Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Bushido : the creation of a martial ethic in late Meiji Japan Benesch, Oleg 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2011_spring_benesch_oleg.pdf [ 1.48MB ]
JSON: 24-1.0071589.json
JSON-LD: 24-1.0071589-ld.json
RDF/XML (Pretty): 24-1.0071589-rdf.xml
RDF/JSON: 24-1.0071589-rdf.json
Turtle: 24-1.0071589-turtle.txt
N-Triples: 24-1.0071589-rdf-ntriples.txt
Original Record: 24-1.0071589-source.json
Full Text

Full Text

BUSHIDO: THE CREATION OF A MARTIAL ETHIC IN LATE MEIJI JAPAN  by OLEG BENESCH  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES (Asian Studies)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  FEBRUARY 2011  © Oleg Benesch, 2011  ABSTRACT This study examines the development of the concept of “bushido,” or the “way of the warrior,” in modern Japan, focusing on the period between the mid-nineteenth century and the early 1930s. The popular view holds that bushido was a centuries-old code of behavior rooted in the historical samurai class and transmitted into the modern period, where it was a fundamental component of Japanese militarism before 1945. In fact, the concept of bushido was largely unknown before the last decade of the nineteenth century, and was widely disseminated only after 1900, especially after the Russo-Japanese War of 1904-5. This study argues that modern bushido discourse began in the 1880s, and was dependent on political and cultural currents relating to Japan’s modernization and the nation’s attempts to redefine itself in the face of foreign “others,” primarily China and the West. Following more than a decade of largely unquestioned thrusts towards modernization and Westernization after 1868, Japanese thinkers looked to their own traditions in search of sources of national identity. The first discussions of bushido at this time were not the work of conservative reactionaries, however, but were conceived by relatively progressive individuals with considerable international experience and a command of Western languages. Some of the first modern writings on bushido clearly posit the concept as a potential native equivalent to the English ethic of “gentlemanship,” which was widely admired in late-nineteenth century Japan, and much of early bushido discourse should be seen primarily as a response to outside stimuli. This study examines the causes and effects of the “bushido boom” that took place between 1898 and 1914, which firmly established the concept not only in Japan, but throughout the world. In this context, this study analyzes the use of bushido by the Japanese military and educational system, as well as its popularization by prominent figures in the early twentieth century. This study also examines the reasons for the decline in the popularity of bushido between 1914 and the early 1930s, thereby providing points of departure for future research on the trajectory of bushido from 1932 to the present day.  ii  TABLE OF CONTENTS ABSTRACT ..............................................................................................................................ii TABLE OF CONTENTS …………………………………………………………………….iii ACKNOWLEDGEMENTS…………………………………………………………………...v INTRODUCTION .....................................................................................................................1 The etymology and historiographical uses of bushidō ..........................................................5 Warrior behavior in Japanese history before 1600 ..............................................................14 Pre-modern writings on bushidō..........................................................................................20 Overview of the existing literature on modern bushidō ......................................................28 The goals of this study.........................................................................................................30 CHAPTER 1 Bridges between Tokugawa and Meiji Bushidō.............................................37 Yoshida Shōin......................................................................................................................38 Yokoi Shōnan ......................................................................................................................45 The curious case of Yamaoka Tesshū .................................................................................51 CHAPTER 2 First Explanations of Bushidō in the Meiji 20s ..............................................68 Theoretical considerations ...................................................................................................68 Development of Japan’s views of itself and its relationships with China and the West .....72 The bushidō of Ozaki Yukio ...............................................................................................76 Fukuzawa Yukichi’s view of martial honor ........................................................................86 Suzuki Chikara: bushidō, militarism, and the rise of Japanese cultural nationalism ..........96 Uemura Masahisa and early connections between bushidō and Christianity....................107 The redefinition of budō and the evolution of the term bushidō between 1885 and 1895 118 CHAPTER 3 The Early Bushidō Boom, 1898-1905 ..........................................................126 Currents in Japanese thought in the late 1890s..................................................................127 The Middle Kingdom becomes peripheral ........................................................................129 Reconsideration of the West and the production of national culture.................................131 A first collection of writings on bushidō ...........................................................................137 Prominent contributors to the Bushidō zasshi ...................................................................142 The impact and limitations of the Bushidō zasshi .............................................................148 Nitobe Inazō and the internationalization of bushidō........................................................151 Looking at Bushido: The Soul of Japan ............................................................................159 Nitobe and contemporary intellectual trends.....................................................................166 Nitobe’s impact on bushidō discourse...............................................................................168 Inoue Tetsujirō: the doyon of 20th-century bushidō .........................................................170 Intellectual currents that flowed into Inoue’s bushidō ......................................................173 Inoue’s entry into bushidō discourse .................................................................................177 iii  A Collection of Bushidō Theories by Prominent Modern Thinkers...................................183 Inoue’s influence on the bushidō boom.............................................................................192 CHAPTER 4 The Late Bushidō Boom, 1905-1914............................................................195 Orthodox bushidō after the Russo-Japanese War..............................................................204 The appearance of bushidō in spiritual education .............................................................212 Bushidō in literature in late Meiji and early Taishō ..........................................................221 Adoption of bushidō for historical and patriotic legitimization ........................................233 The development of bushidō outside of Japan ..................................................................256 Conclusions regarding the late bushidō boom...................................................................270 CHAPTER 5 The End of the Bushidō Boom .....................................................................272 Nogi Maresuke and the end of the bushidō boom .............................................................273 Taishō bushidō...................................................................................................................283 The resurgence of bushidō in Shōwa.................................................................................289 Considering the end of the Meiji bushidō boom ...............................................................298 CONCLUSIONS ...................................................................................................................300 WORKS CITED ....................................................................................................................312  iv  ACKNOWLEDGEMENTS I am deeply indebted to the faculty, staff, and students at the University of British Columbia who have offered me constant support, guidance, and insight from my arrival in Vancouver onward. I owe special thanks to Dr. Nam-lin Hur and Dr. Peter Nosco, whose astute guidance and incisive questions are responsible for any contributions this study may make to the field, and whose excellent practical advice and support has made my research possible. I thank Dr. William Wray for introducing me to valuable and methods and resources for Japanese historical research. I am also indebted to Dr. Tomoko Kitagawa and Hidemi Shiga for their engaging and challenging discussions in various courses and seminars we attended together, and also for directing me towards several sources that proved to be most useful to my project. I owe a great debt of gratitude to Jasmina Miodragovic, whose tireless and outstanding efforts on behalf of graduate students in Asian Studies have allowed this and many other theses and dissertations to be realized. In Japan, I extend my thanks to Dr. Shuntarō Itō for initially encouraging me to conduct research on bushidō as the subject of my MA thesis at Reitaku University. I am extremely grateful to Dr. Yutaka Yoshida for invaluable advice and guidance on conducting historical research in Tokyo, and for providing me with the opportunity to do so. I thank Dr. Toshitada Kitsukawa for introducing me to new sources and approaches in Japanese historiography that have fed directly into this thesis. I am indebted to Helena Simmonds, who has not only graciously and patiently endured, but has even assisted in the completion of this study. I also thank Renate, Walter, and Ilya Benesch, who have both morally and practically supported me through many years of education and research in faraway lands.  v  INTRODUCTION “Bushidō is death.” -Yamamoto Tsunetomo (early 18th century) “Unformulated, Bushido was and still is the animating spirit, the motor force of our country.” -Nitobe Inazō (1900)1 “I believe that the nation as a whole, and every individual Japanese, as well, should now once again return to the bushidō spirit.” -Nakatsugawa Hirosato (2004)2 Since the end of the nineteenth century, the Japanese concept of bushidō has been the subject of heated debate among scholars, politicians, writers, and everyday people both in Japan and abroad.3  The proponents of bushidō have often gone so far as to posit it as the very “soul”  of the Japanese people, while its critics, fewer in number and lesser in public influence, tend to dismiss it out of hand, and seldom devote more than a few pages to the subject. These two diametrically opposed approaches to bushidō have resulted in several hundreds of books and articles dedicated to defining what bushidō is, and a mere handful of articles and passages in larger, peripherally-related works that attempt to show what bushidō is not.  To date, no  definitive works that can lay claim to the title of “authoritative representative” of either position have emerged. Instead, well over a century after its initial publication in 1900, Nitobe Inazō’s enigmatic Bushido: the Soul of Japan, continues to hold sway as the best-known work on the  1  Nitobe Inazo, Bushido: The Soul of Japan, 98. Minutes of National Diet Budget Committee meeting held on February 24, 2004. 3 Although the term “bushidō” (武士道) is frequently translated as “the way of the warrior,” this translation becomes extremely problematic when discussing the history of the subject, as it is only one of many terms found in Japanese texts dealing with the issue. This is discussed in more depth in Part 1, and for the sake of eliminating as much ambiguity as possible, this study will rely on Romanization of the original Japanese terms to the extent that it is practical to do so. 2  1  topic in any language, and has been described as “being a textbook-like standard among books that have been written thus far under the heading bushidō.”4 One immediate difficulty facing interpreters of bushidō lies in the very existence of a historical Japanese warrior code or codes.  The popular view holds that bushidō developed as a  martial ethic along with the rise of warrior power around the Kamakura period, although bushi were too busy fighting to formally codify it before the end of the sixteenth century. According to this interpretation, aspects of bushidō changed as Japanese society changed, until the samurai class was ultimately eliminated in early Meiji. It is commonly held that the ethic was then appropriated and adapted by the Meiji, Taishō, and Shōwa governments as a ruling ideology that redirected loyalty from feudal lords to the emperor.5  After 1945, many thinkers dismissed what  they regarded as corrupting modern developments in bushidō and returned to examinations of the historical samurai to draw conclusions regarding “traditional” Japanese culture and behavioral patterns.  In contrast, especially in recent years, a number of scholars have emerged who  dismiss the notion that a widely-accepted warrior ethic existed in pre-modern Japan, and maintain that bushidō is a much more recent phenomenon.6  Regardless of whether scholars  focus on the modern or pre-modern periods in their discussions of bushidō, most agree that there are significant differences between the martial ethics of these two periods, although the reasons given for this disconnect vary greatly. Discussions of Meiji bushidō have tended to focus on the writings of Nitobe, Uchimura Kanzō, and other Christian thinkers.7  Some scholars, such as Kanno Kakumyō, divide modern  bushidō into competing “Christian” and “nationalistic” types, but this assessment is simplistic  4  Takahashi Tomio also lauds Nitobe’s work as fundamentally marked by “an outstanding understanding and organization of bushidō” (Takahashi Tomio, Bushi no kokoro, Nihon no kokoro, 2: 426-427). 5 According to Takahashi Tomio, imperial troops were expected to unite the virtues of martial valor as well as Confucian decorum and loyalty (Takahashi, Bushidō no rekishi Volume 3, 153). 6 Some of this research discusses government use of bushidō as a militaristic ideology in the years leading up to the Pacific War, when the concept was an important component of military education and propaganda, but there is still a great deal of work to be done in this specific field. 7 For example, Matsumae Shigeyoshi’s 1987 Budō shisō no tankyū.  2  and does not satisfactorily explain the development of either of these strains of thought, nor their relationship with earlier ideas.8  Further, most examinations of bushidō fail to account for the  continued popularity of the concept in Japan today, when the country is not populated substantially by Christians or militant nationalists. This study will examine the modern evolution of bushidō, first by analyzing the state of the concept at the end of the Tokugawa period.  Although several writers on the subject  featured prominently in twentieth-century bushidō discourse, this study will show that this was a retroactive development rather than a continuation of samurai thought from Edo to Meiji.  This  introduction will give a brief overview of pre-Meiji warrior history and thought, but this study does not focus on the evolution of martial ethics during these periods, which has been dealt with in other works.  As I will show, an idealized view of pre-Meiji history provided some  inspiration and points of reference for modern bushidō theorists, but historical texts and events were carefully selected and interpreted to legitimize and support modern agendas.  For this  reason, pre-Meiji samurai history has limited relevance to modern bushidō, and I will restrict discussion of earlier history to those instances in which direct connections between the two were drawn. In this study, I will demonstrate that the development and dissemination of bushidō from the 1880s onward was an organic process initiated by a diverse group of thinkers who were more strongly influenced by the dominant Zeitgeist and Japan’s changing geopolitical position than by any traditional moral code. These individuals were concerned less with Japan’s past than the nation’s future, and their interest in bushidō was prompted primarily by their considerable exposure to the West, pronounced shifts in the popular perception of China, and an apprehensiveness regarding Japan’s relative strength among nations.  Rather than being  conservative reactionaries or expansionist militarists, however, the initiators of modern bushidō  8  Kanno, Bushidō no gyakushū.  3  included men who were among the most socially progressive advocates of diplomatic discourse between nations.  This study argues that this multilateral process was responsible for the  subsequent popularization of bushidō in a “bushidō boom” with international ramifications between 1898 and 1914. This bushidō boom, which I closely examine with regard to its influence on politics, literature, the military, and sport, is further significant as a model for the cycles of bushidō resurgence that occurred during early Shōwa and then again from the 1980s onward. The fact that the concept of bushidō retains an increasingly high profile in Japanese media, politics, and culture seems unusual, given the nation’s traumatic experiences with nationalistic ideologies in the years leading up to the Pacific War, and an explanation for this state of affairs must take into account a wide variety of factors for this development.  The  diverse backgrounds of the formulators of bushidō, the breadth of the ideology’s application from popular media to governmental policy, and the multifarious interpretations and adaptations by those exposed to bushidō ideology in the prewar period mean that an examination of the subject must account for historical, political, social, cultural, technological, and even economic factors.9  Through a multi-faceted approach to bushidō, incorporating a broad spectrum of the  myriad societal and cultural fields in which bushidō discourse took place in the late Meiji and early Taishō periods, this study seeks to understand the development of bushidō in modern Japan, which is central to understanding many of the bushidō theories prevalent in Japan today. The continued resilience of bushidō has, from the first discussions of the concept in Meiji, arguably been primarily attributable to its unique characteristic of having a perceived historical pedigree without concretely verifiable historical roots.  This has endowed bushidō  with a combination of legitimacy and flexibility that few other ideological constructs possess, allowing it to be interpreted and adapted to very different eras and situations. Conversely, 9  The last two are relevant primarily in relation to the implementation of bushido-based “spiritual education” (seishin kyōiku) by the military, especially in the Taishō and early Shōwa periods.  4  when bushidō became too closely identified with a single mindset or ideal, as happened with militarism in the 1930s and 1940s, it was rejected along with its ideological partners.  This  study demonstrates that this same process took place in the years leading up to 1914, when many came to view bushidō as inextricably linked with the Meiji period, and especially the figure of General Nogi Maresuke.  Through this examination, it is possible to gain insight into the  reasons why this ideology was capable of being resurrected in the postwar period, as well as to provide points of departure for future research on the topic.  The etymology and historiographical uses of bushidō In his 1992 Antiquity and Anachronism in Japanese History, Jeffrey Mass pointed out several significant areas of research on Japan that remained relatively untouched. One of these was the study of the “way of the warrior,” one of the most high-profile themes in popular culture both in Japan and overseas.  In almost two decades since Mass pointed out this gap in the field,  while progress has been made, there is still much work to be done.10  Apprehensiveness on the  part of scholars means that much of the field of samurai thought and behavior has been left to popular writers, resulting in the continuation of considerable misconceptions.  Although some  good translation work has also been done in this manner, the selection of texts and passages has been especially problematic, and terms and concepts from different periods are often conflated or altered to create a homogeneity that is not necessarily supported by the source documents.  For  marketing reasons, for example, a variety of terms are uniformly rendered “bushidō,” as this is the best-known term relating to Japanese warrior ethics.11  10  In 1997, Gomi Fumihiko saw considerable progress in research on the historical bushi, but also called for research on modern bushidō, which he felt was still neglected (Gomi Fumihiko, Sasshō to shinkō: bushi wo saguru, 7, 276). 11 For example, Daidōji Yūzan’s Budō shoshinshū has been rendered “Bushido Shoshinshu” in the title of the most popular edition of the text, Thomas Cleary’s Code of the Samurai: A Modern Translation of the Bushido Shoshinshu of Taira Shigesuke. Tokyo: Charles E. Tuttle, 1999.  5  The etymology of the term “bushidō” has been a subject of considerable debate for over a century. In 1912, the renowned Japanologist Basil Hall Chamberlain (1850-1935) argued as follows: “As for Bushido, so modern a thing is it that neither Kaempfer, Siebold, Satow, nor Rein--all men knowing their Japan by heart--ever once allude to it in their voluminous writings.  The cause of their  silence is not far to seek: Bushido was unknown until a decade or two ago! THE VERY WORD APPEARS IN NO DICTIONARY, NATIVE OR FOREIGN, BEFORE THE YEAR 1900. Chivalrous individuals of course existed in Japan, as in all countries at every period; but Bushido, as an institution or a code of rules, has never existed [original emphasis retained].”12  This view is supported by some Japanese scholars, such as Morikawa Tetsurō and Gomi Fumihiko, who agree that bushidō first became a subject of discussion in the Meiji period.13 The textual evidence confirms these claims, and the term “bushidō” does not appear in any texts before the seventeenth century, and only very sporadically after that time.  Even Inoue Tetsujirō  (井上哲次郎, 1855-1904), who dominated prewar bushidō discourse, admitted that it had not been codified before the early Edo period. Nitobe Inazō believed that he himself had invented the term, and was not aware of its use before the completion of his own book in 1899.  It was  only shortly before his death in 1933 that he became aware of earlier occurrences of the term.14 On the other hand, in a recent article on the etymology of “bushidō,” Kasaya Kazuhiko has argued that bushidō was a widely understood term before the middle of the Tokugawa period. “Until the middle of the eighteenth century, bushidō existed in a form that struggled for supremacy with Confucianism…By the end of the Edo Period, the term ‘bushidō’ seems to have lost its status as a commonly used word, although many texts containing it existed…However, if one did not have a chance to see those texts, the actual state of affairs was that even among bushi, bushidō was no longer a usual term.”15  12 13 14 15  Basil Hall Chamberlain, The Invention of a New Religion, 6. Morikawa Tetsurō, Nihon bushidō shi, 3; Gomi Fumihiko, Sasshō to shinkō: bushi wo saguru, 7. Ōta,Yūzō, Taiheiyō no hashi toshite no Nitobe Inazō, 20-21. Kasaya Kazuhiko, “Bushidō gainen no shiteki tenkai,” 271.  6  Kasaya bases his argument on the existence of a handful of texts that mention bushidō, but does not provide compelling evidence that the word was anything more than a sporadically-cited literary term.16  Maruyama Masao similarly argued that bushidō existed in the transitional  period between the Sengoku “learning of one who takes the bow and arrow (yumiya toru mi no narai)” and the Confucian shidō of the Tokugawa period.17 In spite of these latter assertions, the documentary evidence from pre-modern Japan supports the claims made by Chamberlain, Morikawa, and Gomi. The views of most historians familiar with the relevant documents are reflected in Ōtsuki Fumihiko’s Daigenkai (大言海, Great Sea of Words), which cites the Kōyōgunkan (甲陽軍鑑, Martial Records of Kōyō) as the origin of the term.18  This work is a collection of texts recording the tactics of Takeda Shingen,  and is believed to have been compiled in 1656; i.e. well into the Edo period.19  The use of  bushidō in the Kōyōgunkan set the pattern of only occasional use of the term that dominated until the last decade of the nineteenth century.  While bushidō does receive mention in the  Kōyōgunkan, terms such as budō (the martial way), bushi (mononofu) no michi (the way of the warrior), and yumiya no michi (the way of the bow and arrow) are far more common.20 In addition to the Kōyōgunkan, another source that is frequently cited as the origin of bushidō is the large number of kakun, or house codes, that were attributed to the great military families of the late sixteenth century, although many of these documents appear to have been compiled in the middle of the seventeenth century.  These house codes were viewed as a source  16  The main texts listed by Kasaya are Kiyomasa ki, Kōyo gunkan, Hagakure, Budō shoshinshū, Bukō zakki, Kashō ki, and Shōbu ron. 17 Maruyama Masao, Chūsei to hangyaku, 22. 18 Ōtsuki Fumihiko, Daigenkai, 1794. 19 Kasaya Kazuhiko, “Bushidō gainen no shiteki tenkai,” 235. 20 Sagara Tōru ed., Kōyōgunkan, gorinsho, hagakure-shū (Nihon no shisō Vol 9). In this abridged selection of the Kōyōgunkan, bushidō is found only once, on page 83; bushi (mononofu) no michi on page 175; budō appears on pages 83, 94, 172, 173, 175 (4 times), and 176. Saeki Shin’ichi, relying on an indexed version of the entire work, counts 39 mentions of bushidō versus 65 of budō. He does not mention other terms with similar meanings, such as bushi (mononofu) no michi, buhen, or otoko no michi, the latter two of which appear dozens of times in the Nihon no shisō edition (Saeki Shin’ichi, Senjo no seishinshi, 200). Kasaya, in his “Bushidō gainen no shiteki tenkai” counts 30 and 34 occurrences in two different versions, but neglects to quantitatively analyze synonymous terms (p. 236).  7  of bushidō from an early stage in Meiji-era bushidō discourse, and a significant selection of these was published in 1906 under the title Bushidō kakun shū (武士道家訓集, A Collection of Bushidō House Codes).  Interestingly, the term only appears thrice in this 323-page  compendium, as three of the selected house codes include it once each.21  The first of these  three is the well-known kakun attributed to the house of the powerful daimyo Katō Kiyomasa (加 藤清正, 1562-1611), which also uses several other synonyms found frequently in the other house codes, including otoko no michi (the manly way) and the hōkō no michi (the way of service).  The third of the three bushidō-relevant house codes, that of Honda Tadakatsu (本多忠  勝, 1548-1610), similarly makes use of many different terms, such as budō, bushi no seidō (the true way of the warrior), and shidō (the way of the samurai). In addition, bushidō receives little or no mention in postwar scholarship on medieval house codes, including Kakei Kazuhiko’s comprehensive study of the subject.22  The evidence indicates that the association of bushidō  with Japanese military house codes is a product of late Meiji-era interpretations, and the decision to insert the term into the title of the aforementioned 1906 collection must also be considered in this context. The word “bushidō” can be created simply by adding dō (“way”) to the term bushi (“warrior”), and it is surprising that the term does not occur more frequently in documents from the Edo period, when the concept of “way” was commonly used.  Even during this time,  however, most references to bushidō appear to have been inspired by references to earlier texts. In the few texts that do specifically mention bushidō, there are two characteristics that indicate a reliance on older works.  The Hagakure, an early eighteenth-century work which came to be  one of the most significant samurai texts after its eventual publication in the early twentieth century, provides an example of both of these characteristics. The first is a direct mention of an older text that mentions bushidō; for example, the Hagakure makes mention of the Kōyōgunkan, 21 22  Arima Sukemasa and Akiyama Goan eds., Bushidō kakun-shū. “Bushidō” found on pages 138, 141, and 147. Kakei Kazuhiko, Chūsei buke kakun no kenkyū.  8  indicating that Yamamoto may well have discovered the word in this work.23  The second  indicator is the use of a version of the phrase “budō (or bushidō) no ginmi,” i.e. “the investigation of budō (or bushidō),” which is also found in the Hagakure. 24  The first  occurrence of a version of this phrase appears to be in the Budō shinkan, which is dated 1577.25 In this work, the phrase “budō gu goginmi,” or “the investigation of budō implements” is used with regard to the Sengoku daimyō Takeda Shingen (武田信玄, 1521-1573).26  The earliest use  of the exact phrase “bushidō no ginmi” appears to be in a collection of documents associated with Katō Kiyomasa 27 that scholars speculate was largely written in the period from 1658-1661.28  The importance of these two indicators quickly becomes clear when one reads  through the few texts that contain the term bushidō, and the trajectory of their frequency of use throughout the Edo period is also revealing. The pattern of use for bushidō is best examined through study of the definitive Bushidō zensho (武士道全書, Complete Texts on Bushidō), compiled by Inoue Tetsujirō and Saeki Ariyoshi (佐伯有義, 1867-1945) in 1942. This thirteen-volume work brought together the most comprehensive collection of historical documents relating to the samurai class, and many of them have not been published in any other form.  Of 118 pre-Meiji documents comprising  well over 4,000 pages, including hundreds of poems and quotes, the term bushidō appears in thirteen of them.  Among these thirteen, only six use the term more than once, and four use it  with any degree of regularity, including both the Hagakure and Budō shoshinshū. Ten of the thirteen Edo documents containing bushidō were composed after 1700, and seven of these contain one of the two above characteristics indicating that the author relied on an earlier text as his source of the term.  In addition, the frequency of use increases in the later documents,  23  Saiki Kazuma et al. eds., Mikawa monogatari, hagakure (Nihon shisō taikei Vol. 26), 533. Ibid., 237. 25 Saeki Ariyoshi et al. eds., Bushidō zenshō Vol. 1, 358. 26 Ibid., 353. 27 Ibid., 357. 28 Moriyama Tsuneyo, “Katō Kiyomasa denki ‘Zokusen Kiyomasa ki’ no seiritsu to sono tsuika shū no shōkai,” 336. 24  9  contradicting Kasaya and Maruyama’s claims that bushidō was a widespread concept in early Tokugawa that gradually fell into disuse. The evidence from documents dealing with Japanese warrior thought and compiled in collections such as the Bushidō zensho strongly indicates that, during the Edo period, bushidō was a minor literary variant of other, more commonly used terms referring to warrior behavior. An analysis of the spoken language is more difficult, and Kasaya’s assertion that bushidō was a commonly-understood term would ultimately have to be confirmed or refuted on the basis of such an examination.  Large differences between the written and spoken languages in  Tokugawa Japan complicate attempts to make inferences from the former to the latter.29  This is  exacerbated in the case of bushidō by the fact that the texts that do use the term tend to be either the creations of individuals outside of the mainstream, such as the recluse Yamamoto Tsunetomo, or idealized historical documents such as the Kōyōgunkan.30  Instead, useful information in this  regard can be gained from popular literary and dramatic works from the period, whose dialogue can be seen as representative of the parlance of the age. Fortunately, there is a wealth of cultural material that used the samurai class as its focus, making it possible to glean an understanding of the frequency with which the term was used. On the basis of an examination of some of the most popular, samurai-themed works from different periods of the Tokugawa age, bushidō appears to have been even less used as a spoken term than as written one.  To this end, this study closely examined Ihara Saikaku’s Budō  denraiki (1687)31 and Buke giri monogatari (武家義理物語, Tales of Warrior Duty, 1688)32, 29  Terry Eagleton has discussed the difference between “academic” and “media” discourse in the modern United States, and it would seem that this gulf would have been more pronounced in Edo Japan, where scholars often wrote using linguistic structures that were very different from the spoken language. (Terry Eagleton, Figures of Dissent, 35.) 30 The factuality of the Kōyōgunkan is questioned by most scholars as it is believed to have been compiled in the early 17th century (Eiko Ikegami, The Taming of the Samurai: Honorific Individualism and the Making of Modern Japan, 399). For an overview of the debates on the provenance of the work, see Kuroda Hideo, “Kōyōgunkan o meguru kenkyū shi: Kōyōgunkan no shiryō ron.” 31 Asō Isoji and Fuji Akio eds., Taiyaku Saikaku zenshū: budō denraiki. Budō is used on pages 3, 104, 114, 130, 141, and 171; buhen (martial affairs) on pages 194, 195, 224; bushi no michi on pages 144 and 298; bushi no seidō (the true way of the warrior) on page 74.  10  Takeda Izumo’s (the younger; 竹田出雲, 1691-1756) Kanadehon chūshingura (仮名手本忠臣 蔵, A Treasury of Loyal Retainers, 1748)33, Jippensha Ikku’s (十返舎一九, 1765-1831) Tōkaidō hizakurige (東海道膝栗毛, On Shank’s Mare Along the Eastern Sea Road, published in installments between 1802-1822), Takizawa (Kyokutei) Bakin’s (滝沢馬琴, 1767-1848) Nansō satomi hakkenden (南総里見八犬伝, Legends of Eight Dogs)34, and Katsu Kokichi’s (勝小吉, 1802-1850) 1843 Musui dokugen (夢酔独言, Musui’s Story).  These works, covering every  century of the Tokugawa period, are certainly among the most popular samurai-related narratives of the time, indicating that the terminology they contain was widely understood in Japan during the time in which each was published. In addition, the majority of these texts have been described as prime examples of bushidō literature by modern commentators.35  In spite of this,  the term bushidō does not appear in any of these texts. While the complete absence of the term bushidō from some of the most popular samurai-themed works from the Tokugawa period does not preclude the appearance of the term bushidō in other, lesser-known works, it calls into serious question the claim that the word was widely used or even generally understood by Japanese before the Meiji restoration.  32  Asō Isoji and Fuji Akio eds., Taiyaku Saikaku zenshū: buke giri monogatari. Budō is used on pages 15, 53, 148, and 153; buhen on page 50; bushi no michi on pages 53 and 124; bu no michi on pages 3, 34, 54, and 108. 33 The only related term to appear in this text is bushi no michi, which occurs once in the latter half of the work (Takeda Izumo and Namiki Sōsuke, edited by Miyazaki Sanmai, Kanadehon chūshingura, 92). 34 Takizawa Bakin, Nansō satomi hakkenden Volume 1. Not only bushidō, but shidō and budō are all absent from the first 510-page volume of the Hakkenden. Instead, Takizawa frames most of his ethical discussions in Confucian terms relating to filial piety and loyalty, although he still uses the term bushi to refer to his protagonists. 35 Caryl Callahan, who undertook a partial translation of the Budō giri monogatari for publication in Monumenta Nipponica in 1979, writes that “When Neo-Confucian concepts and vocabulary permeated bushidō, the term giri was adopted to express many old concepts and a few new ones. By Saikaku's day, the concept was deeply meshed with bushidō, and fulfilling the requirements of giri had become the highest ideal of the bushi.”(Caryl Callahan, “Tales of Samurai Honor,” 2). In this passage, Callahan appears to look back at Saikaku’s age with the understanding that bushidō was a firmly established concept, reflecting a widely-held view in much of the twentieth century. Similar examples include Miyazaki Sanmai’s (宮崎三昧, 1859-1919) introduction to the Kanadehon chūshingura, which frames the text in the context of bushidō discourse (Takeda Izumo and Namiki Sōsuke, edited by Miyazaki Sanmai, Kanadehon chūshingura), and the Japanese electronic edition of the Encyclopedia Britannica, which succinctly summarizes Takizawa’s Nansō satomi hakkenden, as “Confucian and bushidō-like.” (Britannica kokusai daihyakka jiten: shōkōmoku denshi jisho ban. Tokyo: Britannica Japan, 2004).  11  An additional complication with the term “bushidō” is the confusion between its historical and historiographical usage.  Although it was largely unknown before Meiji, since  that time “bushidō” has come to be commonly used as a broad descriptive term for Japanese warrior behavior, whether it was originally uncodified or labeled with another name. This is especially problematic in the case of translations of historical documents into modern Japanese and other languages, which often render terms such as “budō,” “shidō,” “hōkōnin no michi,” and many others uniformly as “bushidō,” implying a homogeneity and universality that did not actually exist.  This can be seen most clearly in later commentaries and translations of the  Hagakure, both in English and modern Japanese, as it continues to be one of the most popular texts related to samurai thought. Modern commentators who may themselves be convinced of the importance of bushidō often choose to use one term for both “budō” and “bushidō.”36 While this may, in fact, reflect Yamamoto Tsunetomo’s view of the two terms as having the same meaning, it is misleading in the sense that “bushidō” is generally used throughout modern translations in place of the historically far more common “budō.”  This tendency is not found  only in the works of nationalistic prewar bushidō ideologists, however, but also pervades recent scholarly texts.  For example, in Sources of Japanese Tradition, the translation of a section of  the Hagakure includes the following sentence: “The idea that to die without accomplishing your purpose is undignified and meaningless, just dying like a dog, is the pretentious bushidō of the city slickers of Kyoto and Osaka.”37  A glance at the original text, however, reveals that the  36  There is considerable confusion regarding the historical difference between bushidō and budō, two terms that have become well-understood in English. This difficulty stems largely from a shift in the meaning of the budō that occurred in the late nineteenth century. While it had earlier been used as a general term that could refer to any or all aspects of samurai life, including religion, behavior, learning, destiny, and military training, in the Meiji period budō came to be used almost exclusively to refer to individual military training and abilities, which had also been referred to using terms such as bujutsu, bugei, and buhō during the Edo period. It is this meaning, commonly rendered as “martial arts” in English, that budō continues to retain in both languages in the present day. In fact, the entry on budō in the Kokushi daijiten contains nothing more than a note to the reader to “see bugei” (which can be literally translated as “martial arts”). At the same time in late Meiji, the term “bushidō” began to replace “budō” and other words in discourse concerning aspects of the samurai that went beyond practical concerns such as strategy and martial arts (see: Inoue Shun, “The Invention of the Martial Arts: Kanō Jigorō and Kōdōkan Judo,” 163). 37 Barry Steben, “The Way of the Warrior II,” 476.  12  Romanized term in the transcription is not “bushidō” at all: “Moshi, zu ni ataranu toki ha, inuji nado to iu koto ha, Kamigatafū no uchiagari taru BUDŌ naru beshi [my emphasis].”38  Even  Kasaya Kazuhiko, who is currently one of the most prolific and established writers on the subject of bushidō in Japan, transcribes budō as bushidō in an explanation of a brief passage of the Budō shoshinshū.39  These are by no means isolated examples, and less academically rigorous works  are guilty of far more serious transgressions. One such instance is the 1999 Charles E. Tuttle publication of Thomas Cleary’s translation of Daidōji Yūzan’s Budō shoshinshū, which has been given the misleading title Code of the Samurai: A Modern Translation of the Bushido Shoshinshu of Taira Shigesuke in an evident effort to appeal to readers who are more familiar with bushidō than with the term used by the original author.  Another approach taken by some scholars, such as Takahashi Tomio, has been to use the term historiographically while prefacing it with a specifier that limits unsustainable claims to universality.  In his History of Bushidō, Takahashi uses labels such as “Mito bushidō,” “Aizu  bushidō,” and “Satsuma bushidō” to refer to doctrines that are limited to specific temporal and physical spaces.40  Takahashi’s approach presents one possible solution to the dilemma, but the  38  Saeki Ariyoshi et al. eds., Bushidō zenshō Vol. 6, 30. There are several slightly different versions of the Hagakure, such as the Nihon shisō taikei version mentioned above (p. 220), or the Hagakure zenshū published by Gogatsu Shobō in 1978 (p. 17), but all use the term “budō” in the disputed passage. A similar equivocation regarding the same passage is made by Stacey Day in the preface to the Proceedings of the International Symposium on Hagakure. After admonishing previous interpreters of the Hagakure for their lack of meticulousness concerning the use of specific terms, Day proceeds to render both “bushidō” and “budō” as “bushido” in the “improved” translation that he published in collaboration with Koga Hideo (Koga Hideo and Stacey Day eds., Hagakure: Spirit of Bushido, xix-xx. Day repeats the same error on the following page in a translation of Hagakure kiki-gaki 114 (113 in Nihon shisō taikei version, p. 251). This is another sentence containing both terms, and Day simply renders both as “bushido.”). This admonition applies not only to foreign translators, however, and Japanese commentators are equally guilty of misleading transcriptions. In many post-Meiji versions of the Hagakure in Japanese, the above passage is rewritten so that both of the terms in question are rendered uniformly as “bushidō,” with no comment or explanation given. Two examples of this, one each from before and after 1945, are the Hagakure commentaries by Matsunami Jirō (1939) and Morikawa Tetsurō (1975) (Matsunami Jirō, Hagakure bushidō, 19; Morikawa Tetsurō, Hagakure nyūmon, 33). Ōkuma Miyoshi’s modern Japanese transcription goes so far as to use bushidō while budō remains in the original text on the facing page (Yamamoto Tsunetomo; Ōkuma Miyoshi ed., Hagakure: gendaiyaku, 33). 39 Kasaya Kazuhiko, “Bushidō gainen no shiteki tenkai,” 253. 40 Takahashi Tomio, Bushidō no rekishi 3: 84, 110, 128.  13  most effective method of minimizing the confusion between historiographical and historical usages of “bushidō” would be to use either period- or location-specific historical terms, or neutral descriptors such as “warrior morality.”41  In this study, “bushidō” is used exlusively to  refer to the ideology of the same name that developed from mid-Meiji onward, and not as a synonym for any other term or related concept. The reader will soon notice that there are a relatively large number of Romanized and Italicized terms in this work.  Where widely accepted and directly corresponding terms exist, I  use the English term with the Romanized Japanese original at the first mention of the term. Jeffrey Mass has criticized what he sees as John W. Hall’s excessive insistence on using original terms in Japanese historiography, but adherence to the original Japanese terminology is essential for understanding the historical development of bushidō.42  This admonition naturally applies to  Western scholars, but is also directed at Japanese scholars and writers concerned with bushidō, as much of their work is riddled with similar terminological equivocations.  Warrior behavior in Japanese history before 1600 While the available evidence strongly indicates that the term “bushidō” was virtually unknown before 1890, and did not enjoy widespread use until the 20th century, there is a second issue which is more critical to the study of bushidō.  This is the question of the very existence  of a samurai code of ethics under one or several different names, or even no name, before the Meiji period.  It would seem plausible to argue, as many have done, that a widely understood,  centuries-old warrior code of ethics was merely put to paper and given a new moniker by Tokugawa thinkers before being resurrected and redirected by Meiji-era bushidō theorists. However, an examination of source materials and later scholarship relating to samurai morality 41  In a later text, Takahashi argues for the use of the concept “budō tetsugaku (budō philosophy)” to describe periods when bushi were not actually active, and he considers the concept of budō to be primary to bushidō (Takahashi, Bushi no kokoro, nihon no kokoro, 284, 290, 400-401). 42 Jeffrey Mass, Antiquity and Anachronism in Japanese History, 7-8  14  does not reveal the existence of a single, broadly-accepted, bushi-specific ethical system at any point in pre-modern Japanese history. The very identification of warriors as a separate class is problematic before the Edo period, and even after this point regional differences and temporal changes precluded the broad acceptance of a warrior-specific code of conduct, as the character and definition of bushi varied considerably from domain to domain. Beginning in the late Meiji period, some prewar scholars sought the origins of bushidō in the exploits of the legendary first emperors, or even in the earlier Age of the Gods, while more reasoned accounts tended to search for the genesis of a warrior ethic in the Heian or Kamakura period.43  While some bushidō theorists sought to push the origins of their subject as far back  into history as possible in search of legitimacy, there was a broad consensus among scholars that no codified warrior ethic existed before the seventeenth century.  Even Inoue Tetsujirō, who  was bushidō’s most fervent promoter from 1901 until his death in 1944, and who traced the history of the “unique Japanese bushidō spirit” to the mythical Plain of High Heaven, contended that the subject was first codified in the works of Yamaga Sokō in the late seventeenth century.44 Kanno Kakumyō, a more recent proponent of bushidō, also argues that the samurai were too busy fighting in earlier centuries, and only began to concern themselves with ethics in the relatively peaceful Edo period.45  At the same time, most bushidō theorists, including Inoue and  Kanno, have argued that the period of intermittent warfare before 1600 was critical in the formation of bushidō, which is fundamentally a martial ethic. To be sure, texts that are often held up as Edo-period formulations of bushidō, such as the Hagakure and Budō shoshinshū, are strongly colored by a nostalgic view of an earlier age when samurai still frequently clashed on the battlefield.  43 44 45  Saeki Ariyoshi ed., Bushidō Hōten, 10. Inoue Tetsujirō, Bushidō, 41. Kanno Kakumyō, Bushidō no gyakushū, 20-21.  15  It is telling that historians of Japan’s turbulent pre-Tokugawa centuries are almost all silent on the subject of bushidō, and do not frame their studies in the context of warrior-specific moral norms.  When medieval historians do address the subject, the result tends to be critical of  bushidō. Karl Friday, for example, has addressed the issues of loyalty and a stoic attitude towards death, which are among the characteristics most frequently attributed to the samurai. With regard to the former, he writes “however central the willingness to die might have been to twentieth-century notions of bushidō, it takes a considerable leap of faith to connect this sort of philosophy with the actual behavior of the medieval samurai.”  On the subject of loyalty, Friday  points out that “the truth is that selfless displays of loyalty by warriors are conspicuous by their absence.” Friday describes the situation between the rise of the warriors and the seventeenth century as follows: “…the ties between master and retainer were contractual, based on mutual interest and advantage, and were heavily conditioned by the demands of self-interest. Medieval warriors remained loyal to their lords only so long as it benefited them to do so; they could and did readily switch allegiances when the situation warranted it.  In fact, there are very few important battles in Japanese history in  which the defection—often in the middle of the fighting—of one or more of the major players was not a factor.”46  This assessment of medieval loyalty is echoed by Cameron Hurst: “In fact, one of the most troubling problems of the premodern era is the apparent discrepancy between the numerous house laws and codes exhorting the samurai to practice loyalty and the all-too-common incidents of disloyalty which racked medieval Japanese warrior life.  It would not  be an exaggeration to say that most crucial battles in medieval Japan were decided by the defection—that is, the disloyalty—of one or more of the major vassals of the losing general.”47  With regard to the supposed willingness of medieval warriors to offer up their lives, Hurst criticizes the belief that seppuku was a widespread custom, and was instead limited to hopeless 46 47  Karl F. Friday, “Bushidō or Bull?” 341-342. G. Cameron III Hurst, “Death, Honor, and Loyality: The Bushidō Ideal,” 517.  16  situations in which a defeated warrior was certain to be subjected to torture, a common practice at the time.48 The views put forth by Japanese historians are similar to those expressed above.  Hurst  cites Sakaiya Ta’ichi’s comparison of the medieval samurai sense of loyalty with that of modern professional baseball players.  According to Sakaiya, the samurai fought for a cause as long as  they were allied to a certain “team,” but fully expected to move teams or be “traded” during their careers.49  This contractual view of medieval loyalty is borne out by the surviving documents  from the time.50  Warriors carefully recorded the services rendered to their lords, and expected  payment commensurate to their sacrifices.  As Friday has summarized the situation, “Warrior  leaders could count on the services of their followers only to the extent that they were able to offer suitably attractive compensation – or, conversely, to impose suitably daunting sanctions for refusal.”51  Loyalty that was not dependent upon compensation existed in some cases of bonds  of kinship, but these were generally limited to the nuclear family, and even then bonds between siblings were often weak.52 The makeup of the military before 1600 is another factor to consider in discussions of warrior behavior and ethics.  In much the same way as medieval European knighthood was  inseparable from the structure of feudal society, the changes in Japan’s military technology and specialization were concomitant with shifts in the social hierarchies and lifestyles of the Japanese people.  In contrast with China, Japan’s early military evolved from corvee peasant armies to a  much smaller professional class of mounted archers.53  As a result, the sizes of armies shrunk  dramatically, and some of the most famous battles in the 12th, 13th, and 14th centuries were contested by only a few hundred men. For example, at Minatogawa in 1336, during one of the 48  Ibid., 520. Ibid., 518. 50 For an analysis of 14th-century warrior documents, see , State of War: The Violent Order of Fourteenth-Century Japan. 51 Karl F. Friday, Samurai, Warfare, and the State in Early Medieval Japan, 59. 52 Ibid., 57. 53 Karl F. Friday, Hired Swords: The Rise of Private Warrior Power in Early Japan, 39-40. 49  17  largest battles of the 14th century, and also one of the most significant battles in all of Japanese history, only an estimated seven hundred men perished together with the loyalist Kusunoki Masashige.54  Generally, much like their contemporaries in Western Europe, Japanese warriors  during this period tended to emphasize heroic individual actions and personal glory. The gradual breakdown of central authority in the fifteenth century culminated in the form of the Ōnin War of 1467-1477, which resulted in major changes to the military order.  The  inability of super-regional governmental institutions to guarantee any semblance of protection to the lives and property of the citizenry resulted in a diversification and expansion of military entities.  Temples, most of which had always had considerable fighting forces, increased these  to the point that they represented some of the largest armies in the land. Peasants increasingly banded together in religious or secular ikki for protection.  In some provinces, most notably Iga  and Kōga, villages under the headship of jizamurai organized their own unique defenses against outside attackers, including composing constitutions and constructing extensive fortifications.55 Warfare ceased to be the exclusive domain of the professionals, and even bushi families found themselves incorporating irregular units consisting of conscripted peasants into their military forces. As a result, in the mid-to-late sixteenth century, when domainal consolidation increased armies to an unprecedented size, they included a large proportion of non-professionals and had a far higher ratio of pedestrians to equestrians than in previous centuries. The introduction of firearms in the middle of the sixteenth century increased the effectiveness of conscripted troops as they did not necessitate extensive training to use.  Although the arquebus did require a  certain amount of military drill, it did not take the same level of dedicated training required by equestrian archery, and those who used guns as their main weapon were not accorded the elite  54 55  Thomas Conlan, State of War: The Violent Order of Fourteenth-Century Japan, 57. Pierre Francois Souyri, The World Turned Upside Down: Medieval Japanese Society, 190-191.  18  status of the mounted archers of earlier eras.56  The sixteenth-century trend toward larger armies  can be seen in the troop numbers present at the Battle of Sekigahara in 1600, where the competing factions were able to muster an estimated total of roughly 150,000 men.57 Changes in the size and makeup of Japanese armies in the centuries before 1600 naturally affected the relationships within and between feudal houses and other social entities.  When  fighting forces were relatively small-scale, such as during the late Heian, Kamakura, and early Muromachi periods, horizontal relationships were more representative than in later times.  Most  bushi still had direct connections to, and drew their power from, their landholdings, and were able to use this authority as leverage when dealing with other warriors.  Therefore, simple  demands or orders to arrive and fight for an allied house were unusual, and “loyalty” required payment.  Reciprocity was expected at every stage of the process, and merely arriving at the  camp of the leader who had requested one’s presence was duly rewarded.  According to  Thomas Conlan, a fundamental error in the historiography of Japan’s medieval military has been the equivocation of two separate terms from different eras: both chūsetsu and chūgi are often understood to simply mean “loyalty,” with insufficient discrimination between the times and situations in which the two terms were used.58  This misunderstanding results in some puzzling  translations of medieval petitions, which seem to request compensation for different types of loyalty, including “the loyalty (忠節, chūsetsu) of dismembering an enemy” and “the loyalty (忠 節, chūsetsu) of treason.”59  The second of these seems particularly irreconcilable with most  understandings of loyalty, but is made comprehensible when chūsetsu is translated not as “loyalty,” but as “service.” Rather than an abstract concept of loyalty of the type found in  56  Arquebusiers, although more important than the bowmen or pikemen they fought alongside, also had their own units and were ultimately of relatively low rank in the military hierarchy. (Kasaya Kazuhiko, The Origin and Development of Japanese-Style Organization, 50-51) 57 John Whitney Hall, “The Bakuhan System,” 163. 58 Thomas Conlan, State of War: The Violent Order of Fourteenth-Century Japan, 141-164. 59 Ibid, 142-143.  19  Confucian-influenced discourse after 1600, loyalty during this turbulent period was very pragmatic and would shift if not sufficiently rewarded.60  Pre-modern writings on bushidō Before 1945, the historiography of Japan’s medieval period was heavily politicized, especially when it concerned the imperial house. The great controversy that erupted in 1911 concerning the proper interpretation of the period of the Northern and Southern courts, and the legitimacy of the two, is a prime example of this. 61  Scholars were not prohibited from  conducting their own research on these subjects, but public and governmental pressure made it difficult to publish views on sensitive subjects that did not conform to the orthodox interpretations, which were marked by an emphasis on bushidō values such as imperial loyalism, bravery, and self-sacrifice.  Medieval war tales were selectively edited and incorporated into  new dramatic forms in order to bring forth these virtues, and many elements from these prewar interpretations are still widely disseminated. Among historians, revision of the prewar versions of medieval history began soon after 1945.  One example of this is the work of Takayanagi Mitsutoshi (高柳光寿, 1892-1969),  founder of the Japan Historical Society, who struck back at the idealized view of medieval Japan in a 1960 essay on bushidō.  According to Takayanagi, the accounts of Sengoku warriors were  largely products of the Edo period and more a reflection of seventeenth-century society than actual battlefield conduct. 62  Takayanagi further rejected the notion of the existence of  unilateral loyalty to a lord, much less an emperor, before the Edo period, and argued that most samurai would have considered their own lives to be considerably more important than the lives 60  This situation can be seen clearly in the Taiheiki account of the abortive Kemmu Restoration attempted by the emperor Go-Daigo in 1333, which failed largely due to the sovereign’s refusal to monetarily compensate the warriors who fought for his cause, instead expecting them to offer their services purely out of a sense of duty to the imperial house. (Andrew Edmund Goble, Kenmu: Go-Daigo’s Revolution, 79) 61 John S. Brownlee, Japanese Historians and the National Myths, 1600-1945, 118-130. 62 Takayanagi Mitsutoshi, Bushidō: Nihon bunka kenkyū (Volume 8), 5.  20  of their superiors.63  Naramoto Tetsuya presented similar arguments in his 1971 Bushidō no  keifu (Genealogy of Bushidō), citing examples of warlords asking brave fighters to join their side from the enemy ranks as a common practice.64  Naramoto further discussed the repeated looting  of Kyoto as evidence of a lack of ethics, and the great importance warriors placed on appearance as the antithesis of the popular image of the austere and frugal samurai. 65  More recent  expositions of the period are Gomi Fumihiko’s 1997 Sesshō to shinkō: bushi wo saguru (Killing and Faith: Searching for Warriors) and Saeki Shin’ichi’s 2004 Senjō no seishinshi (History of Battlefield Mentalities), which also reject the idea that a widely-understood warrior ethic existed before 1600.  While these works focus on medieval history, Saeki deemed it necessary to  include a discussion of bushidō in modern Japan, and one of the goals of his book is to dispel some of the prevalent bushidō-related myths concerning medieval Japanese warriors and society.66 Aside from a handful of idealized accounts, such as that of Kusonoki Masashige and Godaigo, pre-1600 samurai behavior was not generally deemed suitable by interpreters of bushidō active between Meiji and 1945.  The mercenary nature of bushi during Sengoku was at  odds with the emperor-centered nationalistic core of the “orthodox” modern bushidō that rose to prominence after the Russo-Japanese War.67  In addition, the Edo-period texts that showed the  greatest nostalgia for pre-Tokugawa conditions were carefully selected, condensed, and edited to purge them of those elements which ran counter to the national project in the early twentieth century.  For example, a pocket edition of the Hagakure was widely distributed by the army  beginning in the 1940s, but in a highly abridged form compiled by the philosopher Watsuji Tetsurō (和辻哲郎, 1889-1960).68 63 64 65 66 67 68  It is perhaps most telling, however, that while students of  Ibid., 9-13. Naramoto Tetsuya, Bushidō no keifu, 30-31. Ibid., 33-39. Saeki Shin’ichi, Senjō no seishinshi. This development is discussed in greater depth in chapter three of this study. Unoda Shōya, “Bushidō ron no seiritsu: seiyō to tōyō no aida,” 43.  21  bushidō would sometimes look to medieval history to seek evidence for their theories, historians of medieval history were, and are, seldom drawn to bushidō as an exegetical tool.  Changes in military technology and tactics in Japan between the early thirteenth and the end of the sixteenth centuries contributed greatly to the fluid nature of warrior society during this period.  In turn, the diversity of martial culture defies retrospective attempts to superimpose  codes of warrior morality on anything broader than local groups for more than brief periods. With regard to warrior morality, the importance of Sengoku lies primarily in providing samurai in the Edo period with historical reference points that, in an idealized form, gave considerable legitimacy for their domination of the political order.  In addition, specific incidents or  individuals from Sengoku could be selected by Tokugawa theorists to argue for or against certain points.  One example of this is Yamamoto Tsunetomo’s reliance on anecdotes concerning the  lords of Nabeshima domain, which make up a considerable portion of the Hagakure.  In terms  of modern bushidō, however, the Edo period was by far the greatest source of historical references for thinkers who sought inspiration in earlier periods.  This can be seen in the  Bushidō zensho, with the vast majority of documents found in its 4,000 pages having been composed between 1600 and 1868. While many of them were cited by bushidō theorists in the twentieth century, the large number of writings on samurai behavior and ethics that were composed during the Edo period do not represent a consensus on any specific issue or course of behavior. In this context, the most important theoretical shift from Sengoku to Edo was also a social one.  During the earlier  period, there can be no doubt that a certain martial elite perceived themselves to be primarily warriors, but the distinction between warrior and civilian was less clear among lower-ranking or part-time fighters.69  69  The self-perception of the bushi is significant because, as Kanno Kakumyō  Gomi Fumihiko has discussed the variety of warriors in late Heian and Kamakura, and the difficulties in  22  has summarized warrior ethics, “Ultimately, it can be said that bushidō is the awareness of each individual bushi that ‘I am a bushi.’”70  Kanno presents a compelling argument regarding the  core of what could be called a unique form of samurai thought.  On a theoretical level, that  which set Japanese warriors apart from other social groups, especially before they were given separate legal status by Toyotomi Hideyoshi and Tokugawa Ieyasu, was nothing other than the awareness of being different from non-bushi in some way. During the Edo period, much of the literature that was later integrated into the bushidō canon, such as the works of Yamaga Sokō, was concerned with elucidating the differences between bushi and the non-samurai classes.  While bushi may have themselves been aware of  their bushi status by virtue of their martial duties distinct from those of courtiers, monks, or other non-combatants, they would have been unable to point to any consistent, widely accepted, or unique ethical or moral norm that set their profession apart from the rest of the population. The situation changed somewhat from the Azuchi-Momoyama period (1568-1600) onward, when bushi were at least able to rely on a vague, supposedly innate and usually inherited “warrior-ness,” which was given legal support by successive governments.  As  Douglas Howland and others have argued, it was only at the end of the sixteenth century that the concept of mibun (social status) became important in Japan as a representation of “a conservative wish to reduce social fluidity and to fix social status.”71  At the same time that the social classes  were by and large being fixed by legislation in the late sixteenth and early seventeenth centuries, however, the bushi were quickly becoming removed from their role as active warriors, thereby losing the practical distinction that ostensibly set them apart from other classes.  This created an  awkward situation in which there was little or no application of the martial skills that formed the basis for samurai domination of the political sphere.  For this reason, although Confucian  differentiating between groups of warriors, pirates, and bandits (Gomi, Sasshō to shinkō, 140, 256). 70 Kanno Kakumyō, Bushidō no gyakushū, 225. 71 Douglas R. Howland, “Samurai Status, Class, and Bureaucracy: A Historiographical Essay,” 355.  23  models were applied by Yamaga, Ogyū Sorai (荻生徂徠, 1666-1728), and others to provide a theoretical justification for samurai rule, the parallels drawn between samurai and Chinese gentleman-scholars were not entirely satisfactory, as contemporary Japanese scholars and foreign observers realized full well.72 In addition, changing economic conditions meant that class distinctions were often at odds with social status, a realization that has led to a fundamental rethink of the applicability of the concepts of “class” and “status” in some recent scholarship. The situation is further complicated by regional differences that meant that certain groups were considered samurai in some domains but not in others.73  This can partially account for the great  discrepancies in the percentage of the population that was considered to be samurai in different domains.74  As a result of regional and temporal variations in the character of the warrior class  over the Tokugawa period, which trended heavily towards bureaucratization, a perceived need for definition and legitimization of the role of the bushi in an age of peace became pronounced. It could be argued that the large number of surviving writings that sought to define the significance of the warrior class indicate that the bushi were highly aware of their unique status and desired to understand its nature. Conversely, these same texts could be interpreted as evidence that samurai found their social status increasingly challenged by economically powerful commoners, some of whom were also purchasing or receiving samurai privileges such as the right to wear swords.75  These practices increased greatly as the Edo period drew to a close.  In  this environment, samurai would have felt considerable pressure to identify characteristics that made them different from, and superior to, the other classes. The sense that their position was under threat may also explain the vitriol directed towards commoners in the writings of some  72  Ibid., 356. Ibid., 361-362, 374. 74 According to Sekiyama Naotarō’s analysis of the period 1870-3, depending on the domain, the percentage of samurai ranged from 3.92% to 26.54%, with a national average of 6.40%. (Sekiyama Naotarō. Kinsei nihon no jinkō kōzō, 307-314) 75 For example, the representatives of the Kaitokudō school were granted permission to wear swords when meeting with government officials. Najita Tetsuo, Visions of Virtue in Tokugawa Japan: The Kaitokudō Merchant Academy of Ōsaka, 74. 73  24  samurai commentators, such as Yamamoto Tsunetomo.  On the other hand, especially towards  the end of the Edo period, other commentators of both samurai and non-samurai extraction increasingly rejected the notion that there were fundamental differences between the classes. Depending on the specific domain and period, the variety of stratifications within the bushi and commoner ranks led to situations in which the differences within the classes were often greater than they were between certain members of either, resulting in a degree of overlap, especially in terms of lifestyles and economic status.76 Writers on warrior ethics have often overlooked the ambiguities in the historical situation when considering class status, instead focusing on idealized notions of what a samurai should represent.  In this context, one of the best-known texts on the subject is Yamaga Sokō’s  argument justifying the exalted status of the warrior class: “The tasks of a samurai are to reflect on his person, to find a lord and do his best in service, to interact with his companions in a trustworthy and warm manner, and to be mindful of his position while making duty his focus.  In addition, he will not be able to prevent involvement in parent-child, sibling, and  spousal relationships.  Without these, there could be no proper human morality among all the other  people under Heaven, but the tasks of farmers, artisans, and merchants do not allow free time, so they are not always able to follow them and fulfill the Way.  A samurai puts aside the tasks of the farmers,  artisans and merchants, and the Way is his exclusive duty.  In addition, if ever a person who is  improper with regard to human morality appears among the three common classes, the samurai quickly punishes them, thus ensuring correct Heavenly morality on Earth.  It should not be that a samurai  knows the virtues of letteredness and martiality, but does not use them.  Therefore, formally a samurai  will prepare for use of swords, lances, bows, and horses, while inwardly he will endeavor in the ways of lord-vassal, friend-friend, parent-child, brother-brother, and husband-wife relations. the way of letteredness, while outwardly he is martially prepared.  In his mind he has  The three common classes make  76  For example, in much of the country, hatamoto ranks were hereditary, whereas ashigaru were not. However, there were also ashigaru who were able to pass their positions to their descendants, further blurring the lines. Henry D. Smith relies on the differences within the ranks to explain a serious discrepancy in accounts of the Akō incident of 1703. In this incident, 47 masterless samurai carried out a famous attack, but only 46 of them surrendered to the authorities and were condemned to seppuku in the aftermath. According to Smith, the 47th and lowest-ranked rōnin, Terasaka Kichiemon (寺坂吉右衛門, 1665-1747) was dismissed by the group immediately following the attack as they did not want his low ashigaru status to reflect on the rest of them and cause difficulties or embarrassment. The government responded by simply striking his name from the list of accused. (Henry D. II Smith, “The Trouble with Terasaka,” 5, 38-41)  25  him their teacher and honor him, and in accordance with his teachings they come to know what is essential and what is insignificant. ...Therefore, it can be said that the essence of the samurai is in understanding his task and function.”77  According to Yamaga’s reasoning, one of the major differences between samurai and commoners is that the former have more time to focus on the nature of ethical behavior and can therefore serve a role similar to that of idealized Confucian gentlemen in terms of their position as moral role models for the rest of society. There are two major problems with this view.  In  addition to unemployment or low stipends rendering many bushi unable to make an idealistic “Way” their “exclusive duty,” the specific content of the “Way” outlined by Yamaga was not sufficiently clear or widely accepted that it would serve as a moral guide in real situations.  The  latter reason meant that it was possible for Yamaga to be both lauded and criticized as the teacher of the Akō rōnin, although there is no evidence that he would have supported their actions, let alone acted as their teacher. In comparison, Yamamoto Tsunetomo’s Hagakure provides more specific behavioral guidelines along with a class awareness that is more pronounced than Yamaga’s. According to Yamamoto, the disparity between the classes was also due to the inferiority of non-samurai, but this was due to the innate nature of the individuals in the classes rather than a product of the social structure.78  Whereas Yamaga believed that the moral superiority of the samurai came  from the personal cultivation made possible by their professional situation, Yamamoto felt that their martial nature and readiness to serve gave bushi the right to have the power of life and death over commoners.  Legislation in the form of rudeness-killing laws (burei-uchi) did, in  fact, permit samurai to kill commoners for perceived slights, but the obvious social disorder that this practice would cause meant that it was only seldom applied.79  There was no love lost on  77  Yamaga Sokō. Yamaga Sokō (Nihon shisō taikei Vol. 32), 32-3. This view can also be found in the Kōyōgunkan, which clearly stated that it was not possible for commoners to be like bushi. (Sagara Tōru ed., Kōyōgunkan, gorinsho, hagakure-shū (Nihon no shisō Vol 9), 83) 79 Ikegami Eiko, The Taming of the Samurai, 244-245. 78  26  the other side, either, and the disdain most commoners had for the samurai has been described as “legendary.”80  By the mid-nineteenth century, however, increasing social mobility had further  blurred the distinctions among warriors and between warriors and commoners, and even many influential bushi questioned the innate supremacy of their class.81 Ultimately, the Japanese bushi were a clearly defined class for only the last 250 years of their existence.  During this time, they were politically specified as such and, on the whole, did  not differ significantly from the general populace in terms of ability, religion, geography, or any other readily identifiable characteristic.  Equally, aside from their ostensibly elevated status in  the social order, and pronounced consciousness of the same, there were no other dominant factors that would have tied bushi together.  Since their status was primarily a political and  professional distinction, it is natural that their religious, behavioral, and ethical views were as varied as those of the population at large, and far more likely to be determined by influences other than their status as samurai.  As a result of this diversity, it is possible to select certain  examples of warrior writings and behavior to make a case for almost any interpretation of the “nature” of bushi.  Such discussions tend to be a reflection of the times and situation of their  authors, rather than an accurate depiction of any greater “way of the samurai.”  Many samurai  were naturally aware of their elevated status in the social order, even if they were surpassed financially by wealthy commoners.  However, this consciousness of belonging to an elite varied  greatly depending on time, location, and the specific situation of the individual bushi, and for many of them the differences within their class seemed greater than those between the classes.  80  Leslie Pincus, Authenticating Culture in Japan, 130-132. Other examples include Andō Shōeki (安藤昌益, 1703-1762), who derided the samurai as parasites on society and the National Learning scholar Kamo no Mabuchi (賀茂真淵, 1697-1769) put forth the oft-cited social criticism that the more people one killed, the higher one’s rank, inferring that the shogun was the biggest murderer in the land. (Colin Holmes and A.H. Ion, “Bushido and the Samurai,” 310; Kanno Kakumyō, Bushidō no gyakushū, 39-40). 81 Yokoi Shōnan, for example, argued that the difficult lives of the peasantry made them surpass the warrior class in endurance and mettle, and with weapons and a bit of training a peasant force would be able to handily rout a comparable army of samurai. This view foreshadowed events almost two decades lter, when the new imperial army defeated Saigō Takamori’s former samurai forces in the Satsuma Rebellion of 1877. (Yokoi Shōnan, “Kokuze sanron,” 463)  27  In addition, insofar as class consciousness can be argued to have been widely shared, it did not serve as the basis for a uniform accepted ethic, nor was it possible to easily integrate it into nationalistic modern bushidō ideologies that could serve a supposedly classless society.  Overview of the existing literature on modern bushidō The literature on bushidō, especially in Japanese, is dauntingly vast, reflecting the prominent role of the concept throughout the last century.  Most of these writings deal with  pre-modern Japanese history, and there are very few studies of bushidō in late Meiji, which was the most important period in the development of the concept as it is commonly used today. There are no significant monograph-length studies of bushidō in English, let alone any studies specifically dealing with its modern development.  With regard to specific treatments of  bushidō in English, Cameron Hurst and Karl Friday have given good summaries of pre-Meiji bushidō and its problems, and have discussed its uses in the Pacific War, but do not discuss its development between these periods.82 including many dedicated treatments.  In Japanese, there are far more works on bushidō, Some more recent works are those by Kanno Kakumyō  and Kasaya Kazuhiko, focusing on pre-Meiji thought and behavior.  Kanno’s Bushidō no  gyakushū is one recent example of a work on bushidō that also includes a brief overview of the uses of bushidō in Meiji, but Kanno makes it clear that he does not consider this period to be representative of “true bushidō.” Other works that take a similar approach and make a brief mention of modern developments include Fuji Naotomo’s Nihon no bushidō (Japan’s Bushidō), Morikawa Tetsurō’s Nihon bushidō shi (History of Japanese Bushidō), Saeki Shin’ichi’s Senjō no seishinshi (Spiritual History of the Battlefield), Tawaragi Kōtarō’s Shin bushidō ron (New Bushidō Theory)), and Ozawa Tomio’s Rekishi toshite no bushidō (Bushidō as History). Alexander Bennett’s 2009 Bushi no etosu to so no ayumi (The Warrior Ethos and its 82  G. Cameron Hurst III, “Death, Honor, and Loyality: The Bushidō Ideal,” and Karl F. Friday, “Bushidō or Bull?”  28  Progression) focuses on martial arts in an attempt to trace warrior ethics throughout samurai history, and the seventh chapter of this work discusses “Meiji bushidō” with a dual emphasis on government policies and the role of Nitobe Inazō as a popularizer of the concept. Takahashi Tomio’s three-volume Bushidō no rekishi (History of Bushidō) includes broad discussions of several individuals and aspects of modern bushidō, including Yamaoka Tesshū, Fukuzawa Yukichi, Nitobe Inazō, and Christian bushidō in general, but their influence with relation to greater discourse is not always made clear. There have been relatively few studies that specifically deal with the development of bushidō in modern Japan, and these are generally article-length.  Essays published on the  subject in Japanese include Suzuki Kōshi’s “Meiji ki nihon ni okeru bushidō no sōshutsu” (“The Creation of Bushidō in Meiji Japan”), Iida Kanae’s “Fukuzawa Yukichi to bushidō” (“Fukuzawa Yukichi and Bushidō”), Unoda Shōya’s “Bushidō ron no seiritsu: seiyō to tōyō no aida” (“The Birth of Bushidō Theory: Between East and West”), and my “Hakken sareta dentō toshite no bushidō: Nihon no dokuji no shisō nano ka” (“Bushidō as Invented Tradition: A Uniquely Japanese Ideology?”). Given the scope of the subject, however, these articles have only been able to either provide a general overview or deal with a very specific aspect of modern bushidō. Bushidō is also discussed peripherally in monographs on aspects of modern Japanese history, including works on Japanese cultural nationalism (Minami Hiroshi, Nihonjin ron: Meiji kara kyō made (Japaneseness Theory: From Meiji to the Present)), the development of the concept of Yamato spirit (Saitō Shōji, Yamato damashii no bunkashi (A Cultural History of the Yamato Spirit)), Wang Yangming’s philosophy (Kojima Tsuyoshi, Kindai Nihon no yōmeigaku (The Wang Yangming School in Modern Japan)), and studies of historical figures such as General Nogi Maresuke (Yamamuro Kentoku, Gunshin: kindai Nihon ga unda “eiyū” tachi no kiseki (War Deities: Tracking the “Heroes” Borne of Modern Japan)). The few dedicated books on modern bushidō, such as Matsumae’s study of Christian and Western writers, focus on 29  specific aspects or theorists.  A focus on Christian bushidō theorists is a common theme in  English-language studies of modern bushidō. Ota Yuzo, A. Hamish Ion, Cyril Powles, Simon Edwards, and others have discussed bushidō in studies focused on the spread of the concept abroad and the role of Nitobe Inazō and Christian writers, with Ota also providing a brief overview of late-Meiji bushidō discourse in general.  This study acknowledges and examines  the important role Christians played in bushidō discourse, but is also careful not to overemphasize that role.  Nitobe’s writings on bushidō, for example, were not taken seriously  by most Japanese scholars when his Bushido: The Soul of Japan was published in 1900, and the Japanese edition was not published until 1908. John Tucker and Carol Gluck have provided excellent discussions of certain aspects of modern bushidō in the context of other studies, but have not focused on the subject. To date, there is no monograph-length study of bushidō in English, nor is there a comprehensive study of the modern development of bushidō in any language.  In addition, there  seems to be almost no reference to the development of bushidō between 1880 and 1900, an important period in which the modern concept first developed.  These are major gaps in the  scholarship that need to be filled, for the vast majority of works written on bushidō in the last four decades are unconsciously relying on interpretations and theories that were posited during the late Meiji period. There is a strong tendency for writers on bushidō to unwittingly examine samurai thought and behavior before 1868 through interpretive lenses ground in late Meiji.  The goals of this study Even some of the staunchest supporters of bushidō ideology acknowledge that the term “bushidō” did not gain great currency in Japan until the late Meiji period, when the nationalistic climate following the Sino-Japanese War gave nativist concepts such as bushidō and Yamato damashii (Yamato spirit) an unprecedented degree of exposure.  The work of Nitobe Inazō is 30  mentioned most often in this context, and he is generally portrayed as at least an important popularizer of the concept, even if his depth of knowledge has been called into question. Nitobe’s Bushido: The Soul of Japan appeared in 1900, the year pinpointed by Basil Hall Chamberlain as a watershed in bushidō discourse.  This view is supported by the 1930 military  spiritual education text Virtues of Military Men, which was one of the first and most important works to revive bushidō discourse after it fell out of favor during the Taishō period. According to this text, following victory in the Sino-Japanese War, the world finally looked at Japan and began to research bushidō.  In addition, the authors of the Virtues continued, it was at this  juncture that many Japanese, including Nitobe, noticed bushidō for the first time, and the subsequent military success of the Russo-Japanese War then inspired a large number of novels and other popular works on bushidō.83 These trends are reflected in the Digital Archives of the Meiji Era held at the National Diet Library in Tokyo.  These archives, comprising over 127,000 works published between  1868 and 1912, provide digital images of all of these texts, and include fully searchable bibliographic records and, in many cases, tables of contents.84  A search for “bushidō” reveals  202 works from the period that include the term in the title or table of contents, indicating that bushidō would be a central theme in the text. Only four works in the database, including a translated French novel titled Bushidō, mention the term before 1898. 85  The number of  publications increases from a total of three in 1899 and 1900 to seven in 1901, six in 1902, and dozens per year from 1903 onward.  On the basis of this evidence, the arrival of bushidō in  Japan’s mainstream culture can be dated to the early 1900s, and it is likely that by the end of the  83  Kyōiku sōkan bu, ed., Bujin no tokusō Vol. 1, 26. Search conducted February 2009. There are several articles and other works on bushidō from this period that are not listed in any searchable databases, but the raw figures taken by the records of the National Diet Library are an accurate reflection of general trends. 85 Kuroiwa Ruikō trans., Bushidō ichimei himitsubukuro. This work is a translation of the 1880 mystery novel Les Cachettes de Marie-Rose by the prolific French writer Fortune du Boisgobey. Kuroiwa (1862-1920) was an newspaper essayist and writer who attracted a great deal of notice for his translation of mystery novels into Japanese. 84  31  Russo-Japanese War in 1905, the term would have been familiar to most Japanese, although their understanding of the concept was vague. The broad views of bushidō outlined by Chamberlain and the authors of the Virtues left many questions unanswered.  As this study will demonstrate, the development of modern  bushidō in fact began in the late 1880s, and the foundations for its rapid later development were laid before war broke out in 1894. Most of these early works were forgotten in the twentieth century as bushidō discourse changed, which is why this period has not been examined in great depth before.  These texts are, however, key to understanding the trajectory of later bushidō,  and especially the resiliency of the concept as Japanese society underwent great social and cultural change in the century that followed.  The thinkers who initiated modern bushidō  discourse came from diverse backgrounds and had very different worldviews, but they also had similar perspectives on Japan’s place in the world order relative to China and the West.  This  study will show that modern bushidō developed organically from a cautious nationalism that had not entirely freed itself from a pronounced earlier lack of confidence vis-à-vis the West, and was simultaneously seeking increasing cultural and philosophical independence from China.  The  cautious perspective of some of the earliest bushidō theorists was a significant factor in the dismissal of their works in the more confident environment after 1895, but their influence can be seen in the thought of later writers on the subject. Several scholars have pointed out the disconnect between historical samurai and modern bushidō, and this study agrees that the development of bushidō in Meiji was less a continuation of earlier thought than a response to cultural and geopolitical changes that were occurring at the time.  It must also be pointed out that a bushidō discourse did exist in the waning years of the  Edo period, and was marked by the involvement of the influential activist Yoshida Shōin and several other prominent figures.  One of these, the noted swordsman and Bakumatsu figure  Yamaoka Tesshū, has been credited with a series of lectures on bushidō delivered shortly before 32  his death in 1888.  Takahashi Tomio argues that Yamaoka was one of the few individuals to  bring “feudal bushidō” into the modern age unchanged and undiluted. 86  These thinkers  received varying degrees of exposure in modern bushidō discourse, especially after 1900, and their close temporal proximity to the modern theorists means that they must be placed under consideration as possible bridges between Edo and Meiji bushidō.  For these reasons, the first  chapter of this study is an examination of Bakumatsu bushidō discourse, and the influence that it had on its Meiji counterpart. It will show that this influence, although significant in some ways, occurred after bushidō discourse had already become established around 1900, and that Bakumatsu bushidō theorists did not have a formative influence on the first modern exponents of the subject. Chapter two of this study concerns the Meiji origins of modern bushidō as discussed above.  Specifically, it examines the writings of Ozaki Yukio, Fukuzawa Yukichi, Suzuki  Chikara, and Uemura Masahisa, all of whom were active before the Sino-Japanese War.  Their  contributions towards the development of bushidō were strongly influenced by three major trends in Japanese thought at the time.  The first of these was the maturation of Japan’s  relationship with the West, a process marked by a reevaluation of the idealistic adoration or rejection that defined the attitudes towards the West held by many Japanese thinkers in early Meiji.  The second factor was a change in Japan’s views of China, which became increasingly  negative in the years leading up to the Sino-Japanese War.  The third factor that influenced the  first generation of modern bushidō theorists was a raised interest in their nation’s culture. Whereas Japanese in the 1880s not infrequently claimed to be embarrassed by their culture in front of foreigners, by the early 1890s interest and pride in their own heritage was growing rapidly.  The relationship between these three factors was continually evolving, and they  influenced the individual bushidō theorists to varying degrees, but they were important to all of  86  Takahashi Tomio, Bushidō no rekishi Volume 3, 223-224.  33  them.  In this context, this study will show that the presence of a foreign “other” or “others”  was an essential element in the initiation of modern bushidō discourse and that the first formulators of bushidō were equally or more influenced by current events beyond Japan’s borders than they were by the historical samurai class. The third chapter examines the development of a “bushidō boom” that occurred after 1898, and traces its development through the Russo-Japanese War of 1904-05.  Buoyed by the  success of the Sino-Japanese War, many currents of Japanese thought became increasingly nationalistic, and it was natural that a “native” ethic such as bushidō would gain currency during this period.  Whereas bushidō theories before this time could be considered more  “internationalist” than nationalistic, the tone of discourse changed considerably after the war. The newly confident, and sometimes chauvinistic, bushidō that marked the bushidō boom of late Meiji built on the earlier foundations but quickly superceded them.  This change in tone even  led Uemura Masahisa, an earlier writer on bushidō, to criticize the appropriation of the concept by nationalistic and militaristic elements in 1898.  Uemura’s frustration at the “misuse” of  bushidō reveals one of the greatest strengths of the ethic: resiliency.  The legitimacy bestowed  on the concept by its alleged relationship with the historical samurai, combined with a lack of concrete historical roots that could be used to define or refute it, meant that bushidō was an ideal vehicle for nationalist sentiments of the type that came to the fore in the years around 1900. Bushidō combined easily with other nationalistic terms such as yamato damashii and kokutai to form nationalistic and militaristic ideologies. In this context, this study examines the roles of the 1898 journal Bushidō zasshi, Nitobe Inazō, and Inoue Tetsujirō in the spread and development of bushidō. It will be shown that Nitobe’s significance to Meiji bushidō theory was not nearly as great as his current reputation would indicate.  Instead, Inoue Tetsujirō was  the undoubted primate of bushidō from 1901 until 1945, and the “orthodox” bushidō interpretation he developed became a dominant ideology from the Russo-Japanese War onward. 34  It was only during the second half of the bushidō boom, from 1905 until 1914, that bushidō became a widely popularized subject both in Japan and abroad. Chapter four examines the processes by which “orthodox” bushidō became firmly established, and how it and other bushidō interpretations spread throughout the diverse fields of literature, academia, sport, and religion.  Due in no small measure to support from Inoue, bushidō came to play a central role in  military and civilian education, especially with the growth of spiritual education programs used to indoctrinate the troops with the desired virtues of loyalty and self-sacrifice. “Orthodox” bushidō also came to play a key role in the national ethics education program known as “National Morality” (kokumin dōtoku), outlined by Inoue in a series of articles and books beginning in 1908.  At the same time, the popularity and unquestioned patriotic credentials of  bushidō led to its frequent mention by writers of both literature and pulp fiction, while academics wrote many volumes on the subject.  Members of religious orders and promoters of various  types of sport, both traditional and foreign, called upon bushidō to popularize their causes and give them the patriotic legitimacy that was deemed so important at the time.  Foreign  interpreters of Japan also showed great interest in the bushidō, further raising its profile.  This  study will demonstrate that by the end of the Meiji period, Japan was saturated with bushidō and, in a reversal of the situation from 1900, there would have been few Japanese or foreigners interested in Japan who would not have heard of bushidō and had a general idea of its meaning. At the time of the Meiji emperor’s death in 1912, it seemed as though bushidō would continue to expand its reach, but this was not to be the case.  Chapter five discusses the  unexpected decline that bushidō experienced around 1914, which was closely tied to the end of Meiji and specifically the dramatic death of General Nogi Maresuke. In addition to examining the influences that led to this change in bushidō’s status, this study will provide an overview of bushidō discourse in the years before its ultimate revival in the early Shōwa period. Although a detailed analysis of bushidō’s role in the “dark valley” of the 1930s is beyond the scope of this 35  study and is a subject for future research, it will be shown that the strengths and resiliency that defined modern bushidō from its origins in late Meiji made a resurrection of the concept not only possible, but highly likely. This theme is also significant in the conclusions and suggestions for future research at the end of this study, in which subsequent cycles of bushidō popularization and decline are examined in the historical context of the subject’s modern development.  36  CHAPTER 1  Bridges between Tokugawa and Meiji Bushidō  Developments in the Bakumatsu period, specifically the late 1850s, had a strong influence on modern bushidō discourse. Several developments in the use of the term “bushidō” occurred almost simultaneously, even if their significance was not noted at the time. Bushidō suddenly appeared in the works of several writers within a span of only a few years, although two of them were acquainted, and it is likely that they influenced one another with regard to their use of bushidō.  Three writers whose surviving works contain reference to bushidō are Yoshida  Shōin, Yokoi Shōnan, and Yamaoka Tesshū. Their usage of the term is consistent with many of the same patterns seen in earlier works, but two factors make their writings relevant to the rapid development of bushidō in the late 1890s. The first is that all three men had an influence on the leaders of the new Meiji government.  Yamaoka was less prominent than Yoshida or  Yokoi, at least before 1868, but served both the Tokugawa and Meiji governments, allowing him to develop his concept of bushidō over the course of several decades.  Shortly before his death,  Yamaoka reportedly gave a series of lectures attended by several influential government officials, and these talks may be the first major expressions of bushidō as their central theme.  However,  recent scholarship has cast considerable doubt on the accuracy of the records of these talks, which were not published until 1902, and it is questionable whether the talks actually took place. This chapter examines the bushidō thought of these three individuals, and compares it with their impact on late-Meiji bushidō discourse.  Their writings are at times inconsistent with  their posthumous roles, and their status as bridges between Tokugawa and modern thought is not straightforward.  As we shall see, the writings of Yoshida, Yokoi, and Yamaoka tended to be  used less as inspiration for later writers than as carefully selected sources for the historical legitimization of modern agendas.  37  Yoshida Shōin The first writer in the bakumatsu period to use the term bushidō in his writings was Yoshida Shōin.  Yoshida was born in Hagi in 1830, and was adopted into the family of his  uncle at five years of age.  The Yoshida were in the employ of the Mōri family as teachers of  the Yamaga school of military science, and Shōin began studying strategy at a very young age. The breadth of his studies expanded greatly in his early twenties, when he traveled throughout Japan and met with scholars from many different schools of thought.  The most influential of  these was Sakuma Shōzan (佐久間象山, 1811-1864), who introduced Yoshida to Western learning and supposedly encouraged him in his plans to stow away on an American ship in 1854. This led to Sakuma being arrested along with Yoshida when the latter was captured.1  After  Yoshida’s plan failed, the two men were fortunate to escape execution, which was probably achieved only thanks to Sakuma’s political connections.  Instead, Yoshida was sent back to his  domain, where he was imprisoned and placed under house arrest for almost two years, although he was still able to study and lecture during this period. Upon his release in 1856, Yoshida began to agitate more openly for the overthrow of the Tokugawa, and was arrested again in Chōshū in 1858 out of concern that he may have been involved in assassination plots against several high shogunal officials.  The following year, he was sent to Edo, where he was executed  during the Ansei Purge carried out by the Tokugawa government under the direction of Ii Naosuke (井伊直弼, 1815-1860). Although he only lived for thirty years, Yoshida’s influence as a teacher was significant, and many future Meiji leaders studied at his Shōka Sonjuku, including Yamagata Aritomo (山県有朋, 1838-1922), Itō Hirobumi (伊藤博文, 1841-1909), and Kido Kōin (木戸孝允, 1833-1877).2 As the adopted son of a family that had traditionally served as teachers of Yamaga Sokō’s military strategies, it was natural that these would become one of Yoshida’s primary 1 2  Shinobu Seizaburō, Shōzan to Shōin: kaikoku to jōi no ronri, 156. Thomas M. Huber, The Revolutionary Origins of Modern Japan, 40.  38  sources of inspiration.  The most important of these was the Bukyō zensho (武教全書,  Complete Writings on Martial Teaching), a textbook of Yamaga’s writings used by his school of military strategy.3  The Bukyō shōgaku (武教小学, A Small Martial Teaching), which was often  used as the introduction to the Bukyō zensho, attracted considerable interest until well into the Shōwa period.  This was due to the emphasis the Bukyō shōgaku placed on the virtues of  Japanese warriors, giving the text an appeal to nationalist thinkers even in later eras. In keeping with the tradition of his adoptive family, Yoshida Shōin devoted a great deal of time to the study of the Bukyō zensho, and supposedly began to lecture on the subject at the age of eleven. Yoshida was praised by the daimyō of his domain for a series of lectures in 1849-50,4 and in 1852 he went beyond lecturing to write the commentary Bukyō zensho kōshō (武教全書講章, Lecture Section on the Bukyō zensho).5  These early lectures and writings remained fairly loyal  to Yamaga’s original texts on military strategy, both in content and terminology, and dealt primarily with practical applications, such as defending and attacking castles, as well as managing troops (yōshi, 用士). In his writings on strategy and tactics, Yamaga did not tend to use the term shidō (士道), and Yoshida’s early commentaries on Yamaga’s work followed this same pattern.  For example,  Yoshida’s 1852 commentary does not include shidō, and although it does mention budō, the term is used to refer to military training rather than ethical prescriptions.6  Like Yamaga, Yoshida  made no mention of bushidō in these early commentaries, nor in his other works before 1857. In that year, Yoshida composed yet another commentary on the Bukyō zensho, titled Bukyō zensho kōroku (武教全書講録, A Record of Lectures on the Bukyō zensho).  Unlike Yoshida’s  previous writings, which had focused on Yamaga’s teachings on military strategy, this work was almost exclusively a commentary on the Bukyō shōgaku, and had little to do with applied 3 4 5 6  Hirose Yutaka ed., Yamaga Sokō zenshū Volume 1, 479. Yoshida Shōin, Yoshida Shōin zenshū Volume 1, 15-16. Ibid., 14, 19. Ibid., 86.  39  military science.  The section headings of the Bukyō zensho kōroku correspond exactly to the  nine sections of the Bukyō shōgaku, with an additional introduction and conclusion. In this introduction, Yoshida lauded Yamaga’s teachings, and especially the influence that Yamaga was widely believed to have had on Ōishi Yoshio (大石良雄, 1659-1703) and his accomplices in the Akō incident. Yoshida went on to posit shidō as a way of “cultivating the body and correcting the heart, as well as ruling the nation and pacifying all-under-heaven.”7 Yoshida also discussed the existence of shidō in China, which he claimed had a different national polity (kokutai) than the land of the gods (shinshū, i.e. Japan), leading Yoshida to criticize those people who attempted to learn about Japan by reading foreign texts.8  This  association between shidō and China led Yoshida to cease using the term, for it is not found in the remainder of the Bukyō zensho kōroku after being used five times in the opening paragraphs. Instead, Yoshida replaced shidō with nativist synonyms, including budō, bushidō, and bushi no michi.  In his later works, Yoshida tended to use budō synonymously with bushidō, and not  merely to refer to military training as he had in the Bukyō zensho kōshō.9  In addition to  nationalistic considerations, Yoshida believed that the concept of bu (martiality) was essential in an age of crisis and conflict, such as Japan was experiencing during the late 1850s following Commodore Perry’s arrival.  Yoshida criticized prevailing attitudes towards military service,10  and emphasized bu as a type of all-encompassing concept of proper attitudes and behavior. Specifically, Yoshida stated that bu contained both bun (letters; civil) and bu, that bu was the same as chūkō (loyalty and filial piety), and that the teachings of bu (bukyō) contained Confucianism as well as all other teachings.11  7  Yoshida Shōin, Yoshida Shōin zenshū Volume 4, 208. Ibid., 208-210. 9 For an example of this broader usage, see Hirose Yutaka ed., Kō-Mō yowa, 267. 10 Yoshida Shōin, Yoshida Shōin zenshū Volume 4, 222. 11 Ibid., 215, 266. 8  40  Yoshida’s sudden elucidation of bushidō in commentaries on a subject he had discussed frequently throughout the previous decade is significant in the history of the concept.  There are  two factors that help explain Yoshida’s interest in bushidō and his use of the term.  First,  Yoshida simply combined the concept of bu, which was the focus of this commentary, with the shidō he was reluctant to use due to its association with China.  Yoshida’s associate Yokoi  Shōnan would repeat this lexicographical composition in his own work shortly thereafter. Second, Yoshida at this time had an interest in the Kōyōgunkan, citing it in the Bukyō zensho kōroku.12  These two factors may well have acted in combination, with Yoshida coming across  the term in the Kōyōgunkan at the same time that he was considering a new, bu-centered interpretation of Yamaga’s shidō theories. The new concepts in Yoshida’s 1857 Bukyō zensho kōroku, and his increased interest in the Bukyō shōgaku, reflect a further radicalization of his thought during his imprisonment from 1854-6.  His later thought was marked by an increased emphasis on death and an apparent  willingness to sacrifice his life in the name of a great cause (taigi).  Yoshida’s time in prison  also led him to write his most influential work, the Mencius commentary Kō-Mō yowa (講孟余 話), the inflammatory content of which meant that it was distributed in manuscript form and published only after 1871.13  In spite of its title, this text tended to disregard those sections of  the Mencius that did not agree with Yoshida’s jingoistic emperor-worship.14  The Kō-Mō yowa  was similar to the Bukyō zensho kōroku in that both were cases of Yoshida liberally borrowing from and reinterpreting well-known texts to expound his own theories.  However, the Kō-Mō  yowa does not mention bushidō, even though the two works were composed only months apart,  12  Ibid., 264. Ng Wai-ming, “Mencius and the Meiji Restoration,” 53. 14 The Kō-Mō yowa is a confirming example of Nakamura Hajime’s analogy regarding the Mencius, which he felt must have had a difficult voyage from China, as a few of its less agreeable books seem to have been washed overboard and thus failed to reach Japan. One example of this is the Mencian idea of just revolt against an emperor who has lost the mandate of Heaven, a notion that did not find favor in Japan. Nakamura Hajime, Ways of Thinking of Eastern Peoples, 470-471. 13  41  indicating that Yoshida either discovered the word in the interval between these two works, or that he did not deem the subject sufficiently significant to mention in his best-known text. In the short term, Yoshida’s use of bushidō in the Bukyō zensho kōroku did not have an impact beyond its possible influence on Yokoi Shōnan’s use of the term. factors.  This is due to several  For one, although Yoshida mentioned bushidō more than other similar terms in this  text, he still considered this work to be an updated commentary on shidō, as he mentioned in his introductory paragraphs.  Shidō, bushidō, and bushi no michi are not defined as separate  concepts in the Bukyō zensho kōroku, and budō is also used in a context that indicates that its meaning largely overlaps with these other terms.15  A second factor is that Yoshida did not  carry this use of bushidō over into any of his other writings, which encompass twelve volumes in one authoritative collection, but rather limited it to this one relatively minor text.  This has led  Wai-ming Ng and others to state that Yoshida did not use the term bushidō, reflecting the minor role the concept had in his theories.16  Another factor is that the political influence of Yoshida  Shōin in the bakumatsu period is difficult to ascertain clearly.  While Yoshida became one of  the most important figures of the Restoration in later historiography, there were many factions clamoring for change at the time. There were countless “political cliques,” as Albert Craig has termed them, with diverse motivations ranging from reform to complete overthrow of the regime.17  These cliques were aligned along political fault lines,, but they also had strong  regional loyalties.  For example, when Yoshida heard that sonnō jōi movements from other  domains planned to assassinate Ii Naosuke, he convinced the members of his Chōshū clique to kill a bakufu emissary first in order to enhance their standing among the various loyalist factions.18  On the other hand, Thomas Huber made Yoshida the focal point of his study on the  15  Yoshida Shōin, Yoshida Shōin zenshū Volume 4. In both his use of bushidō on page 220 and budō on page 232, Yoshida recommends that they be “polished” in order to serve one’s lord and nation, and their meaning is broader than mere “martial arts.” Also see p. 222. 16 Ng Wai-ming, “Mencius and the Meiji Restoration,” 46, note 3. 17 Albert Craig, Choshu in the Meiji Restoration, 155-159. 18 Ibid., 159-160.  42  origins of the Meiji Restoration, arguing that Yoshida was working on a “blueprint” for reform that closely resembled the policies implemented in early Meiji, when his former students held powerful posts in the government.19  As these works show, even modern historiography has not  been able to make the role of Yoshida in the Restorationist movement completely clear. The treatment of Yoshida’s legacy during Meiji is another factor that calls into question the positing of Yoshida as the primary forethinker of the Restoration. Yoshida was certainly respected in his home domain after his death, and is widely regarded as the primary exponent of the sonnō (emperor-revering) ideology that would later become the dominant interpretation,20 but there do not seem to have been movements to commemorate him on a larger scale until the mid-1880s.  A Yoshida Shrine was built in Tokyo’s Setagaya ward in 1882, and the Sonjōdō  memorial hall was built in his honor in Kyoto in 1887, where it now serves as a repository for documents and other materials related to the restorationist movements.21  From 1887 until 1900,  there was a considerable amount of official activity commemorating Yoshida, including his induction into Yasukuni Shrine (1888), posthumous awarding of the Senior Fourth Rank (shōshi i, 1889), and visits to Yoshida’s rebuilt school and another shrine by the future Taishō emperor (1898).22 In addition to the temporal trajectory of government ceremonies and physical monuments to Yoshida, the publication records of his own texts and other research concerning his historical role are also revealing.  According to Tanaka Akira, the Meiji period saw the  publication of 14 books and magazines concerning Yoshida, in addition to 38 editions of his own writings. By comparison, the Taishō period, which was only one-third as long, had 20 research publications and merely two editions of his writings, while publications during the first decade of Shōwa (to 1936) included 34 research works on Yoshida, and eight editions of his own 19 20 21 22  Thomas M. Huber, The Revolutionary Origins of Modern Japan, 42. See, for example, Albert Craig, Choshu in the Meiji Restoration, 162-163. Yoshida Shōin, Yoshida Shōin zenshū Volume 1, 47. Ibid. 48.  43  writings.23  The first work to discuss Yoshida did not appear until the 24th year of Meiji (1891),  and the most significant early work on the subject, Tokutomi Sohō’s (徳富蘇峰, 1863-1957) Yoshida Shōin, was not published until 1893.24  In this latter work, the popular rights activist  Tokutomi portrayed Yoshida as a type of social revolutionary comparable to the Italian activist Giuseppe Mazzini (1805-1872).  This view of Yoshida as a revolutionary nationalist later  inspired right-wing groupings such as those formed by young officers in the 1930s, and also provided a model for leaders of Chinese revolutionary movements, such as Confucian activist Kang Youwei (康有為, 1858-1927).25 Tokutomi underwent a profound change in his ideological outlook during the Sino-Japanese War, adopting an imperialistic political philosophy that was opposite to his earlier thought.  This shift was apparent in the second edition of his Yoshida Shōin, which he revised  and published in 1908 after the success of the Russo-Japanese War.  The chapter comparing  Yoshida to Mazzini was left out of the new version, while new chapters titled “Yoshida Shōin and National Polity Theory (kokutairon),” “Yoshida Shōin and Imperialism,” and “Yoshida Shōin and Bushidō” were added.26  Soon after the reissuing of Yoshida Shōin, there was a flood  of publications accompanying the fiftieth anniversary of his death. These works argued for both the populist and imperialist interpretations of Yoshida, with the latter becoming the dominant view in the following three decades. On the basis of publications and commemorative activities, it seems that interest in Yoshida Shōin reached significant levels only in the middle and end of the Meiji period, not immediately after the Restoration. Miyazawa Seiichi sees the lionization of Yoshida in the context of a larger “bakumatsu-Restoration boom” that began in the late 1880s and sought to  23 24 25 26  Tanaka Akira, “Yoshida Shōin zō no hensen,” 35. Ibid., 26-27. Miyazawa Seiichi, Meiji ishin no saisōzō, 19. Tanaka Akira, “Yoshida Shōin zō no hensen,” 30.  44  reevaluate the importance of notable figures on both sides of the Restoration struggles. 27 Yoshida’s relatively obscure writings on bushidō remained unknown during the hectic period immediately before and after 1868. In fact, the connection between Yoshida and bushidō would not be drawn until 1901, when Inoue Tetsujirō used Yoshida’s writings as an example of the influence of Yamaga Sokō, who Inoue called the “sage of bushidō.”28  Yokoi Shōnan Yokoi Shōnan’s use of bushidō was more limited and specific than Yoshida’s, but it was also more important to the development of the concept before 1901. for this.  There are two reasons  First, although Yokoi only used bushidō twice, he clearly defined the term when he  used it in his best-known work, the Kokuze sanron.  And second, Yokoi lived into the Meiji  period and was an important figure for almost a decade after Yoshida was executed.  Yokoi,  like Sakuma Shōzan, was a generation older than Yoshida, and was already a respected thinker in reformist circles when the two became acquainted.  Unlike Yoshida, Yokoi was more  interested in practical reforms than revolutionary action, and supported attempts to avoid conflict through the unification of court and bakufu (kōbu gattai) shortly before the Meiji Restoration. This position was later responsible for his assassination by extreme loyalists.29 Although Yokoi met and exchanged ideas with Yoshida and other young “men of spirit” (shishi), after the early 1850s his thought differed from theirs in important ways, which may 27  Miyazawa Seiichi, Meiji ishin no saisōzō, 17. This is discussed in greater depth in Chapter 3. 29 As William G. Beasley explains, the term “loyalist” is somewhat confusing, for it is often difficult to determine exactly who would fit into this category. The thought of many reformers and activists could at some point be labeled loyalist, but changes in personal ideology during this time were not infrequent. In addition, the activities and views of many restorationists were retroactively changed or exaggerated by either themselves or their biographers during the Meiji and later periods (William G. Beasley, The Meiji Restoration, 159). These difficulties are not limited to the loyalist label, however, and many different groupings and affiliations in the Edo Period would become subject to considerable historical revisionism during Meiji and beyond. One of the most complex examples of this process is the retroactive affiliation of thinkers and activists with the teachings of Wang Yang-ming, whether or not they were actually interested in his writings. For a discussion of this process, see Oleg Benesch, “Wang Yangming and Bushidō: Japanese Nativization and its Influences in Modern China.” 28  45  largely be attributed to his greater age and experience.  As William Beasley has pointed out,  only 14 of the 121 loyalists in Satsuma and Tosa in 1862-3 were 35 years of age or older, a figure which also seems to be representative of loyalists in other han.30  In contrast, Yokoi  Shōnan turned sixty that year. The youth and eagerness of the shishi was reflected in a desire for “action,” although the content of this was rarely defined or agreed upon.  Even when calling  upon established philosophy, as in Yoshida Shōin’s commentaries on the Mencius, the shishi selectively interpreted and cited classical works to support their own revolutionary aims. Additionally, in the case of the sonnō factions, the emperor was posited as the focus of absolute loyalty, freeing the shishi from the traditional bonds of lord-vassal relations.31  More like a  transcendent deity than a direct superior or even daimyō, the emperor’s inaccessibility made it possible to justify a wide scope of action on his behalf. As a result, the thought of Yoshida and many of the younger loyalists did not contain practical, constructive recommendations in case their revolutionary aims succeeded, leading some scholars to describe it as nihilistic.  As  Maruyama Masao argued, loyalty to one thing simultaneously represented rebellion against its alternatives, and for many of the shishi, absolute loyalty towards an idealized emperor represented absolute rebellion against the old order, resulting in the assassinations of Yokoi Shōnan, Sakuma Shōzan, and others whose views were considered too moderate in the 1860s.32 In spite of their frequent collaboration and exchange of ideas during the 1840s and 50s, the later Yokoi Shōnan can be viewed as a conservative thinker in comparison with Yoshida and the loyalists.33  During the earlier period, Yokoi’s thought contained many elements that were  considered radical, earning him status as a venerated theorist among the younger reformers and activists. Yokoi’s practical approach and interest in technological and institutional adoptions from other nations was present from the outset, as recorded by Motoda Eifu (元田永孚, 30 31 32 33  William G. Beasley, The Meiji Restoration, 158. Ikegami Eiko, The Taming of the Samurai, 324. Maruyama Masao, Chūsei to hangyaku, 11. Alistair Swale, The Political Thought of Mori Arinori, 9.  46  1818-1891), but even as late as 1853 he insisted that death would be preferable to the shame of making any concessions to foreigners, such as allowing them to land in Japan. 34  The  conviction with which Yokoi argued for the preservation of the nation’s honor deeply impressed younger loyalists such as Yoshida.  Three decades after Yokoi’s assassination, Katsu Kaishū  (勝海舟, 1823-1899) remembered the fearless intensity Yokoi was capable of displaying, noting in his memoirs that Yokoi and Saigō Takamori (西郷隆盛, 1827-1877) were the only truly fearsome men he had ever encountered.35  In spite of this intensity, Yokoi’s pragmatic nature  allowed him to realize when circumstances made his beliefs untenable. As a result, soon after the arrival of Perry’s fleet, Yokoi’s thought underwent a major change, and he advocated opening the country to trade with other nations, although he stipulated that this be limited to trading partners who would do so in a respectful manner.36 The show of force put on by the American fleet certainly helped convince Yokoi that his earlier insistence on military resistance to the foreign threat was not practicable.  Many scholars  believe, however, that a more important factor was Yokoi’s discovery of the Kaikoku zushi (海 国図誌, Haiguo tuzhi, Illustrated Treatise on the Maritime Countries), a Chinese overview of the nations of Europe and America. This text, written by Wei Yuan (魏源, 1794-1857) in the aftermath of the Opium War in 1844, is credited with convincing Yokoi that Japan could not compete with the advanced weaponry of the European powers.37  This text also addressed  another problem that Yokoi and other Confucian-influenced thinkers in Japan faced when dealing with the West: the notion that the Europeans were barbarians and could therefore not be negotiated with.  Yokoi and others realized that military resistance was futile, but rejected  dealing with those nations that lacked righteousness and justice.  They were able to overcome  this obstacle with the aid of the Kaikoku zushi, which included a very positive assessment of 34 35 36 37  D.Y. Miyauchi, “Yokoi Shōnan's Response to the Foreign Intervention in Late Tokugawa Japan,” 270-1. Katsu Kaishū, Hikawa shinwa, 12. Miyauchi D.Y., “Yokoi Shōnan's Response to the Foreign Intervention in Late Tokugawa Japan,” 274-5. Minamoto Ryōen, “Yokoi Shōnan ni okeru jōi ron kara kaikoku ron he no tenkai,” 208-9.  47  Western politics and government.  Russia, especially, was described as an enlightened nation  with a ruler comparable to the Confucian sages.38  Yokoi adopted this interpretation and,  according to Motoda, later drew comparisons between George Washington and the sage-kings Yao and Shun.39  This view of the West, which was almost as exaggeratedly laudatory as the  traditional view had been unrealistically negative, eliminated many difficulties by providing a basis for negotiations with Europe and America. This pragmatic approach differed greatly from the thought of the younger loyalists such as Yoshida, for whom negotiation with the West remained unthinkable. Yokoi’s willingness to adapt resulted in his assassination by sonnō jōi activists, but his practical thought also earned him positions of great influence.  After thoroughly impressing the daimyō of Echizen,  Matsudaira Yoshinaga (Shungaku; 松平慶永, 1828-1890), Yokoi was summoned to Edo when the former was installed as highest government official (seiji-sōsai) in 1862. For roughly five months, Yokoi held a position that has been described as the “brain” of the shogunate.40 Although his period in charge of political matters was limited, Yokoi had a major influence on the reform of the sankin kōtai (alternate attendance) system, a move made to reduce wastage that also severely weakened one of the last vestiges of control over the domains held by the central government. Yokoi’s proposals for the reform of currency policies, military defenses, agriculture, and especially trade were outlined in lectures and texts such as the Kokuze sanron, setting his approach apart from the “negative” one attributed to many loyalists. This difference can also be seen in the attitudes of Yokoi and Yoshida towards the condition of the samurai.  Both men  considered the samurai of their time to have degenerated from an ancient ideal, largely due to the misguided separation of bun and bu that had occurred during the Tokugawa peace.  38 39 40  While  Ibid. 203-4 D.Y. Miyauchi, “Yokoi Shōnan's Response to the Foreign Intervention in Late Tokugawa Japan,” 275. Okazaki Masamichi, “Yokoi Shōnan no seiji shisō,” 115.  48  Yoshida emphasized the importance of reintroducing bu, as shown in the Bukyō zensho kōroku, Yokoi was not convinced that there was a place for all samurai in the reformed military he proposed, and their numbers had become a burden on the state. Instead, many samurai should contribute to society by pursuing other professions, such as silkworm raising, fishing, metalworking, and sericulture.41  In contrast, Yoshida focused on the necessity of samurai to be  willing to die for their fathers, lords, domains, and especially their nation (闔国, kōkoku), and would not stoop to discussing the possibility of them engaging in mundane tasks such as mulberry farming.42  As Yoshida stated, the primary meaning (dai-ichi gi) of his commenting  on the Mencius was to explain that the people (shinmin) have a duty to die for the imperial state (皇国, kōkoku).43  In other words, whereas Yokoi took a more practical approach to the  difficulties facing the warrior class, Yoshida argued that any problems, including hunger, cold, and other tangible concerns, could be solved by strengthening the spirit and inculcating martial virtues.44 In the context of purely military matters concerning the samurai, Yokoi and Yoshida’s thought was less divergent, and they may well have influenced one another.  Yokoi realized that  it would not be possible to incorporate all samurai into a modern military, and he was also aware of the necessity of military technology, especially in naval warfare. However, he did not believe that technology alone could win conflicts, and argued that the proper spirit was essential for troops to succeed. Although many samurai had become “arrogant soldiers” (kyōhei) and could not currently compete with the West, it was possible to recover the lost martial spirit.45 With regard to his usage of bushidō, Yokoi had been acquainted with Yoshida Shōin since at least 1853, and they had several discussions as well as written correspondence.46 41 42 43 44 45 46  Satō Shōsuke et al. eds., Nihon shisō taikei Volume 55, 443-4. Hirose Yutaka ed., Kō-Mō yowa, 23, 193. Ibid., 336. Ibid., 36. Minamoto Ryōen, “Yokoi Shōnan ni okeru jōi ron kara kaikoku ron he no tenkai,” 221. Huber claims that Yokoi and Yoshida first met when they were both in Kumamoto in 1850, but other sources  49  Yokoi may have been familiar with the Kōyōgunkan, which Yoshida certainly knew, and he referred to Katō Kiyomasa and Honda Tadakatsu as examples of martial valor and ability.47  It  is also possible that Yoshida was the source of Yokoi’s usage of bushidō, as the former’s Bukyō zensho kōroku appeared in 1857, a little more than two years before the 1860 Kokuze sanron. The Bukyō zensho kōroku was not published at the time, but it seems to have reached a large audience for a hand-written document, making it likely that Yokoi would have been aware of its existence.  Yokoi and Yoshida also had many common acquaintances, including Sakuma  Shōzan, Katsu Kaishū, and Aizawa Seishisai (会沢正志斎, 1782-1863).  The timing of Yokoi’s  use of bushidō is further indicative of Yoshida’s influence, for the term is not found in any of Yokoi’s earlier works that predate the Bukyō zensho kōroku. Yoshida did not specifically differentiate between bushidō, bushi no michi, shidō, and other terms, but his emphasis on the concept of bu may well have led Yokoi to create the term bushidō in the Kokuze sanron.  Yokoi frequently used the term shidō, and it appears on almost  every page of the Kokuze sanron, and is the title of the third section. However, Yokoi felt that existing shidō thought was excessively focused on letters, weakening the warrior class from its heyday in the Sengoku period. According to Yokoi, “Originally, bu was the principal content of shidō.  Therefore, if one knows what a bushi is, he cannot fail to understand bushidō.”48  Through a reconciliation of military and civilian virtues, Yokoi believed that it would be possible to realize his goals of enriching Japan while strengthening her military so that the country could contend with the increasing foreign threats. Yokoi only mentioned bushidō twice in the Kokuze sanron, but his role in the development of the concept is significant for the fact that he was the first to clearly differentiate the term from other concepts that had hitherto been used synonymously.  Like Yoshida, Yokoi did not have an immediate impact on the use of  indicate that the two missed one another on this occasion (Thomas M. Huber, The Revolutionary Origins of Modern Japan, 150). 47 Satō Shōsuke et al, eds., Nihon shisō taikei Volume 55, 459. 48 Satō Shōsuke et al, eds., Nihon shisō taikei Volume 55, 456, 461.  50  bushidō, but his writings entered into discourse on the subject in late Meiji.  Both writers were  introduced into modern bushidō by Inoue Tetsujirō, who cited Yokoi’s contention that Western studies was merely management and economics, and was lacking the moral aspect of Chinese (Asian) thought.49  For Inoue, Yokoi was more important for his perceived links with Wang  Yangming’s teachings than his specific statements on bushidō, although Inoue tended to conflate the two subjects. While Yoshida and Yokoi’s usage of bushidō did not set off a “bushidō boom” of the type that occured from the late 1890s onward, the existence of their texts, as well as their posthumous status as Restoration heroes in the Meiji period, provided evidence and justification for those who developed, historicized, and politicized bushidō from the end of the nineteenth century onward, with Inoue Tetsujirō the most prominent among this group.  The curious case of Yamaoka Tesshū50 Yamaoka Tesshū, who has been credited with developing bushidō and introducing it to the modern age through a series of lectures, was born in 1836 as Ono Tetsurō, the fifth son of a shogunal retainer in Edo.  From a young age, Yamaoka was interested in military matters,  studying strategy and swordsmanship under several renowned teachers.  In 1855, Yamaoka  Tesshū entered the bakufu military academy (kōbusho) and began to train under the guidance of Yamaoka Seizan (山岡静山, 1829-1855), the older brother of Takahashi Seiichi (Deishū; 高橋 49  Inoue Tetsujirō, Nihon yōmeigakuha no tetsugaku, 558. Like many other famous historical figures, including Yoshida Shōin and Saigō Takamori, Yamaoka has been the subject of great public interest, which has resulted in a many different accounts of his life. Yamaoka appears to have been a giant of a man, and is often described as a great swordsman and peerless drinker. While many of the anecdotes relating to Yamaoka may be based on fact, others are almost certainly apocryphal, such as the belief that, when he was ready to die, he assumed the zazen position and expired without moving. Although there are fewer fantastic accounts regarding Yamaoka than there were concerning Yoshida and Saigō, Yamaoka’s lesser status also meant that there were fewer reliable accounts by witnesses who could refute the more questionable claims. Historical events in which Yamaoka played an important role and disinterested accounts by contemporaries provide useful information, but unfortunately these do not add up to a full biography of the man. This is compounded by the fact that many writings attributed to Yamaoka are very likely partial or complete forgeries. As an authoritative and factually verifiable biography of Yamaoka is beyond the scope of this study, this chapter limits its discussion of Yamaoka’s life and work to those aspects that are corroborated by the greatest number of sources, and indicates when an event or text appears to be apocryphal or is not sufficiently supported by independent accounts.  50  51  泥舟, 1835-1903).  After Yamaoka Seizan drowned in the Sumida River, Tesshū married his  eldest daughter and took the Yamaoka name.  In late 1859, Yamaoka Tesshū joined Kiyokawa  Hachirō (清河八郎, 1830-1863) and a handful of other shishi to form a sonnō jōi party known as the Kobi no kai (“Tiger-tail Society”).51 Although Yamaoka does not seem to have been actively involved in the jōi actions of his compatriot shishi, members of the Kobi no kai were responsible for several high-profile assassinations of foreigners in the 1860s, including the killing of Townsend Harris’ translator, Henry C.J. Heuskens, in January of 1861.52  Many of the  Kobi no kai’s members, including Kiyokawa, were killed by the shogunate or rival factions during the 1860s, and at least seven of them were given posthumous imperial ranks after the Restoration.53 Yamaoka’s association with the group seems to have been based primarily on personal friendships with some of its members, rather than a desire to participate in its activities. If anything, Yamaoka’s motivation in joining the shishi seems to have been to prevent open conflict between them and the bakufu, a strategy that he communicated in a letter to Matsudaira Yoshinaga.54  The bakufu adopted this approach in 1863, creating a special army consisting  primarily of 300 itinerant shishi in order to placate the renegade elements and better control them. The government soon regretted this decision, however, as the new forces threatened to become uncontrollable, and later that year bakufu assassins killed Kiyokawa and dispersed most of the shishi.  The day after the assassination, one of Kiyokawa’s followers was able to recover his  head and bring it to Yamaoka, who buried it in the temple in which he was currently staying.55 Shortly thereafter, Yamaoka was arrested by the shogunate along with Takahashi, although both were soon released.56 51 52 53 54 55 56  In spite of these connections with radical elements, Yamaoka was more  Katsube Mitake, Yamaoka Tesshū no bushidō, 20. Reinier H. Hesselink, “The Assassination of Henry Heusken,” 344. Ibid., 351. Yamaoka Tesshū, Kuzū Yoshihisa ed., Kōshi Yamaoka Tesshū: denki sōsho 242 (Part I), 181-182. Reinier H. Hesselink, “The Assassination of Henry Heusken,” 350. Yamaoka Tesshū, Kuzū Yoshihisa, ed. Kōshi Yamaoka Tesshū: denki sōsho 242 (Part I), 202-203.  52  of an establishment man than a revolutionary, both before and after the Restoration.  His  forthright personality, which has been described as “stupidly honest” (baka shōjiki),57 certainly made him better-suited to official service than revolutionary intrigue. Regardless of these problems in the early 1860s, Yamaoka’s closeness to the last Tokugawa government can be seen in his progression through various official positions until he was appointed ōmetsuke (chief inspector) in 1868.  Soon after, his brother-in-law Takahashi  Seiichi introduced Yamaoka to the new shogun, Tokugawa Yoshinobu (徳川慶喜, 1837-1913). Through this meeting, Yamaoka was appointed as an aide to Katsu Kaishū, the commander of the bakufu military, and it was in this capacity that Yamaoka participated in the event which would make him most famous.  With the Satchō armies approaching Edo, Katsu directed  Yamaoka and another member of the Kobi no kai, Masumitsu Kyūnosuke (益満休之助, 1841-1868), to deliver a letter of surrender to the enemy commander, Saigō Takamori. Upon reaching Saigō’s camp in Sunpu (Shizuoka), they submitted the bakufu proposal for surrender on the condition that there would be no bloodshed when Saigō’s forces entered Edo.  The  agreement of these terms, which sealed the fate of the Tokugawa government while limiting the amount of potential violence during the civil war, was seen as Katsu and Yamaoka’s most significant achievement, making the latter famous overnight. The peaceful surrender of Edo was widely lauded, but there were also doubts regarding the honorableness of this action. Katsu’s role, especially, became the subject of debate in late Meiji, when his decision to capitulate was reassessed by Fukuzawa Yukichi in the context of bushidō.58 Through the fulfillment of his mission, Yamaoka had impressed not only Katsu, but also Saigō, leading to Yamaoka’s appointment to various important posts over the remaining twenty years of his life.  In Meiji, Yamaoka served as sanji (councilor) of Ibaragi Prefecture and kenrei  (prefectural head) of Imari Prefecture, before being appointed the personal bodyguard of the 57 58  Katsube Mitake, Yamaoka Tesshū no Bushidō, 14-5. See Chapter 2.  53  young Meiji emperor and then, in 1875, kunai daijō (position of fourth rank in the early Meiji imperial court).  During his time in the imperial household, Yamaoka earned the favor of the  emperor, who is said to have shared Yamaoka’s fondness for drinking parties.59  Although he  reached the position of kunai daisuke (third rank), Yamaoka retired from palace duties in 1882 after ten years of service.  During his time in the imperial court, Yamaoka continued to train in  swordsmanship and study Buddhism, as he had been a devoted follower of the Boddhisatva Kannon since his youth.60 After retiring, Yamaoka focused more intensely on these activities and is renowned as the founder of the mutō (no-sword) school of kendō.  In addition,  Yamaoka’s interest in Buddhism led him to found a Rinzai Zen temple, Tesshū-ji, in the Shimizu area of Shizuoka in 1883.61  Yamaoka succumbed to stomach cancer in 1888, with the disease  usually attributed to his legendary love of good food and strong drink. During the final two years of his life, when his health had begun to deteriorate, Yamaoka wrote prolifically, and was said to have written well over 100,000 pages in this period.62  Shortly before his death, he  received visits from the emperor and empress, with both of whom he had maintained a warm relationship. Yamaoka was busy in the year before his death, founding the conservative nationalist group Nihon kokkyō daidōsha (The Society of the Great Way of Japanese National Teachings) together with Torio Koyata (鳥尾小弥太, 1847-1905). 63  In addition, Yamaoka has been  credited with delivering a series of lectures on the subject of bushidō that have been described as the point of origin of the concept in the modern sense.64  These talks were not published until  59  Donald Keene, Emperor of Japan: Meiji and his World 1852-1912, 288-9. Yamaoka Tesshū, Kuzū Yoshihisa ed., Kōshi Yamaoka Tesshū: denki sōsho 242 (Part I), 88. 61 Ibid., 139. Yamaoka’s interest in Buddhism, and religion in general, was quite broad, and he first began to study Zen at the age of 13 (p. 87). 62 Ibid., “Kaisetsu,” p. 3. This staggering figure has to be qualified, for most of Yamaoka’s “writings” were kigō (drawings or calligraphy) made to raise money for various causes. For example, the Feb. 27, 1887 edition of the Yomiuri shimbun carried an article announcing that Yamaoka intended to “write” 200,000 kigō for the benefit of Matsudaira Tadanori (松平忠敬, 1855-1919). 63 Donald H. Shively, “The Japanization of the Middle Meiji,” 107. 64 Anatoliy Anshin, “Yamaoka Tesshū no zuihitsu to kōwakiroku ni tsuite,” 103. 60  54  fifteen years after Yamaoka’s death, but many prominent figures were listed in the audience, indicating that they would have been known to much of Tokyo’s elite.  Yamaoka’s writings  indicate that he had been interested in the concept of bushidō for almost three full decades and, of the three individuals that are the focus of this chapter, Yamaoka’s use of the term bushidō was the most developed, as shown by the various ways in which he used the word after discovering it in older texts. Yamaoka first mentioned bushidō in a brief text written in 1858, using the term synonymously with more common phrases such as bushi no michi.  Yamaoka initially used the  term in conjunction with ginmi and fuginmi, following the pattern of bushidō usage established throughout the preceding two centuries. 65  The term ginmi is found in combination with  bushidō in the majority of the few texts that mention bushidō, including works associated with Takeda Shingen and Katō Kiyomasa, as well as the Hagakure, Bugaku keimō (1801), and Shijin roku (1840). Kasaya Kazuhiko has argued on the basis of an 1860 text that Yamaoka was not familiar with the Kōyōgunkan or other early works containing bushidō, and must have heard the term in conversation, as Kasaya believes that it was commonly used at the time.66  However, in  the 1860 text, Yamaoka specifically stated that he was the originator of the term, casting doubt on arguments that bushidō was widely disseminated in the middle of the nineteenth century. Certainly, Yamaoka did not consider bushidō to be an established term, but he overstated his case by claiming that he was the first to discuss the subject.  As the above analysis of  Yamaoka’s earliest mention of bushidō indicates, it is highly likely that he derived the term from a much older text.  In this sense, Yamaoka may have felt that he was the first to introduce the  term into contemporary discourse, for bushidō was an obscure linguistic relic that could be freely appropriated.  65 66  Yamaoka Tesshū, Kuzū Yoshihisa ed., Kōshi Yamaoka Tesshū: denki sōsho 242 (Part II), 8. Kasaya Kazuhiko, “Bushidō gainen no shiteki tenkai,” 264.  55  Although he mentioned bushidō in earlier texts, Yamaoka first set about appropriating and properly defining the term in the 1860 text, written when he was in the midst of his involvement with the Kobi no kai sonnō jōi movement.  In this brief essay, Yamaoka began his  argument as follows: ”For us Japanese, there exists a subtle (bimyō) way of ethical thought. Confucianism, nor is it Buddhism.  It is not Shinto; it is not  Integrating the three ways of Shinto, Confucianism, and  Buddhism, there is a way of ethics (dōnen), the remarkability of which can be seen since ancient times especially in warrior houses (bumon). Tetsutarō (Yamaoka) has given this a name and called it bushidō.  However, I have closely read the old existing texts, but have seen nothing that is written  and transmitted in the sutras.  Along with transitions in my affairs and various experiences, it is a  type of morality (dōtoku) that has been born from my feelings and senses (kannen). Today, I am giving shape to this moral way (dōnen), but this should certainly not be a task of one morning and one evening.  I have gone in and out between life and death in various ways, and have melted, cast, and  cultivated these experiences so that today I have for the first time reached this point.  I now work to  actively develop this Way, for although in the past people have not spoken of it, it was secretly in their hearts and let them be without anxiety.”67  Yamaoka continued by explaining that bushidō, although unspoken, could be put into practice. As it was undocumented, Yamaoka argued that it could not be understood by merely studying books and writing poetry like a scholar or technician, but that practice was all-important. However, Yamaoka emphasized that “practice” of bushidō did not necessarily mean “action,” and that the heart/mind (kokoro) should take precedence over form/action (katachi). Furthermore, “only knowing victory, and never knowing defeat, is not bushidō.”68  According  to Yamaoka’s relatively pacifistic outline of bushidō, “tragic incidents” such as revenge killings and assassinations were against bushidō, and were the misguided actions of those who had strayed from the correct path.  Yamaoka cited Tokugawa Ieyasu as stating that “repaying  compassion (on) with compassion is a weighty task, while repaying compassion with violence is  67 68  Yamaoka Tesshū, Kuzū Yoshihisa ed., Kōshi Yamaoka Tesshū: denki sōsho 242 (Part II), 28-9. Yamaoka Tesshū, Kuzū Yoshihisa ed., Kōshi Yamaoka Tesshū: denki sōsho 242 (Part II), 30.  56  easy.”  Bushidō meant emphasizing the original kokoro, and making katachi submit to the  former when necessary. This essay provides an insight into why, despite his prolific writing and relatively high personal profile, Yamaoka’s bushidō theories did not gain wide acceptance.  Although he was a  member of the Kobi no kai sonnō jōi group and close friend of Kiyokawa and other extreme radicals, Yamaoka revealed pro-establishment leanings by citing Tokugawa Ieyasu using his usual honorific Edo-period title of shinkun (god-ruler).  Furthermore, Yamaoka’s bushidō  theories, like most writings attributed to him, were filled with Buddhist terminology and ideas, giving them a degree of compassion and pacifism that differed greatly from the writings of shishi such as Yoshida and Kiyokawa.  Yamaoka’s emphasis on Buddhism may not have been  problematic in the 1860s, but the political climate of early and mid-Meiji was expressly hostile to Buddhism and would have made it difficult to find an audience for overtly Buddhist texts. This may explain why there does not seem to be any existing material concerning Yamaoka’s Buddhist-influenced bushidō theories from a period of about two decades from the mid-1860s. By the late 1880s, however, the project of State Shinto had begun to show clear weaknesses, and official hostility towards Buddhism had begun to fade.69  On a more personal  note, Yamaoka’s retirement gave him time for other pursuits, and there are many newspaper reports from the 1880s mentioning his activities promoting swordsmanship and Zen Buddhism.70 In addition, the rapid deterioration of his health after 1886 would have been a consideration in completing and presenting his lectures on bushidō.  The bushidō that Yamaoka is said to have  presented in 1888 was more developed than the brief outline he had composed in 1860, but his ideas were fundamentally unchanged.  Buddhism continued to be the dominant theme of a  69  Klaus Antoni, Shintō und die Konzeption des japanischen Nationalwesens (Kokutai), 189-206. See, for example, the Yomiuri shimbun editions from May 17, 1882, regarding Yamaoka opening a Zen school at Tōzenji (p. 2), and Oct. 15, 1887, announcing Yamaoka’s pledge to write on 100,000 folding fan papers to raise funds for a fencing school (p. 2).  70  57  relatively pacifistic bushidō theory, and Yamaoka emphasized compassion to an even greater degree than in his earlier works.71 According to Kuzū Yoshihisa and Abe Masato, Yamaoka gave a series of four to six lectures on bushidō in the year before he died, attended by his old friends and acquaintances, including Inoue Kowashi (井上毅, 1843-1895).72  Abe Masato reports that he gathered the  records of these seminars and asked Katsu Kaishū to comment on them ten years later, and Abe compiled and published all of these documents 1902.73  By this point, bushidō had become  widely understood in Japan, and the market for this type of work was favorable, given the status of both Katsu and Yamaoka