Ikki

Social Conflict and Political Protest in Early Modern Japan

James W. White

• Publisher: Cornell University Press
• Year: 2016
• eBook ISBN: 9781501704581
• Topic: History
• Subtopic: Japanese History
• Index: History

The reign of the Tokugawa shoguns was a time of statebuilding and cultural transformation, but it was also a period of ikki: peasant rebellion. James W. White reconstructs the pattern of social conflict in early modern Japan, both among common people and between the populace and the government. Ikki is the first book to cover popular protest in all regions of Japan and to encompass nearly three centuries of history, from the beginnings of the Tokugawa shogunate in the 1590s to the Meiji restoration.

White applies contemporary sociological theory to evidence previously unavailable in English. He draws on the long historical record of peasant uprisings, using narrative interpretation and sophisticated quantitative analysis. By linking the texture of conflict to the political and economic regime the shoguns created, he casts doubt on competing interpretations of a contained, orderly society.

PART I

THE CONTEXT OF CONTENTION

1

THE POLITICAL CONTEXT

Popular contention in Tokugawa Japan was not a violent eruption from a staid and stable context, nor was it born from an archaic and stifling context as the precursor or agent of new identities and world views.1 Contention flowed from context, and without an understanding of context no understanding of contention is possible.

My treatment of context is selective. I am interested in aspects of the overall Tokugawa context conducive to stability and order or to contradiction, instability, and conflict. I am also arbitrary. My division of context is artificial, for the Tokugawa polity was a political economy, and society and culture were also part and parcel of the polity; but I hope that analytical clarity justifies the distinction. Some factors of interest have multiple relevance. For example, the official ideology of “benevolent government” was embedded in both policy and popular culture. I treat these separately in each context.

The Institutional Structure of Tokugawa Politics

For several centuries prior to the establishment of the Tokugawa regime Japan had been governed under a series of bakufu, or quasi-feudal systems featuring a nominal emperor at the top, a supreme warlord/aristocrat, or shogun (nominally the emperor’s vassal, but actually the ruler of the country), and a hierarchy of enfeoffed vassals and their vassals under him (Mass and Hauser 1985). The last of these systems disintegrated into civil war in the late fifteenth century, and for a century government was in the hands of regional powerholders. In the last half of the sixteenth century a series of leaders unified progressively larger pieces of the country, and in 1600 Ieyasu of the Tokugawa house defeated the last rival coalition and asserted military control over the entire country. A bit of pressure on the imperial court led to his appointment as shogun; thus empowered, he proceeded to divide up the country into Tokugawa house lands and fiefs allocated to allied lords, former enemies, Buddhist temples, and Shinto shrines. At the top he built his own bakufu, designed both to govern the house lands directly and to maintain control over all the lords.

The Tokugawa regime was an unprecedentedly centralized form of government, but it did not spring full-blown from the hand of Ieyasu. The previous bakufu had already built “power analogous to the power of a state” (Wakita 1982: 346), and Ieyasu’s immediate predecessors had taken major steps to dissolve independent guilds and free cities; impose national land surveys; disarm the common people and fix them on the land as a closed social class; make, unmake, and move lords around and raze their castles; move the samurai off the land and transform them from landed warriors to stipended urbanites; and establish new local-level political structures and revenue systems (Hall 1981: 15ff., 194ff.; Wakita 1982).

Nor did the regime reach maturity under Ieyasu. It was consolidated during the first half of the seventeenth century under his successors (Nakane and Oishi 1990: 22) and continued to grow in institutional sophistication into the first half of the eighteenth century (Nakai 1988: x–xi), achieving a level of bureaucratic impersonality, legalization, administrative rationality, and central control which, compared with early modern European states, is impressive (Mass and Hauser 1985; Totman 1967; White 1988c). The institutionalization of the regime continued throughout the era; the most important such developments for us were the Tribunal, a supreme judicial structure established in 1635; the Magistracy of Accounts, which also assumed judicial functions; roving inspectors and censors; and a substructure of lawyerlike roles and officially designated litigants’ lodgings in the shogun’s capital of Edo.2

The power to invade the jurisdictions of the lords had initially been coercive; after the imposition of Tokugawa rule it became increasingly administrative, founded on Tokugawa house lands (bakuryō or tenryō), which accounted for roughly one-quarter of the country’s total productive capacity (Kitazawa 1982), although gifts of territory to retainers shrank this resource base during the era (Tokyo Hyakunen Shi 1973, 1:1352). Control of the house lands, too, became increasingly administrative: roughly 40 percent were parceled out as domains to two thousand plus immediate retainers of the Tokugawa, or hatamoto, while 60 percent (the chokkatsu-chi or okurairi-chi) remained under the jurisdiction of the Magistracy of Accounts, which administered them through some fifty intendsants (Nakai 1988: 117ff.; Kitazawa 1982: 36–37; Hall 1991: 169–71). The intendants, however, were transferred frequently, and most lived in Edo and delegated to assistants; grass-roots control was in the hands of the village headmen. There were usually only a few dozen assistants, in territories equivalent to medium-sized lordly domains governed by a thousand officials or more (Totman 1967: 62–73; Rekishi Kagaku 1974: 75–76). The bakufu never possessed, and never tried to possess, enough officials to govern the house lands closely; it remained a weak presence and frequently entrusted its lands to the administrative control of nearby lords (Hall 1991: 205).

Tokugawa Ieyasu distributed fiefs to a variety of lords and retainers; these domains were often fragmented, and many parts of Japan were patchworks of jurisdictions with few or no coercive resources. Holding the top status were the lordly houses related to the Tokugawa family.3 Then came the immediate vassals, the hatamoto, followed by some 150 “hereditary lords” (fudai) who had been on Ieyasu’s side at the decisive battle for control in 1600, and finally about a hundred “outer lords” (tozama) whose houses had been on the losing side in that battle (Hall 1991: 152; Totman 1967: 111ff.). But status did not correlate with capabilities: the related houses held big domains but had no tradition of strength, being basically creations of the state; the hatamoto were fragmented and underadministered; and the hereditary lords’ domains were often widely dispersed in small parcels. The outer lords, by contrast, often remained in control of the large holdings they had held before 1600.

Thus status and power were dissonant, but what are of more interest to us are the fragmentation of jurisdictions and inconsistency of administrative practice and capability across jurisdictions. At one extreme, entire provinces were under the rule of a single lord; at the other, single counties bore literally dozens of separate bakufu, lordly, hatamoto, and clerical jurisdictions. In fact, sometimes single villages were divided among a dozen or more rulers, each of whom was entitled to a portion of the village taxes and often had a headman in his part of the village; unsurprisingly, no single lord was eager to take responsibility for social control in the village as a whole.

All lords were themselves under the thumb of the bakufu. After the battles of 1600 the Tokugawa eliminated or reduced the holdings of almost a hundred lords; the first five shoguns together eliminated 200 lords, created 172, transferred 280, and increased the size of 200 holdings, reallocating in the process roughly half of the country’s entire productive capacity (Hall 1991: 150–52; Aono 1988: 23). One family was moved ten times in four generations; one lord was moved three times in twenty-nine years (Aono 1988: 30). Additionally, each lord was limited to one castle in his domain; new lords had to be invested by the bakufu; weddings linking lordly families had to be approved by the bakufu; and all the lords were required to live in alternate years in Edo and, when absent, to leave their families hostage there. Clearly, when all these measures were in place the Tokugawa house had little to fear even from its old enemies, much less its vassals. The administrative prerogatives of the lords were limited by Tokugawa laws or hatto (Hall 1991: chaps. 4, 5; Harafuji 1978); every domain’s fiscal, economic, and social data were at the disposal of the state (Hall 1991: 158–59; Iinuma 1991: 102); and when the lords gathered in Edo there was “not a whisper of parliament” about them (Totman 1967: 36).

The Tokugawa state was in control: the autonomy of the domains was “as good as nil” (Iinuma 1991: 102). Most bakufu offices were occupied by hatamoto, though hereditary lords held those at the top; nevertheless, the hereditary lords saw themselves more as lords than as bakufu officials, and the Tokugawa reciprocated: they showed them no special favors, moved them more frequently than they did the outer lords, and exempted them from no regulations (Bolitho 1974: chap. 2, 116–18). Lordly domains possessed more extractive and coercive capabilities than did the house lands, though the hereditary domains were more poorly staffed, dependent, and smaller than the outer (Hall 1966: 357; 1955: 23; Bolitho 1974: 46). And the hatamoto lands practically invited dissidence: small to begin with, they became smaller and smaller until not one in ten was an economically and administratively viable unit.4

The Tokugawa polity was both cause and effect of popular contention. And one prime aspect of the polity was the balance of power between the bakufu state and other political elites. In the case of Japan the others were the lordly domains, or han—hence the common appellation, the bakuhan system. In each of the bakufu states that had ruled Japan since the twelfth century the center was limited, but it was also able to establish checks between the lords and manipulate the fragmentation inherent in the feudalistic structure (Mass and Hauser 1985; Berry 1987; Maruyama 1974).

The two questions to be asked about all such systems were What was the bakuhan power balance? and Why was it what it was? I maintain, first, that the Tokugawa state was unquestionably supreme and, second, that the limitations on state power were largely voluntary, in that the Tokugawa government decided early on not to attempt to build a fully centralized modern state, not to try to force the biggest domains to obey every one of its edicts (Brown 1993).

Concerning the first question, observers such as Harold Bolitho (1974: 38), Albert Craig (1961: 4, 36), and Mary Elizabeth Berry (1986) have stressed the nondevelopment of institutions typical of modern states and the breadth of domain autonomy and concluded that, with the exception of establishing “a monopoly of war” (Berry 1986: 245), the Tokugawa took few steps beyond a basically feudal distribution of power. I and others, however, have held that the bakuhan system was at least as centralized as any of the so-called absolutist states of early modern Europe.5

In the first place, a statistical profile of relative power bases is instructive: the Tokugawa house lands accounted for roughly 15 percent of the nation’s total productive capacity, and the hatamoto for another 10 percent.6 The related houses occupied roughly 15 percent, the hereditary lords 25 percent, and the outer lords roughly 35 percent (Totman 1967: 33; Craig 1961: 15; Murakami 1965). Thus the lords far outweighed the bakufu and its retainers, but there were some three hundred of them; they jealously protected their prerogatives vis-à-vis one another, and none posed any sort of a threat to the Tokugawa during the formative seventeenth century. I have already mentioned some of the state restrictions on lordly power and domain autonomy. Others include the requirement that bakufu law be incorporated into domain law; state regulation of currency, religion, and commerce; precedence of bakufu over domain courts; state-approved trusteeship, rather than ownership, of domains by lords; and frequent (albeit irregular) extractions from the lords by the state.7

But a taxonomical question remains: Of the different familiar types of states, which did the Tokugawa state most resemble? There are those who term it feudal,8 but I find the category of limited help (Murakami 1985: 403). It seems to me that the closest analogue to the Tokugawa state was the “absolute” state of early modern Europe—not because either was in fact absolute but because they exercised strikingly similar degrees of control over rival elites and their own common people and because the “statelike” attributes missing from the Tokugawa state were also missing from the European. The developmental level I have in mind is that of France immediately before Louis XIV, Prussia before the Great Elector, Russia before Peter, and England before Henry VIII.9

None of the “absolute” states of Europe could dispose at will of the property or liberties of nobles; the Tokugawa had unlimited power of attainder. The bakufu never had to buy the nobles off as did Frederick William of Prussia in 1653, trading state military and fiscal rights for nobles’ autonomy within their estates; the Tokugawa never achieved such fiscal authority, but they could intervene in a variety of ways between a lord and his people, as we shall see. Tokugawa Japan had no body equivalent to the French parlements or the English parliament, able to check the state; Japan after 1600 certainly never saw anything resembling the noble revolts of the Fronde and the English Revolution. One of the defining trends of the early modern state was the separation of state sovereignty from the ruling class, on the one hand, and the person of the monarch, on the other; I have already noted the separation of the bakufu from the lords, and by 1600 the notion of the ruler as an absolute and impersonal “public authority,” or kōgi, subordinate neither to lords nor to monarch was already current (Hall 1991: 89, 94; Berry 1986).

I have drawn a fragmentary picture, to be filled out hereafter, but the parallels between state structure and performance and popular contention in Tokugawa Japan and early modern Europe are sufficient to recommend the analogy. My framework of comparison is limited to but one period of European history, however. Why did this comparability not continue? That is, why did the Tokugawa state, after attaining ascendancy in 1600 and unquestioned national jurisdiction by 1650, not continue to centralize?

There is nothing inevitable about governmental growth. In the development of European nation-states the incentive for growth most often cited is defense against rivals both foreign and domestic. In the words of Charles Tilly (Evans et al. 1985: 170), “War makes states.”10 But Japan faced no foreign threats until the mid-1800s, and thus no pressure to maintain a large standing army or develop an all-encompassing taxation system. In light of the threats to civil order posed by armies themselves (Bercé 1980: 147) and the near-universal resistance to enhanced revenue measures, it seems quite plausible that the Tokugawa state opted to forgo full centralization once its own security was assured. It is possible, in turn, that the low military and extractive posture of the Tokugawa state contributed to its longevity. The state enjoyed the luxury of simply backing off when new revenue initiatives stimulated resistance, of letting the economy go its own way, and of leaving the lords alone as long as they kept their own domains quiescent and did not defy Tokugawa authority. The contrast to England and France, where fiscally desperate states pushed their countries clear to revolution, is sharp. So is the situation in Japan in the mid-1800s, when one can see a hint of how the Tokugawa state might have developed in a context of war. In the mid-nineteenth century foreign states began to nose around Japan’s shores, and the bakufu in response began to impose coastal defense burdens on the lords; these burdens were reflected in fiscal strain and heavier domain taxes, which in several instances provoked popular resistance.

A plausible alternative motive force for state building is domestic political competition, and here again the mid-nineteenth century offers a telling example of what the early Tokugawa had been lucky enough to escape. In the mid-18...

Table of contents

• List of Figures
• List of Tables
• Acknowledgments
• Note on Orthography
• Introduction
• Part I: The Context of Contention
• Part II: The Texture and Content of Contention
• Part III: The Correlates and Causes of Contention
• Part IV: Consequences and Conclusions
• Appendix 1: The Aoki Kōji Data
• Appendix 2: Magnitude and Type of Contention
• Bibliography