the manner appropriate to its station. While involving the entire populace in some way, the ritual at the same time thus clarified and consolidated the social hierarchy. This premise was fundamental to Aizawa’s vision of the daij2sai. It also figured in his identification of an analogue to the she.
In the rituals related to the she, Aizawa pointed out, the ruler honored the land, on which the populace depended for its livelihood, and the deities of the land, whom the people revered. The ruler further conjoined with these sacrifices rites to those who had performed meritorious deeds on behalf of the populace. The efficacy of these practices lay in their capacity to unify the people and secure their allegiance. “If the emperor takes the lead in offering rites to these deities, there is a means by which to extend a regulating authority over the hearts of the people, and their allegiance will be focused on one [rather than divided].” It was precisely with this intent, Aizawa proposed, that Sujin, together with moving the mirror from his residence and creating a public framework for his rites to Amaterasu, had set up ritual structures for the worship of Yamato no Okunitama and another deity, Omononushi 大物主. Aizawa categorized
“Esoteric” and “public” in late Mito thought 369
both of these deities as local founders similar to Okuninushi. They had brought benefits to the people of the Yamato region, where the court was based, by making the land habitable. In taking measures to ensure the maintenance of appropriate sacrifices to them, Sujin had manifested his own reverence for these deities of importance to the populace of the capital region. He thereby had shown the latter that he “took the concerns of the people as his own and brought them to entrust their hopes to the court.” This was just what the ancient Chinese rulers had accomplished through the she.34
As outlying regions were brought under the sway of the court, Sujin had also, Aizawa argued, paid homage to the meritorious founder deities of those areas, such as Okuninushi. This action, which again resembled that of the Chinese sages, had made it possible to temper uncivilized customs and had established the ground for incorporating the populace as a whole into the framework of h2hon hanshi.35 Pursuing lines of interpretation of this sort, Aizawa not only played down the elements that had been used to construct esoteric approaches to deities such as Okuninushi, he also gave those deities a carefully defined place within a hierarchically ordered framework of public ritual centered on the court.
Efforts at implementation
Through the types of interpretive strategies touched on above, the late Mito scholars tried to show that in intent and effect the ritual forms of Japanese antiquity matched those described in the Chinese classics like the “two sides of a tally.”36 They did not undertake this demonstration simply as an abstract academic exercise. To the contrary, for them it was a preliminary step towards establishing a comparable unification of rites and governance in their own day. In the political context of the late Tokugawa period, carrying out such a project on a national scale was beyond the realm of immediate possibility. Even at the local level of the domain, efforts at implementation of more modest elements of the Mito ritual program were thwarted by factionalism and by the sharp ups and downs between 1829 and 1860 in the political situation of the scholars’ patron, the strong-minded Daimyo Tokugawa Nariaki 徳川斉昭 (1800–60).37 But even if partial and sporadic, the attempts of the Mito scholars and Nariaki to put their vision into practice throw into relief some of the implications of the Mito rejection of secrecy and concept of a system of public ritual.
The creation of a system of ritual that “would extend from the palace and be felt throughout the realm” ultimately required action by the rulers of the day. Aizawa hoped, for instance, that the court would revive the long-abandoned ritsury2 practice of sending offerings to important shrines throughout the land on the occasion of major court rites such as the kinensai (toshigoi no matsuri) 祈年祭, the spring rites to pray for good370 Kate Wildman Nakai
crops, and the niinamesai 新嘗祭, the autumn rites of thanksgiving. Like the open-ended divination no longer performed on the occasion of the daij2sai, the offerings sent out to shrines, he pointed out, were crucial means of bringing the realm as a whole within the transformative aura of court ritual. In the absence of the sought-for action from above, the efforts of the late Mito thinkers to implement their vision of the unifica tion of rites and governance were directed primarily at devising ways to “gather the devotion and reverence of the multitudes” and “focus it around the seat of rule.” To this end, in 1834, Aizawa compiled an expla nation “for the ordinary people of rural districts and villages who cannot read characters” of what he regarded as the most important of the rites observed at court and by the domain rulers. One needed to take care not to be disrespectful in listing up the details of court ceremonies, he acknowledged; nevertheless, these were also precisely the things that “the populace should look up to with reverence.”38 Nariaki’s forced retirement in 1844 and the eclipse of the scholars associated with him put an end to a plan to publish this work, titled S2en kagen 草偃和言 (Gentle Words [to Make] the Grass Bend), and to distribute throughout the domain a simplified version of Aizawa’s ritual calendar adapted to village life.39 But from these projects we can glean a picture of the direc tion in which the Mito scholars hoped to move.
As with Tekiihen, likewise meant for popular edification, Aizawa wrote S2en kagen in a straightforward Japanese rather than the ornate kanbun, laden with allusive references to the Chinese classics, that he favored for pieces such as the well-known Shinron 新論 (New Theses, 1825). The parallels with Tekiihen go beyond style. In Tekiihen, as we noted above, Aizawa based his selective account of kamiyo on that in Kitabatake Chikafusa’s Jinn2 sh2t2ki. In S2en kagen, for basic information about the nature of the court rites that he described, he similarly made use of a concise fifteenth-century compendium of the yearly round of court rituals compiled by the then young court noble and scholar (and later regent) Ichij3 Kaneyoshi 一条兼良 (1402–81). Aizawa pointed out that this work, titled Kuji kongen 公事根源 (Fundamentals of Court Ceremo nial), was already “distributed widely.” By making it the foundation of his own account, he thus could preserve a balance between discretion in discussing matters of the court and enlarging general awareness of the things that “the populace should look up to with reverence” (S2en kagen, p. 275). But, just as with Jinn2 sh2t2ki, Aizawa was also highly selective in his use of Kuji kongen. Through what he omitted as well as his further explanations of the significance of what he included, he modified and shifted the orientation of Kaneyoshi’s work even as he built upon it.
In Kuji kongen Kaneyoshi listed month by month a round of more than 170 rites important to court life. These included major court ceremonies dating back to the ritsury2 period like the kinensai and niinamesai, festivals of various shrines, ceremonies of court appointments, various elegant
“Esoteric” and “public” in late Mito thought 371
activities and miscellaneous events, and a large number of Buddhist related ceremonies. From these Aizawa selected a little fewer than thirty. As we might expect, he excluded all ceremonies of Buddhist provenance as well as those having to do with court appointments and elegant amusements. He also omitted many of the shrine festivals and various rites that had a magical import or were intended to dispel malevolent forces, such as the Gion no gory2e 祇園御霊会 and the tsuina 追儺 festivals. Where rituals of this sort were part of the original ritsury2 state rites, Aizawa mentioned them. He kept, for instance, the hishizume no matsuri 鎮火祭 (intended to prevent fires) and the michiae no matsuri 道饗祭 (meant to pacify and keep away malevolent entities). But he left out various details to be found in Kuji kongen, instead stating pointedly that these were minor rites.40
By contrast, Aizawa put prime emphasis on the major state rites dating from the ritsury2 period (in his view, of course, many could be traced to kamiyo or the reigns of the first emperors). Above all he elaborated on rites like the spring kinensai, wherein the emperor sought the assistance of the deities for the successful growing of the year’s crops, and the autumn kannamesai 神嘗祭 and niinamesai in which the ruler made offerings of gratitude for the harvest. In his explanation of the import of these rites, Aizawa repeatedly stressed that the divine blessings of grain and cloth bequeathed originally by Amaterasu were what ensured the people’s sustenance and that the emperor performed these rites of gratitude on their behalf as a means of fulfilling the imperative of h2hon hanshi.
Aizawa not only tried to enlarge his readers’ awareness of these circumstances, he also urged that they engage at the local level in activities that would echo the rites performed at court. On the fourteenth day of the fourth month, for instance, the emperor sent specially woven cloth offerings to Ise. This, Aizawa noted, was in gratitude for Amaterasu’s bequest of the fibers that made it possible for the populace to ward off the cold in winter. For their part, the people, “so as not to forget this sacred benefice, should put on clean clothes, whether newly made or freshly washed, and make a pilgrimage to their local shrine.” Similarly they should hold rites of thanksgiving in their local shrines in the ninth month, when the emperor sent offerings of grain to Ise for the kannamesai, and on the occasion of the niinamesai in the eleventh month, when the emperor personally offered the first fruits of the harvest to the heavenly deities. Gathering together with their relatives and fellow villagers to celebrate these occasions, the people should renew their sense of the debt of obligation that they owed to Amaterasu and to the emperor for performing rites of recompense to her on their behalf (S2en kagen, pp. 291, 303–04, 310). In a similar vein, the simplified version of Aizawa’s ritual calendar meant to be distributed to villages in the domain stipulated that, as the kinensai was held on the fourth day of the second month to pray for the successful growing of the five grains, it would be appropriate for villagers
372 Kate Wildman Nakai
to hold the Inari 稲荷 festival (likewise associated with food and grain) on this day (Mito-han shiry2, vol. 5 [bekki, ge], p. 51).
Dilemmas of practice
By bringing the entire nation within the aura of rituals emanating from the court, the Mito scholars hoped to focus the minds of the populace on one source and foster unstinting service to superiors. In an ironic paradox, however, various tensions embedded in that effort compro mised precisely what the Mito thinkers hoped to accomplish. The height ened emphasis on an Amaterasu-emperor axis as the pivot of a national structure of ritual raised questions, for one thing, about the relationship of that structure to the contemporary bakufu-centered political order. Mito was one of the three main collateral Tokugawa houses, and the late Mito scholars did not question bakufu rule. They in no way intended devotion to Amaterasu to detract from the fulfillment of feudal obligations to the bakufu. To the contrary, shogunal reverence for the emperor, who offered dedicated service to Amaterasu, would, they argued, in turn encourage renewed commitment to the shogun by the daimyo and their retainers. Properly oriented reverence would have a chain effect of fostering loyalty and whole hearted service to superiors throughout the feudal hierarchy.41 Reflecting their acknowledgment of the importance of the obligations owed the bakufu, the Mito scholars called for expressions of homage to Tokugawa Ieyasu as well as to Amaterasu. Aizawa, for instance, included the anniversary of Ieyasu’s death in S2en kagen. It was owing to Ieyasu’s efforts that the people of the present age had enjoyed over two hundred years of peace and security, he wrote. On that day the people should gather with their relatives and neighbors to reflect with gratitude on Ieyasu’s accomplishments (S2en kagen, pp. 292–94). We may read in such arrangements an attempt to strike a balance between commitment to an Amaterasu-emperor axis, on the one hand, and to the bakufu, on the other. The strife between pro-court and pro-bakufu factions that left a bloody imprint on the domain’s final years suggests, however, that balance was in actuality not easy to maintain.
Inherent in the emphasis on Amaterasu as the source of blessings for the entire populace and on the necessity for the realm as a whole to offer recompense was likewise the danger that it might encourage contravention of the very social hierarchy it was supposed to reinforce. The late Mito thinkers held that this should not be the case, that by specifying an appropriate role for each participant, a national structure of rites centered ultimately on reverence to Amaterasu should confirm the place of each within a class-based hierarchical order. The Mito scholars remained sensi tive, nevertheless, to the potentially negative consequences of the populace expressing devotion to Amaterasu in an unmediated form. One of the grave faults of Christianity and Buddhism, Aizawa asserted, was
“Esoteric” and “public” in late Mito thought 373
encouraging the masses to worship Heaven or an overarching deity directly. In so doing, these doctrines inevitably sowed the seeds of disruption of the social hierarchy. Recognizing that the emperor in making offerings to Amaterasu was also acting on their behalf, the people should not presume to duplicate his role. Rather, they should feel a doubled sense of gratitude, to the emperor as well as to Amaterasu (Tekiihen, pp. 259–60). Referring to the ancient prohibition of private worship at Ise, Fujita T3ko emphasized that the performance of rites to Amaterasu was the prerogative of the court; ordinary people should not engage in “private” rites to her.42 Even as the Mito scholars exalted Amaterasu as a font of blessings to the populace, they thus also called for the preservation of a respectful distance.
On the level of practice, the late Mito thinkers promoted several ways of bridging the contradiction in this situation. One was the mechanism of replication that Aizawa advocated in S2en kagen. By coordinating their local village rites with the court ritual calendar, the populace should echo the rites conducted by the emperor. But they were to make their offerings to the deities of their own area, not directly to higher deities such as Amaterasu, honored by the emperor. In effect they were to engage in what we might call “honoring vicariously”: that is, they were to offer rites to hierarchically appropriate deities who could in turn be linked to service to Amaterasu. As Aizawa put it, the tutelary shrines of local areas were dedicated to deities “who had all in antiquity assisted the Sun Deity (hi no kami 日神) in her heavenly task and had pacified and nurtured the local populace. To pay homage to these deities thus accords with the principle of thanking the Sun Deity for her blessings.”43 The choice of the deity to enshrine at the K3d3kan 弘道館, the domain school founded under Nariaki, was based on this notion. The deity Nariaki and his advisers ultimately selected was Takemikazuchi 建雷. They saw him as particularly suitable because he was the deity of Kashima 鹿島, the leading shrine (ichi no miya 一宮) of Hitachi province, where Mito was located (although Kashima itself was not within Mito domain territory). As a military deity, Takemikazuchi was also an appropriate object of homage for a warrior house. Most important, his action in pacifying various unruly deities and paving the way for Ninigi’s descent had provided vital assistance in the heavenly task of securing the welfare of the realm. As such it exemplified the height of service to Amaterasu. For the retainers of the Mito domain to offer rites to Takemikazuchi would inculcate in them awareness of the hierarchically correct forms of honoring Amaterasu as well.44
Such devices did not fully suffice, however, to contain the momentum generated by the focus on reverence for the imperial line and Amaterasu as the ultimate mechanism for ensuring that “the multitudes would be of one mind.” Even among the Mito thinkers wavering can be seen on this point. In discussions in 1834 over the choice of the deity to honor at the
374 Kate Wildman Nakai
K3d3kan, for instance, Nariaki initially favored enshrining Emperor Jinmu, Ojin 応神, or Tenji 天智, or a combination of the three. A year earlier he proposed as a separate project establishing a shrine to Jinmu within the domain. Aizawa and Nariaki’s other advisers demurred on the ground, in the first case, that it was not appropriate to enshrine a figure as awesome as the emperor in a school, and, in the second, that it did not accord with correct ritual for a feudal lord to offer rites to an emperor.45 Some years later, however, Aizawa had second thoughts about this issue. Noting in 1848 the urgency of creating proper venues for meeting the spiritual needs of the populace and thus forestalling the corrupting influence of Buddhism and heterodox forms of popular worship, he called for the establishment of new shrines of orthodox antecedents. Central among those he proposed was a shrine to Jinmu at Kashihara 橿原 in Yamato, near the first emperor’s putative place of burial. At the same time Aizawa also expressed regret that the plan to build a shrine to Jinmu in Mito had not come to fruition.46
Similarly, parallel to his emphasis on honoring Amaterasu vicariously through hierarchically differentiated modes, Aizawa also pursued ways of expressing reverence to her in a more immediate manner. As a key means of doing so while avoiding the problems of direct worship, he advocated showing “reverence from a distance” (y2hai 遙拝). The ante cedents of this practice could be traced to the second lord of the domain, Mitsukuni 光圀 (1628–1700), who every New Year, attired in the court dress appropriate to his rank, had bowed toward the emperor in Kyoto. In 1837 Aizawa proposed to incorporate an extension of this custom in the ritual to be employed at the K3d3kan. He recommended that a special altar (dan 壇) be erected to the rear of the shrine to Takemika zuchi as a site for manifesting reverence to Amaterasu and to the emperors from Jinmu down to the current occupant of the throne. On the first day of classes for the year, to be established as the day for the shrine festival, the domain lord and his retainers were to wear formal dress and from this altar bow towards Kyoto. Thereafter those present were to enter the school hall and bow in homage, first to a scroll inscribed with the gene alogy of the imperial line, beginning with Amaterasu, and then to other scrolls depicting the lineage of the Tokugawa shogunal house from Ieyasu on and the lineage of the domain lord. In confirmation of the unity of Confucian and Shinto teachings supporting these expressions of devotion, the participants were then to listen to a set of ceremonial lectures on the passage from Nihon shoki describing Amaterasu’s bestowal of the regalia on Ninigi, the liturgy (norito 祝詞) recited in honor of Amaterasu on occasions like the kinensai, and the first verse of the Analects.47
Through the creation of a structure of “public” ritual centered on Amaterasu, the Mito scholars aimed to reinforce the Tokugawa social and political order, not weaken it. In their eyes, the prevalence of “secret transmissions” purveying esoteric interpretations of Amaterasu and the
“Esoteric” and “public” in late Mito thought 375
events of kamiyo served only to encourage “private speculation” and a presumptuous manipulation of things that individuals should not treat as a personal possession. Secrecy was an invitation to social chaos. Promotion of “public” ritual centered on Amaterasu would counter the threat of chaos and inculcate throughout society the dynamic of h2hon hanshi. Such a public ritual would thereby also redound, they believed, to the benefit of the shogun and the Tokugawa order. In fact, however, the Mito scholars’ exaltation of Amaterasu acted as an undertow, pulling their project in a direction beyond the parameters of their vision. As such it contributed, albeit inadvertently from their perspective, to the disinte gration of the ideological foundations of the Tokugawa feudal system as they conceived it.
The Mito project had another unintended consequence: the move in the Meiji period and later towards the formulation of a system of public ritual that presumed mass involvement of the populace. The forms such involvement took hardly preserved the Mito scholars’ ideal of a ritual structure that would embody and reinforce a class-based social hierarchy. Yet embedded in the modern imperial state was the legacy of many aspects of the late Mito deconstruction of secrecy and call for ritual embodiment of h2hon hanshi – from Aizawa’s ritual calendar to the estab lishment of a shrine to Jinmu and the institution of y2hai of Amaterasu and the emperor as national practices.
Notes
1 For a discussion of this incident and the relevant sources, see Hori 1964, pp. 39–54.
2 Hayashi Razan, Shint2 denju, p. 19.
3 Scheid 2002, p. 306. See also Taira 1972, p. 518.
4 See Takashima 1992, p. 3; Ooms 1985, pp. 221–86.
5 Ogy5 Sorai, Bend2, pp. 26–28. Here and below, all citations to the Nihon shis2 taikei (NST) edition of works written in kanbun are to the kakikudashi 書き下し version.
6 I have previously discussed some of the links between Sorai’s thought and the Mito school (and have also taken up from a different perspective a number of the points addressed below) in Nakai 2002. See also Bit3 1973; Wakabayashi 1986, pp. 117–19.
7 Aizawa Seishisai, Shinron, p. 52. See also the translation of Shinron in Wakaba yashi 1986, p. 152.
8 Aizawa Seishisai, Doku Naobi no mitama, p. 36.
9 聖人以神道設教、而天下服矣. Explanation of the guan 觀 (Jap. kan) hexagram; translation (slightly modified) from Wilhelm 1967, p. 486.
10 Shinron, p. 140; Wakabayashi 1986, p. 252.
11 Fujita T3ko, K2d2kanki jutsugi, p. 264. In Hitachi obi 常陸帯, T3ko described the “so-called Shinto of the day” in terms similar to those used by Dazai Shundai: it had been fashioned by “incorporating the theories of Yin Yang and five-elements cosmology or by in effect adopting the ideas of Confucian [metaphysics] and Buddhism.” Fujita T3ko, Hitachi obi, p. 120. The language T3ko used in K2d2kanki jutsugi to describe the corruption of Shinto sums up
376 Kate Wildman Nakai
the Mito understanding of the phenomenon referred to in this book as “esotericism”: it meant “wild private speculation” (midari ni shichi o motte 妄り に私智を以て), “allegorical” (g4gen 寓言) interpretations that turned the records of kamiyo into “riddles” (s2ji 辞) or a “secret code” (ingo 隠語), and the concealing of the shallowness and dubious nature of such things by
putting them in the form of “secret transmissions” (hiketsu 秘訣). 12 The late Mito scholars’ approach to kamiyo was part of a broad shift from allegorical to “historical” interpretations that took place in the course of the Tokugawa period. I have discussed these developments in a contribution to a forthcoming volume based on a conference held on Maui in November 2000 that was jointly sponsored by the International Research Center for Japanese Studies, Kyoto, and the University of Hawai’i.
13 English requires gendered pronouns, and, in accordance with the assump tion generally accepted today that Amaterasu was female, here I will use “she” and “her” in referring to this deity. We should keep in mind, however, that the Mito thinkers, like many other medieval and early modern figures, may not have conceived of Amaterasu in these terms. Aizawa took issue with thinkers who asserted Amaterasu was female, arguing that Nihon shoki and Kojiki were not explicit on this point. See, for instance, Doku Kuzubana, p. 22; Doku Maganohire, p. 8. As noted above and discussed further below, the Mito thinkers often alluded to Amaterasu as Tenso. Although this term could be traced back to early texts such as Kogo sh4i 古語拾遺, the Mito scholars used it in a manner evocative of the posthumous names of the founders of Chinese dynasties. As such it also conveyed an image of Amaterasu as male rather than female. For details, see pp. 362–63 and note 19.
14 Shinron, p. 56; Wakabayashi 1986, p. 158; Aizawa Seishisai, Kagaku jigen, p. 30b.
15 Kitabatake Chikafusa, Jinn2 sh2t2ki, p. 60; Varley 1980, p. 77. 16 Jinn2 sh2t2ki, p. 41: 天祖ハジメテ基ヲヒラキ、日神ナガク統ヲ伝給フ . The editors of the NKBT edition gloss the characters Tenso as Amatsumioya. See also Varley 1980, p. 41.
17 Note 6 on p. 41 of the NKBT edition of Jinn2 sh2t2ki indicates that Chikafusa followed Watarai Shinto ideas in substituting tenso for tenjin 天神 and speci fying tenso to be Kuni no Tokotachi.
18 Shinron, p. 53; Wakabayashi 1986, p. 154.
19 Tekiihen, pp. 254–55. Compare Jinn2 sh2t2ki, pp. 60–61; Varley 1980, pp. 76–77. Aizawa also elided phrases where Chikafusa specifically identified Amaterasu as female (compare Tekiihen, p. 253; Jinn2 sh2t2ki, pp. 52–53; Varley, p. 66). These omissions support the supposition that Aizawa likely preferred to think of Amaterasu as male rather than female (see note 13).
20 Tekiihen, pp. 257–59. Chikafusa notes in passing that the rice people eat everyday is an imperial benefice originating from the seeds sown by Amaterasu (Jinn2 sh2t2ki, p. 163; Varley 1980, p. 230). But the image of Amaterasu as a source of material blessings is not nearly so central to his account as it is to the Mito view.
21 Nihon shoki, vol. 1, pp. 150–51; Aston 1972, vol. 1, p. 80.
22 Yamazaki Ansai, Jindaikan k2gi, pp. 166–70; Takashima 1992, pp. 577–80. 23 See, for instance, Hirata Atsutane, Tama no mihashira, pp. 72–77. 24 Shinron, pp. 142, 148; Wakabayashi 1986, pp. 257, 264.
25 See Kagaku jigen, p. 43b; Mito-han shiry2, vol. 5 (bekki, ge), p. 317. 26 Liji 11 (“Jiaotesheng” 郊特牲), vol. 2, pp. 394–95, 398–99; Legge 1964, vol. 1, pp. 425–26, 429–31.
27 Aizawa Seishisai, Taishoku kanwa, pp. 237–38, 241–42. The late Mito scholars were, of course, not the only ones to find fault with existing forms of Shinto.
“Esoteric” and “public” in late Mito thought 377
Their near contemporaries, the Kokugaku thinkers, likewise attacked Yoshida Shinto and the early Edo Confucian Shinto schools as corruptions of the true Japanese tradition. The Kokugaku and Mito thinkers differed sharply, however, in their conception of what constituted the true tradition. The Mito scholars criticized in particular the Kokugaku denial of its compatibility with Confucian norms and the methods of rule of the Chinese sages. According to the Mito scholars, a key aspect of the lack of balance in the perspective of the Kokugaku thinkers was their rejection of the “culture” (bun) that was an essential element of Confucian learning and their overexaltation of “simplicity.” The “simplicity” of the supposedly “ancient” outlook that the Kokugaku thinkers claimed to recover was, the Mito scholars argued, in fact little more than a fictive creation of their own that drew from Taoism and even Christianity. See K2d2kanki jutsugi, pp. 264–65; Hitachi obi, pp. 102–03.
28 Nihon shoki, vol. 1, pp. 238–39; Aston 1972, vol. 1, pp. 151–52. 29 Shinron, pp. 140–41; Wakabayashi 1986, pp. 253–54.
30 As I have noted in Nakai 2002, p. 289, Aizawa traced the first performance of the daij2sai to the rites that Jinmu, according to Nihon shoki, conducted at Tomiyama 鳥見山 following his succession. Nihon shoki, vol. 1, pp. 214–15; Aston 1972, vol. 1, p. 134.
31 Shinron, p. 55; Wakabayashi 1986, p. 157. See Liji 24 (“Jiyi” 祭義), vol. 2, pp. 701, 719; Legge 1964, vol. 2, pp. 211, 226.
32 Kagaku jigen, p. 36b. See also Shinron, p. 150; Wakabayashi 1986, pp. 267–68. 33 For a succinct statement of Aizawa’s views on this aspect of the power of a public system of ritual, views which owe much to the ideas of Ogy5 Sorai, see Kagaku jigen, p. 26a.
34 Shinron, p. 141; Wakabayashi 1986, pp. 254–55.
35 Shinron, pp. 142–43; Wakabayashi 1986, pp. 256–57.
36 Shinron, pp. 148. As he notes, Wakabayashi omits from his translation the passage containing this phrase. See Wakabayashi 1986, pp. 264, 313–14. 37 On the Mito reform program pursued by Nariaki and his advisers, see Koschmann 1987.
38 S2en kagen, pp. 318, 309.
39 Aizawa revised S2en kagen in 1843 in preparation for its publication; in a post script he took note of plans to distribute the published version among the village leadership stratum. The priest of a major domain shrine ultimately sponsored its publication in 1852.
40 Kuji kongen, for instance, noted of both the michiae no matsuri and the hishizume no matsuri that they were managed by the Urabe 卜部 lineage and of the latter that “many secret techniques (hijutsu 秘術) are said to be used during this rite.” Aizawa omitted these details. See Ichij3 Kaneyoshi, Kuji kongen, p. 76; S2en kagen, p. 301.
41 Fujita Y5koku, Seimeiron, p. 13. See also Shinron, pp. 153–54; Wakabayashi 1986, pp. 271–72.
42 K2d2kanki jutsugi, p. 316. Elsewhere T3ko described direct worship of Amate rasu by the populace as “tantamount to going over the head of the lord of one’s domain and making a direct appeal to the court. It is an extreme abro gation of proper ritual forms.” Hitachi obi, p. 120.
43 S2en kagen, p. 304. Implicit in such a strategy, of course, was the necessity to clarify the origins of local tutelary deities and to free them of syncretic corrup tions. See Hitachi obi, p. 120.
44 K2d2kanki jutsugi, pp. 316–17; Hitachi obi, pp. 105–06.
45 Mito-han shiry2, vol. 5 (bekki, ge), pp. 264, 266; Mito-han shiry2, vol. 4 (bekki, j2), pp. 200–01.
378 Kate Wildman Nakai
46 Aizawa Seishisai, K2ko fuken, pp. 529–31. Aizawa urged as well creating shrines to Sujin and Tenji. There already were shrines to Ojin (i.e., those to Hachiman 八幡), he acknowledged, and these would suffice if purged of Buddhist elements. Ibid.
47 Aizawa Seishisai, Taimon sansaku, p. 183. To indicate the unity of Confu cianism and Shinto and the indivisibility of the civil and military arts (bun 文 and bu 武), Nariaki and his advisers also established a hall to Confucius in the K3d3kan parallel to the shrine to Takemikazuchi. Aizawa recommended that on the day of the sekiten 釈奠 service performed in honor of Confucius, a cere mony similar to that conducted on the first day of classes, with lectures on the same texts, should be held before a scroll inscribed with Confucius’s name. Ibid.
References
Primary sources
Bend2 弁道, by Ogy5 Sorai 荻生徂徠. In Ogy4 Sorai 荻生徂徠 (= NST 36), 1974. Bend2 sho 弁道書, by Dazai Shundai 太宰春台. In Nihon shis2 t2s2 shiry2 日本思想闘 争史料, vol. 3. Tokyo: Meicho Kank3kai, 1969.
Doku Kuzubana 読葛花, by Aizawa Seishisai 会沢正志斎. In Nihon jurin s2sho 日本儒 林叢書, vol. 4. Tokyo: H3 Shuppan, 1971.
Doku Maganohire 読末賀能比連, by Aizawa Seishisai. In Nihon jurin s2sho 日本儒林 叢書, vol. 4. Tokyo: H3 Shuppan, 1971.
Doku Naobi no mitama 読直毘霊, by Aizawa Seishisai. In Nihon jurin s2sho 日本儒 林叢書, vol. 4. Tokyo: H3 Shuppan, 1971.
Hitachi obi 常陸帯, by Fujita T3ko 藤田東湖. In Kikuchi Kenjir3 菊池謙二郎 (ed.), T2ko zensh4 東湖全集 Tokyo: Hakubunkan, 1940.
Jindaikan k2gi 神代巻講義, by Yamazaki Ansai 山崎闇斎. In Kinsei shint2 ron 近世神 道論 (= NST 39), 1972.
Jinn2 sh2t2ki 神皇正統記, by Kitabatake Chikafusa 北畠親房. In NKBT 87, 1965. Kagaku jigen 下学邇言, by Aizawa Seishisai. Privately published, 1892. K2d2kanki jutsugi 弘道館記述義, by Fujita T3ko. In Mitogaku 水戸学 (= NST 53), 1973.
K2ko fuken 江湖負暄, by Aizawa Seishisai. In Mitogaku 水戸学 (= ST, vol. Ronsetsuhen 論説編 15), 1986.
Kuji kongen 公事根源, by Ichij3 Kaneyoshi 一条兼良. In Jikkinsh2, Kuji kongen 十訓 抄 . 公事根源 (=Nihon bungaku zensho 日本文学全書, vol. 22). Tokyo: Hakubunkan, 1892.
Liji (Raiki) 礼記. Shinshaku kanbun taikei 新釈漢文大系, vols 27–29. Tokyo: Meiji Shoin, 1977.
Mito-han shiry2 水戸藩史料, 5 vols. Tokyo: Yoshikawa K3bunkan, 1970. Nihon shoki 日本書紀. NKBT 67–68, 1965–67.
Seimeiron 正名論, by Fujita Y5koku 藤田幽谷. In Mitogaku 水戸学 (=NST 53), 1973. Shinron 新論, by Aizawa Seishisai. In Mitogaku 水戸学 (= NST 53), 1973. Shint2 denju 神道伝授, by Hayashi Razan 林羅山. In Kinsei shint2 ron 近世神道論 (= NST 39), 1972.
S2en kagen 草偃和言, by Aizawa Seishisai. In Mitogaku 水戸学 (= ST, vol. Ronsetsuhen 論説編 15), 1986.
“Esoteric” and “public” in late Mito thought 379
Taimon sansaku 対問三策, by Aizawa Seishisai. In Mitogaku 水戸学 (= ST, vol. Ronsetsuhen 論説編 15), 1986.
Taishoku kanwa 退食間話, by Aizawa Seishisai. In Mitogaku 水戸学 (=NST 53), 1973. Tama no mihashira 霊の真柱, by Hirata Atsutane 平田篤胤. In Hirata Atsutane 平田篤胤 (= NST 50), 1973.
Tekiihen 迪彜編, by Aizawa Seishisai. In Shinron, Tekiihen 新論 ·迪彜編. Tokyo: Iwanami Shoten, 1941.
Modern sources
Aston, W. G. (tr.) (1972), Nihongi: Chronicles of Japan from the Earliest Times to A.D. 697. Tokyo and Rutland, VT: Charles E. Tuttle (reprint; 11896 in 2 vols). Bit3 Masahide 尾藤正英 (1973), “Mitogaku no tokushitsu” 水戸学の特質, in Mitogaku 水戸学 (= NST 53), pp. 556–82.
Hori Isao 堀勇雄 (1964), Hayashi Razan 林羅山. Tokyo: Yoshikawa K3bunkan. Kaempfer, Engelbert (1999), Beatrice M. Bodart-Bailey (ed. and tr.), Kaempfer’s Japan: Tokugawa Culture Observed. Honolulu: University of Hawai’i Press.
Koschmann, J. Victor (1987), The Mito Ideology: Discourse, Reform, and Insurrection in Late Tokugawa Japan, 1790–1864. University of California Press. Legge, James (tr.) (1964), The Li Ki (= The Sacred Books of the East, vols 27–28). Delhi: Motilal Banarsidass (reprint; 11885).
Nakai, Kate Wildman (2002), “Chinese Ritual and Native Identity in Tokugawa Confucianism,” in Benjamin A. Elman, John. B. Duncan, and Herman Ooms (eds), Rethinking Confucianism: Past and Present in China, Japan, Korea, and Vietnam. Los Angeles, CA: UCLA Asian Pacific Monograph Series, pp. 258–91.
Ooms, Herman (1985), Tokugawa Ideology: Early Constructs, 1570–1680. Princeton: Princeton University Press.
Scheid, Bernhard (2002), “Shinto as a Religion for the Warrior Class: The Case of Yoshikawa Koretaru.” Japanese Journal of Religious Studies 29/3–4, pp. 299–324. Taira Shigemichi 平重道 (1972), “Kinsei no shint3 shis3” 近世の神道思想, in Kinsei shint2 ron 近世神道論 (= NST 39), pp. 507–58.
Takashima Motohiro 高島元洋 (1992), Yamazaki Ansai: Nihon Shushigaku to Suika shint2 山崎闇斎:日本朱子学と垂加神道. Tokyo: Perikansha.
Varley, H. Paul (tr.) (1980), A Chronicle of Gods and Sovereigns: Jinn3 Sh3t3ki of Kitabatake Chikafusa. New York: Columbia University Press.
Wakabayashi, Bob Tadashi (1986), Anti-Foreignism and Western Learning in Early Modern Japan: The New Theses of 1825. Cambridge, MA: Council on East Asian Studies, Harvard University.
Wilhelm, Richard (tr.) (1967), The I Ching. Princeton, NJ: Princeton University Press.