Toward a genealogy of the police idea in imperial Japan: a synthesis

Abstract

This paper explores how Japanese officials and others conceptualized police power at particular junctures in imperial Japanese history (1868–1945). It does so by synthesizing prior scholarship on the Japanese police into a broader genealogy of the police idea in prewar Japan, beginning with the first translations and explanations of police in the Meiji period, the changing perceptions of the police in the 1910s, and the evolution from the “national police” idea in the 1920s to the “emperor's police” in the late 1930s. The essay proposes that the police idea in Japan (and elsewhere) can be read as a boundary concept in which the changing conceptions of police power demarcate the shifting relationship between state and society. Indeed, it is the elusiveness of this boundary that allows for police power – and by extension, state power – to function within society and transform in response to social conditions. Approached in this way, the essay argues that the different permutations of the police idea index the evolving modality of state power in prewar Japan, and thus allows us to reconsider some of the defining questions of imperial Japanese history.

Introduction: police as boundary concept

In this paper, I sketch out a framework for analyzing the different formulations of the police idea in modern Japan and how they reveal the changing modality of police power in different historical periods. I take inspiration from recent critical histories of early modern Europe written by Neocleous (2000) and Campesi (2016) which, through Marxist and Foucauldian approaches respectively, trace the development of the police idea from the disintegration of feudalism in the fifteenth and sixteenth centuries, its development in the political discourse of the “science of police” in the seventeenth and eighteenth centuries, through to the police idea's reformulations during the reforms in Europe in the nineteenth century that produced the modern police forces we recognize today. Although no similar extended historical study of the police idea in modern Japan has been conducted, there have been a few studies which address unique theorizations of police power in different periods in Japan's prewar history, including that of Obinata (1993), Kawashima (2013), and Umemori (2002). Here, I will build upon and synthesize this scholarship in order to develop a longer genealogy of the police idea, drawing upon other institutional studies of the Japanese police including that of Westney (1987), Mitchell (1992) and Tipton (1990, 1997) where necessary. And since a full exploration of the police idea and its various permutations in modern Japan is beyond the parameters of an article, I will focus on the imperial period (1868–1945) and how an analysis of the police idea sheds new light on important developments in prewar Japanese history.

Informed by Mark Neocleous's thesis that “the history of police is the history of state power” and that it is “through policing that the state shapes and orders civil society,” I argue that the history of the police idea in imperial Japan can be analyzed as indexing the changing interventions of state power in social life as well as the changing social conditions that the police were tasked to manage between 1868 and 1945.¹ And although the majority of Anglophone scholarship on Japanese policing is focused on the postwar period,² a few studies that have explored policing in imperial Japan have centered on either the adoption of western organizational models in the late-nineteenth century, or the institutional transformations of the police institution during the political tumult of the interwar period and whether it constituted a “police state” in comparison with the German gestapo or Soviet Union's GPU.³ However, as I will demonstrate, an important aspect that requires further exploration is how both of these developments were predicated on nuanced conceptions of police power that were drawn from particular understandings of state formation, capitalist development, social order, national sovereignty, and empire.⁴ In the simplest sense, an analysis of the police idea illuminates how officials and others understood the function of the police in particular social and political circumstances in imperial Japanese history. In broader terms, however, the different formulations of the police idea can be read as revealing how the mediations of police power between state and society were reconceived at specific moments, and as such can offer a unique lens into the important developments in imperial Japanese history, “including the central role of police power in nation-state formation, capitalist development, colonial expansion, and the transformations of police power in response to the interwar cycles of economic and political crisis.”

In order to construct a genealogy of the police idea in imperial Japan, I will approach it as what Timothy Mitchell has called a boundary concept, which demarcates the shifting distinction between state and society, not as discrete, a priori objects, but as a “line drawn internally within the network of institutional mechanisms through which social and political order is maintained.”⁵ Mitchell's concern is with political techniques that produce what he calls the “state effect,” or “the ability to have an internal distinction appear as though it were the external boundary between separate objects” – here, state and society – which, for him, “is the distinctive technique of the modern political order.”⁶ Extending Mitchell's thesis, I argue that police is one of most important “techniques of the modern political order”: through its exercise of social administration and control, the police manifest state power within the social field, thereby reproducing the distinction between and mediating state power within society. Indeed, it was the rapid development of the Japanese police and the intensity of their interventions into everyday life by the early twentieth century that produced the sense of a state standing above and intervening into society – what was critiqued at the time as constituting a “police sate” in the 1910s, and which prompted officials to respond with new formulations of the police idea in the 1920s and 1930s.⁷ These different permutations of the police idea thus index the historically changing mediations of police power between the imperial state and Japanese society, and grounds them in the material conditions that the police were tasked to manage. But, before exploring the police idea in Japan, it is first necessary to clarify the methodological stakes of this approach and its significance for historical inquiry.

The police idea as history

Although many studies apply the term “police” and “policing” to pre-Meiji security forces and practices, the police idea did not exist until leaders of the new Meiji state decided to quickly establish such a system as one of the first steps in state formation following the Meiji Restoration (1868).⁸ It was not just that the term “police” (initially transliterated as polisu; later keisatsu) had not been translated until then, but that as James B. Leavell describes “Tokugawa society was an unpoliced society,” in that there did not exist a unified set of security measures nor a single apparatus across the territory of the Japanese islands with the mandate for administering social order.⁹ Rather, security in Tokugawa society was highly decentralized, not only geographically due to hundreds of semi-autonomous feudal domains which were ruled by feudal lords (daimyo) and their samurai retainers but also socially, in that each social group – samurai, peasant, artisans, merchants, and others – maintained their own internal socio-cultural orderings and punitive measures based on prescriptions unique to their class.¹⁰ And although investigators (yoriki) and detectives (dōshin) from the samurai military class investigated incidents in urban centers and domain castle towns, this was not a single or homogenous system, but a responsibility for samurai retainers of a specific domain.¹¹ This is to remind us that the idea of the police is unique to the formation of the modern state and needs to be understood as a particular mechanism through which a state administers a population within its sovereign territory. In other words, the anachronistic application of “policing” to all premodern security measures overlooks the historically specific function of the police and its relationship to modern state power and capitalist social relations.

Here, scholarship on European political history is instructive. Scholars of Europe have analyzed the development of the police idea, focusing on the political discourse of the “science of police” in the seventeenth and eighteenth centuries. This discourse presented a new understanding of social order during the disintegration of feudalism and bequeathed the term “police” to be revised and reformulated during nation-state formation and later.¹² Such an approach was inspired by Michel Foucault's lectures on the early French discourse on “polizei” and its coalescing into the “science of police” in continental Europe between the seventeenth and nineteenth centuries. Foucault notes that in these writings, police was not described as an agency for the prevention or punishment of crime, but rather as “the set of means by which the state's forces can be increased while preserving the state in good order.”¹³ That is, rather than a repressive form of power to control disorder, police was presented as the policies that constituted good governance of a territory and population, measured by a state's wealth and good order – what Foucault identified as the “state's splendor” – which it was believed would create “equilibrium” with other states.¹⁴

Building from Foucault's re-reading of “polizei,” Pasquale Pasquino concludes that in the science of police discourse, police power was conceived as a “great labou[r] of formation (mise en forme) of the social body, or rather a labour whose principle result is what today we would call society or the social body and what the eighteenth century called the good order of the population.”¹⁵ In other words, more than securing order, Pasquino argues that the new mode of power identified with police was to produce the social in which order was to be cultivated and secured. His thesis not only prompts us to rethink the conventional association between police and social order, but also where to locate the operations of state power, for the state's emerging statistical surveys and reports of splendor and well-being thus rendered power relations “wholly within the interior of the social body” thereby turning the state into a “sort of topographical survey, or rather the name given to this survey.”¹⁶ Recalling Mitchell's theory mentioned above, Pasquino thus finds in early theories of police power the unique “techniques of the modern political order” which produced the “effect” of a state standing above and producing the new forms of relations of modern society.

Along similar lines, other scholars including Alf Lüdtke (Prussia), Roland Axtmann (Hapsburg Empire), Franz-Ludwig Knemeyer, (Germany), and F. M. Dodsworth (England) among others have noted that, in its various regional and linguistic permutations, “police” encapsulated attempts to locate a new form of governance and order in the early stages of nation-state formation and capitalist development, and that the emerging police institutions were tasked with defining the contours of social practice and behaviors toward such ends.¹⁷ It was only later, in the early nineteenth century, that the function of police in Europe narrowed to signify crime prevention and law enforcement, symbolized by what has come to be epitomized in the Anglo-American policing model. But, even in this later refinement, the police idea continued to be associated with the administration of social order, for as Mark Neocleous has argued that police power operates along a “law-and-administrative continuum” in which “the “criminal law… [becomes] just one resource among many which a police officer uses…to achieve a well-ordered civil society.”¹⁸ As we will see, when Japanese officials deliberated organizing a police force in the early-1870s, it was this relatively recent combination of social administration and crime prevention in European policing that they turned to. In other words, the translation of the police idea in Japan and its institutionalization in the early Meiji period was part of a wider global process in which police power was being developed in the entwined processes of state formation and capitalist development on a global scale, and functioned as a political technology to produce – or what Neocleous terms “fabricates” – the kind of social order that was conducive to these projects.¹⁹

The historical relationship between policing and the formation of the modern state is nowhere more explicit than in the case of Japan, in which one of the earliest institutional developments of the new imperial state was the creation of a national police force in 1874. As Umemori Naoyuki has pointed out, the “political uncertainty” following the overthrow of the Tokugawa Bakufu inspired the new “Meiji leaders to establish intense and extensive security institutions and practices in order to consolidate their authority.”²⁰ These initial security institutions coalesced into a national police force in 1874 modeled on the French police, which predated many other elements that came to constitute the new state's governmental and prefectural system. As D. Eleanor Westney has argued, the “Japanese officials of the early Meiji government unhesitatingly viewed…[the police] as an essential part of the apparatus of the modern state,” and as such implemented innovations in the police in the 1870s and 1880s, which “extended the capacity of the [Japanese] police beyond that of most of the European systems the Japanese were monitoring so closely,” thereby creating a police system with “a level of [social] penetration unmatched in Europe.”²¹ It is, therefore, surprising to find that, although they had created such an extensive system by the 1880s, a decade earlier officials did not know exactly what a police agency was or what kind of power it exercised. And it was in their quest to understand this mode of modern power that we find the first formulations of the police idea in Japan.

Fukuzawa Yukichi and the translation of “police”

Umemori Naoyuki explains that as leaders “recognized the necessity of introducing a police system” after the Meiji Restoration, they “realized that no one had enough information on what the ‘police’ was.”²² In late-1870, the new government in Tokyo issued a formal inquiry into “establishing police” (polisu secchi), and quickly set out to gather information on what constituted such an institution and its function in modern society.²³ Apart from a few limited reports from samurai who were sent abroad in the last years of the Edo period or observations of security forces set up by foreigners in the recently opened port town of Yokohama, the new government did not have much information on the police. To rectify this, senior councilor (sangi) of the Great Council of State (Daijōkan) Hirosawa Saneomi (1834–1871) commissioned the intellectual Fukuzawa Yukichi (1835–1901) to explain the history and the function of the police in western countries.²⁴ Fukuzawa issued what is believed to be the first official document outlining the meaning of police in October 1870, entitled “Torishimari no hō,” or as Umemori translates: “The Rules of Policing.”²⁵ This document is an abridged translation of the entry for “police” from The New American Cyclopedia (1866–1868) that defined the police as “a judicial and executive system and an organized civil force for the preservation of order and the enforcement of the laws.”²⁶ However, it is important to remember that the conception of police as preserving order, preventing crime, and/or enforcing laws was fairly recent in Europe and America at the time, and was grafted upon the enduring administrative functions attributed to the police in the earlier “science of police,” particularly in their endurance in the continental police model.²⁷ As we will see, it was these administrative functions that would most appeal to the new Meiji leaders, as they would soon set out to orient the behaviors, morals, and thinking of the population toward such projects as “rich country, strong defense” (fukoku kyōhei).

Translating the entry for “police” in The New American Cyclopedia, Fukuzawa begins by introducing the history of the prosecution of crime in ancient Greece and Rome to present-day, and then outlines the contemporary police systems of France, Britain, and America, with particular emphasis on the centralization and administrative responsibilities of the French police.²⁸ Umemori highlights two important aspects of Fukuzawa's translation. First, he points to a brief explanation that Fukuzawa added which read “civil force means a force that is not military” (jōmu no kenryoku ha heiryoku ni arazaru o iu), and argues this was not simply a terminological clarification but should be interpreted as Fukuzawa suggesting that police constituted “a new form of power, not to be confused with a traditional military force” (i.e., samurai rule).²⁹ Second, this civil power included administering various aspects of modern societies, including commerce, traffic safety, weights and measurements, public amusement, publications, among other things, which Umemori notes did not exist in Japan in the 1870s to the degree or extent of Europe. Umemori asserts that, in this different context Fukuzawa's translation can be read as suggesting the “intimate relationship between police power and the development of ‘civil society’ [my emphasis].”³⁰ In other words, this meant the police's essential mandate would be what Pasquino identified as “the great labour of [social] formation,”³¹ rendered most explicit in Fukuzawa's revision of the first police task, in which he translated the securing of “the safety of traffic” as that of the “safety of commerce” (shōbai o anzen ni seshimuru koto).³² Umemori notes that this translation of the police's productive function can be read in connection with Fukuzawa's more general theory of civilization as the expansion and intensification of commerce, communication, and circulation, correlating with the diversification of wants and needs of a populace.³³ Read in this way, Fukuzawa's translation suggests that the police were envisioned to play a central role in facilitating this circulation, diversification, and expansion, and thus was central to producing civilization – what Umemori summarizes as police power operating as “an agent of civilization.”³⁴ And as we shall see in the next section, the first chief of the Tokyo Metropolitan Police, Kawaji Toshiyoshi (1834–1879), understood the police function in this way.

Kawaji Toshiyoshi and police as “nursemaid”

Soon after issuing a call to explore policing systems, the Meiji government established a civil police force in the new capital Tokyo in 1871, called the Torishimari gakari, or Metropolitan Control Office under the authority of the Ministry of Corrections (Gyōbushō) which was itself soon replaced by the newly created Justice Ministry (Shihōshō) the same year.³⁵ This Metropolitan Control Office replaced a contingent security force consisting of separate samurai groups patrolling the new capital since 1868 under the supervision of the Ministry of Armed Forces (Heibushō).³⁶ Evidence of the rapid changes in institutional affairs, the Metropolitan Control Office was reorganized in 1872 and renamed the Rasotsu, or Constabulary, based on a general understanding of western models.³⁷ Also in 1872 a Police Bureau (Keihoryō) was created in the Justice Ministry to oversee the organization of the prefectural police system.³⁸ The first chief constable of the Constabulary and then Chief of Police (Daikeishi) of the Police Bureau was Kawaji Toshiyoshi, who in 1872 was sent abroad to study the various police systems in Europe. Kawaji later became the first Chief of the Tokyo Metropolitan Police (Keishichō) and is considered the “father” of the modern Japanese police system whose writings are still read by police cadets today.³⁹

Kawaji's report to the Meiji government upon his return from overseas in 1873 can be said to be the first, fully elaborated conception of police power in Japan. Prefacing his report, Kawaji explained that the police was “an absolute necessity in strengthening the state” since it functioned as a “daily cure and remedy to a state, as everyday hygiene is to an individual.”⁴⁰ Police were not only important simply “to protect good citizens,” but as they “nurture the vitality [kiryoku] of a country” the police also contribute to foreign relations by making “imperial powers glorious” and making it possible to “annex surrounding countries,” as in the example of Prussia.⁴¹ If Fukuzawa had made the distinction between police and military to distinguish between modern police power and earlier samurai rule, here Kawaji was connecting the police's domestic function to cultivate productive power to the state's ability to exert this new power in the arena of imperialist competition. With the unequal treaties in mind, Kawaji concluded his report by noting that since Japan was treated as a “semi-sovereign” (hanshu naru mono) state by western powers, it “must establish a strict police system, dispense policemen and promulgate a law for the organization of courts of justice so that we are not despised by foreigners.”⁴² This not only shows, as Umemori has noted, why the Meiji government was willing to consider the formation of “a large security institution despite its strained finances,” but also exemplifies how, from its initial formation, the police were conceived as inseparably linked to the Japanese state's sovereign power both domestically and abroad.⁴³

Kawaji's report was instrumental in the formation of the Japanese police and other institutions of the new state. As he suggested, the Home Ministry (Naimushō) was created in November 1873, and within months the Police Bureau was transferred from the Justice Ministry to the Home Ministry.⁴⁴ Replicating the functional differentiation of the French police, the Home Ministry would oversee the police's administrative (gyōsei) functions, while its judicial functions (shihō) remained under the Justice Ministry. Also modeled on the French police was Kawaji's recommendation that a metropolitan police force be created in Tokyo to serve as the center of a prefectural police system, and that unemployed samurai be recruited into the new force similar to decommissioned soldiers in Europe. It was the continental models of police – namely, French and Prussian – that most appealed to Kawaji. He argued that the particular circumstances in Japan required a particular police force: whereas London was a “prosperous metropolis with wealthy citizens and lively commerce,” allowing Londoners “to contribute with pleasure to the support of their police,” in Japan Kawaji lamented that “present conditions of our country do not bear comparison” and thus requires a centralized national police force with extensive administrative powers.⁴⁵ Kawaji saw the police as instrumental in generating the prosperity that would lead to stronger support for the police among the populace.

As Elise K. Tipton summarizes, “the Meiji founders employed the term ‘police’ in the broad seventeenth and eighteenth-century sense of all internal administration, rather than the narrow sense of crime prevention and detection” as in the Anglo-American systems.⁴⁶ When the Justice Ministry produced a preliminary proposal for police regulations (keisatsu kisoku an) in June of 1873 – a few months before Kawaji's reforms were implemented – justice officials apparently copied verbatim the French regulations, with the help of a French jurist in Japan by the name of Georges Hilaire Bousquet. Similar to Fukuzawa's translation discussed earlier, the Justice Ministry's proposal delegated to the police many administrative tasks that did not correspond to the social realities of Japan at the time, including controlling modern transportation systems, monitoring extensive urban commerce, and industrial production.⁴⁷ Umemori concludes that the list included many “useless, irrelevant duties for mid-nineteenth Japan,” suggesting that the police were first going to be involved in “fabricating” (Neocleous) many of the socio-economic elements that police administered in the West.⁴⁸

The list for administrative policing also indicated how deeply their biopolitical interventions were to penetrate into the population, as they were mandated to monitor the health of the “masses” (shūsho), surveil food markets, observe family and marital conditions, and notice anything that might threaten the health and well-being of the population.⁴⁹ This administrative orientation would soon be institutionalized in the 1880s with the formation of the so-called “hygiene police” (eisei keisatsu) in the Tokyo Metropolitan Police department which was tasked to enforce hygienic practices in the everyday lives of the population and which extended its control in monitoring sanitation and hygiene practices into the twentieth century, fabricating what Ruth Rogaski has called in the Chinese context “hygienic modernity” in prewar Japan.⁵⁰

In the 1876 manual for new police officers Police Methods (Keisatsu shugan), Kawaji invoked the language of civilization in order to explain the police's administrative function in ordering society:

A nation is a family: the government is the parents, its people are the children, and the police are their nursemaid [hofu]. People who have not yet been fully civilized like in our country must be regarded as mere infants. It is the obligation of the nursemaid to nourish the infants. This is the reason why police is the most urgent problem for our country today.⁵¹

We can interpret Kawaji's metaphor of “nourishing” as indicative of the active “formation” (Pasquino) of society, and legitimizing the many administrative interventions that the police were soon to make into society in order to “civilize” the population, including dress, public behavior, hygiene, appropriate forms of entertainment and leisure, among other aspects of daily life.⁵² Kawaji recognized that such active interventions required that the police apply their efforts “day and night, without sleep or rest,” refrain from consuming alcohol even while off duty, do not become involved in politics, get permission from their commanders to marry, and thus serve as civilized exemplars to their fellow countrymen.⁵³ As Umemori argues, the disciplinary function that Kawaji envisioned started with the police's own behaviors and dispositions, an “effect of the ‘civilizing mission’” the imperial state invested in the new force.⁵⁴

By the turn of the twentieth century, the Japanese police had assumed responsibilities to not only manage social behavior but to also determine acceptable forms of political participation in the new polity.⁵⁵ Earlier, prefectural police were ordered to monitor political gatherings in the 1870s, had reinforced the army when quelling the Satsuma Rebellion in 1877, formed a department to monitor political radicals called the Higher Police (Kōtō keisatsu) based on the French model in 1886, and were tasked with managing the political tumult surrounding the promulgation of the Meiji Constitution in 1889 and the first elections of the Imperial Diet in 1890.⁵⁶ As we will see in the next section, this policing of politics took on new forms in the 1900s, and, combined with the police's expanding social interventions, produced critiques of the police that officials had to respond to with new formulations of the police idea.

The police and the crowd: the police idea into the 1910s

By the turn of the century, the problem concerning policing was not necessarily the perceived lack of a public ethos adequate to meet the twin demands of capitalist development and loyalty to the imperial state – i.e., “civilization” – as Kawaji's earlier nursemaid analogy suggested. Rather, by this time, “civilization” had generated its own unique forms of social and political contradictions that required police management. Thus, we find in the early 1900s a discourse on, following German scholars at the time, the “social problem” (shakai mondai) circulating in official and academic quarters, which influenced how the police function was understood.⁵⁷ In addition to inspiring greater attention to German policing, these concerns further expanded the police's administrative mandate by overseeing the application of a variety of labor laws, including the Factory Law, minimum wage and labor accident compensation laws, and the Health Insurance Law, in addition to managing the prostitution license system, publication laws, among many other aspects of urban social life.⁵⁸ And it was the increasing interventions of the police into everyday life that produced the sense of a state standing above and intervening into society – what Mitchell theorizes as the “state effect.”⁵⁹ This ubiquity also gave protestors an immediate target to direct their anger toward in the 1900s and 1910s, for as Robert Spaulding argues, state power was “not personified by the remote and seldom-seen elite” in the higher-echelons of government, “but by the ubiquitous civil police.”⁶⁰ It was them who became the immediate targets of the public's discontent.

In addition to administering social and cultural practices, Japanese police were also managing the political frictions produced from the demands of imperial patriotism, including taxation, military conscription (the so-called “blood tax” or ketsuzei), and sacrifices during Japan's wars with China (1894–1895) and Russia (1904–1905).⁶¹ Such frictions led to increasing tensions between the imperial state and the public, most clearly displayed in the periodic urban riots that started to occur after 1905 over a variety of political and socio-economic grievances, including tram fare increases (1906), tax increases (1908) as well as the inflation of rice prices following the Great War (1918).⁶² However, no matter the different grievances that ignited the riots, their immediate target was often the same: the police. This marked a shift in the police idea in Japan, from the notion of a civilizing power in society exercised by the state, to an instrument used by the state to repress the urban masses.

The most famous of the urban riots was the Hibiya Riot, which occurred on September 5–7, 1905 in Tokyo. The riot started from organized protests against the terms of the Portsmouth Treaty that concluded the Russo–Japanese War. For protesters, Japan did not receive adequate indemnity, and by signing the Treaty the government had gone against the supposed will of the emperor and his loyal subjects who suffered through the hardships of rationing and conscription. A gathering organized for September 5th in Hibiya Park in Tokyo was canceled that morning and police were deployed to close the park. Defying police orders, a crowd of around 30,000 assembled, broke through police barricades and held a short rally. Later in the afternoon a smaller contingent of protesters attempted to deliver an appeal to the emperor imploring him to reject the Portsmouth Treaty, which the police blocked. From there, fighting broke out between protesters and the police, which spread to other parts of the city and continued into the night.⁶³ On the 6th, martial law was declared in Tokyo and the Imperial Army's first division was called in to assist the police. Once the unrest subsided on the 7th, a total of 350 buildings were destroyed, 17 people had died, and 450 policemen and 500 civilians were injured. Importantly, the aftermath revealed that the police were the primary target for the protesters' anger: two main police stations and nine branch stations were destroyed, along with 70 percent of the police boxes (kōban) throughout the city.⁶⁴

The Hibiya Riot has inspired much debate among historians over its political and historical significance. Some Japanese scholars such as Inoue Kiyoshi have interpreted this riot as marking the historical emergence of the “masses” (taishū) as a political subject and thus the start of the period of so-called “Taishō democracy.”⁶⁵ In an important essay, Okamoto Shumpei complicates this interpretation by pointing out that rather than democratic ideals or an emerging class-consciousness, rioters were primarily inspired by nationalism and imperialism, concluding that the riot “was little more than a blind outburst, and its basic orientation was toward nationalistic chauvinism” expressed through loyalty to the emperor and against the government who signed the Portsmouth Treaty.⁶⁶

Be that as it may, it is important to recognize that urban riots continued to occur in Tokyo and elsewhere throughout the 1910s, with the police serving as the immediate focus of people's anger.⁶⁷ These frequent riots had different causes – including, increases in train-fare, or the defense of constitutional government – but the majority of them targeted the police. Surveying these riots, Andrew Gordon remarks that the Tokyo Metropolitan Police Department, “under the jurisdiction of the Home Ministry, was the least popular and most besieged institution,” and that the “hundreds of small, hard to protect, two- or three-man police boxes scattered throughout the city were easy prey, and crowds stoned or burned them in five” of the nine major riots that occurred in Tokyo between 1905 and 1918.⁶⁸ Gordon interprets the targeting of the police and other government offices as “betray[ing] a…new political sensitivity” of the urban masses.⁶⁹ Thus, although Okamoto argues that protesters saw themselves as loyal imperial subjects and thus understood the police as “interfer[ing] with the people's execution of their duty as loyal subjects of the emperor,” such an estimation does not account for the frequency of riots or explain why they continually targeted the local police.⁷⁰ The public's perception of the police and its close connection to the state through the Home Ministry evidences an important transformation in the police idea and thus the police's mediation and reproduction of the boundary between the imperial state and society.

For example, immediately following the Hibiya Riot in September 1905, newspapers reported that public attitudes toward the police had shifted from something to be respected to “something to be feared.”⁷¹ Obinata Sumio summarizes that such reports documented a “lost confidence” in, and “wide distrust” of the police.⁷² In this context of growing resentment against the police, local politicians, lawyer groups, and newspapers mounted a campaign to “abolish the Metropolitan Police” (keishichō haishi). For example, 5 days after the Hibiya Riots were over, the Tokyo-based Jiji shimbun published an editorial calling to “Abolish the Metropolitan Police” on September 12th, which was quickly followed by similar calls from other newspapers. These concerns coalesced to produce a committee to investigate the matter, which concluded by proposing that the Metropolitan Police be moved under the supervision of the Tokyo prefectural government (Tōkyō-fu), replicating other prefectural police forces.⁷³ Proponents reasoned that with its unique organizational structure directly linked to the Home Ministry, the Metropolitan Police force function as a particularly repressive state institution. Thus, if they could be turned into a local police force they would have a different disposition toward the local population.⁷⁴ The proposal was discussed in both the Imperial Diet and Tokyo municipal and prefectural legislatures in January and February 1906. However, the new Home Minister Hara Takashi opposed the idea of reorganizing the Municipal Police Department, and quashed such proposals.

However, officials were prompted to respond to the negative image of the Metropolitan Police force after the 1905 riot. Calls for the police to not be “overbearing” (ibaru) and to be “kind and polite” (shinsetsu teinei) were proposed as one way to build a new relationship with the public.⁷⁵ Counselor of Home Affairs, Mizuno Rentarō, lamented that while in Europe the police “are respected by society,” in Japan such a “public sentiment” does not exist. Mizuno thus called for cultivating the notion in Japan that the police “are not here to harass, but rather to protect [hogo].”⁷⁶ As can be expected, police leaders envisioned their administrative tasks as the best method for spreading this idea among the population and transforming the relationship between the police and masses so that riots such as those of the 1910s did not proliferate. However, with increasing social unrest in the economic recession following World War I and the resurgence of radical politics in the 1920s, police officials redoubled their efforts to formulate a conception of the police that would both respond to such criticisms while also extending the police's administrative interventions further into society.

1920s: the “national police” idea and the dispersion of police power

Following the 1918 Rice Riots and into the recessionary 1920s, state officials increasingly focused on the social nature of uprisings, poverty and what they called “social crime” (shakaiteki hanzai), as Ken Kawashima has analyzed.⁷⁷ In this context, the police's preventative mandate was emphasized thus inspiring a radical rethinking of police power in society. Spearheading such a rethinking was a new cadre of police officials that had risen through the ranks of the police and Home Ministry. Unlike earlier police officials who had been recruited as ex-samurai in early-Meiji, many of the new officials had graduated from Tokyo Imperial University, traveled overseas to observe colonial and western police forces, and kept up on the most recent police studies published in other countries, in particular experiments in what would now be called “community policing” in Britain and American cities.⁷⁸ These new officials propagated a new formulation of the police idea and sought to develop new organizations and methods to transform police–community relations.

At the head of this effort was Matsui Shigeru (1866–1945), who in 1919 assumed the directorship of the Police Training Center (Keisatsu kōshū sho) and used this office to disseminate a new conception of the police's function and relationship to society. Matsui was the first division chief of the Metropolitan Police Department during the 1905 Hibiya Riot, and whose posthumously published recollection of the riot provides one of the only accounts of the riot from the police's perspective.⁷⁹ Such experiences clearly influenced his subsequent conceptualization of the police and his work in the Police Training Center. Immediately following the 1905 riot Matsui set out to address police–society relations through internal publications on police training and studying police education in other countries.⁸⁰

Around the time of the 1918 Rice Riots, Matsui proposed the idea of “national police” (kokumin keisatsu), which blurred the distinction between the police and the populace, sublating both into the national polity. To Matsui and other proponents, this would not only entail organizational reforms – such as delegating police powers to civic groups in order to cultivate cooperation between the police and community – but more importantly for each member of the community to identify themselves with police power and for the police themselves to reimagine their role within the community.⁸¹ As Ken Kawashima summarizes, “the reorganization of the police during the immediate years following World War I thus abandoned the notion of the police state and instead followed the banner of the national police, which was said to operate in the name of defending society.”⁸² Obinata Sumio argues that this was when the police were no longer presented to be “the police for the present government” but rather “police for society, for the masses.”⁸³ This signaled a new stage in the police idea as officials set out to reconceptualize the police's function to mediate between state and society.

Throughout the 1920s, Matsui published articles in both popular magazines as well as police journals in order to advocate for his idea of a “national police” and how it would correspond to the increasingly complex conditions of Japanese society.⁸⁴ The first step required disseminating the “police idea” (keisatsu no shisō) among the populace.⁸⁵ Once internalized, the police idea would inform the practice of “self-policing” (jikei), turning every individual into police and thus allowing for closer collaboration between formal police agencies and the wider community.⁸⁶ Simultaneously, this required the police to reconsider their function as entwined with the community, blurring the boundary between the agency and civilians. Matsui famously called this double movement “the policification of the masses and the massification of the police” (minshū no keisatsuka to keisatsu no minshūka). In a June 1921 essay published in the magazine Taiyō, Matsui argued that it was necessary to cultivate an “idea of an unified ethics” (rentai dōtoku no gainen) between police and the community, in which the police were seen as “for the masses” (minshū no tame) and the people must themselves awaken to a “police-like self-conscious” (keisatsuteki ni jikaku).⁸⁷ Matsui explained that “the essence of the police is to develop [the conditions] for a type of communal living in which the masses are happy and content,” because the state itself “is the ultimate protector of the masses” and, recalling Kawaji's family analogy cited earlier, “state power is exercised similar to a father's love and affection towards his children.”⁸⁸ It follows that, as a state agency, the police attempt to guard society against harm through preventative and administrative measures. However, in the complexities of modern life, the police cannot detect and prevent all sources of harm within the community. This required that civilians awaken to “police-like” consciousness and identify themselves with police power in order to practice “self-policing.” One important campaign to disseminate the “national police” idea to the populace was exhibitions organized by prefectural police departments. These were held in order to, as a 1923 Chiba Police Exhibition report stated, to “have the people understand the police,” “evoke the principle of self-policing and collective prevention [jikei kyōei no nen o kanki shi],” cultivate “cooperation between the people and officials,” and ultimately “to achieve the so-called reality [jitsu] and essential aims of the police.”⁸⁹

Such an idea informed new institutional experiments as well. For example, in 1921, the Home Ministry's Police Bureau developed new programs to abrogate police power to civilian groups, which would not only function to further “secure the public peace” (kōan o iji) in times of emergency but also to imbue the population with the “spirit of self-defense/self-policing” (jiei･jikei no nen).⁹⁰ The emergency situation following the Great Kantō Earthquake of September 1923 provided the first large-scale implementation of delegating police powers to neighborhood groups, and, in view of the pogroms carried out against the Korean community (often with police acknowledgement if not also assistance), we must question what such violence reveals about the “police-like consciousness” that such groups were to manifest.⁹¹ Such horrendous episodes also remind us of the violence that attends to how the police have reproduced the racial and ethnic exclusions constitutive of modern nationalism and imperialism.⁹²

Nonetheless, Matsui's “national police” marked a major transformation in the police idea in imperial Japanese history, in which the boundary the police mediated between the imperial state and society became blurred, leading to major institutional and procedural reforms such as the formation of civilian self-policing groups. By 1930, Matsui could explain in a textbook for police that the “era of the police state” (keisatsu kokka jidai) was over, and that a new era in the state-police–society relationship had taken shape.⁹³ As Kawashima has noted, Matsui believed that the earlier “vertical relationship” (tate no kankei) between the Meiji “police state” and people had turned into a “horizontal relationship” (yoko no kankei) between state, police, and society.⁹⁴ It is important to recognize, however, that Matsui's theory was not a call for the “democratization” of the police, as some scholars have described.⁹⁵ Rather, Matsui's theory entailed diffusing police power into the fabric of society, and as such, extending state power into the minutiae of everyday life. If earlier the police idea had sustained the conceptual boundary between state and society, Matsui's theory blurred this distinction by eliciting all civilians to awaken to police-like consciousness and to carry out the administrative and preventative functions of police in their daily lives. As Kawashima has noted, Matsui's colleague Maruyama Tsurukichi went so far as to claim that as the masses internalized the police idea and policed themselves, this would produce “the social condition in which the police no longer exists,” which for Maruyama was the “final and ultimate goal of the police itself.”⁹⁶ In other words, for Maruyama and other officials the ultimate objective of the police was to so thoroughly delegate the police function to civilians that the police ceased to exist as a distinct agency of the state, approximating what Fredric Jameson has theorized elsewhere as a “vanishing mediator.”⁹⁷ Maruyama's theory is one of the few instances in which the ideological telos of policing has been articulated so explicitly. Of course, the police did not disappear in the 1920s. Rather, the police idea evolved within the turbulence of the 1930s, producing another conception of the mediations that the police were to perform between state and society.

1930s: reimagining the state/society boundary in a time of crisis

If in the 1920s formulations of the police idea had emphasized the dispersal of state power into society through the police's administrative interventions, during the political-economic crises of the 1930s we find new formulations of the police idea that emphasized the need to harness this dispersal to mobilize the public for state initiatives. In 1930, Security Section chief of the Police Bureau, Ishihara Tsunejirō, identified the police as the “state police” (kokka keisatsu), tasked with the responsibility of “protect[ing] the life of the state” against dangerous ideological threats.⁹⁸ And since the emperor was sovereign of the imperial state, the “state police” were thus also conceptualized as the “emperor's police,” or what Matsui Shigeru had come to call “His Majesty's police” (heika no keisatsukan) by the mid-1930s.⁹⁹

Some scholars characterize this as the point when the Japanese police turned away from the “people” and abandoned earlier democratic inspirations of the 1920s in order to acclimate to the rising nationalism and militarism following Japan's seizure of Manchuria in 1931. For example, Elise Tipton argues that the notion of the emperor's police signaled “a changed role of the police” which “mirrored the decline of ‘Taishō democracy’… [and] the ascendance of the military and of national socialist ideas over Western democratic ideas.”¹⁰⁰ Elsewhere she explains this as a result of external competition between the Home Ministry, Justice Ministry and Military Police concerning who had primary jurisdiction for protecting the imperial state from political threats.¹⁰¹ However, we can understand the ascendance of “emperor's police” in the 1930s as also coalescing from within the police's own evolving theories and practices, including from the earlier “national police” idea, since of course the emperor was sovereign and the national polity (kokutai) comprised of his imperial subjects. In other words, for Matsui and others, the earlier idea of the “policification of the masses” became the means through which to mobilize the national polity to serve the objectives of the imperial state in a time of crisis.¹⁰²

Another internal factor animating the ascendency of the “emperor's police” idea in the 1930s was that this was a decade in which regular officers were increasingly told to guard the imperial polity from ideological threats such as communism, as exemplified in Ishihara's argument cited above. Previously policing political threats fell within the special jurisdiction of the Special Higher Police (Tokubetsu kōtō keisatsu, or Tokkō).¹⁰³ However, in the political turbulence of the 1930s regular police officers were instructed to remain vigilant of ideological threats in their locale, a process that scholar Ogino Fujio has called the increasing “Tokkō-ization of regular police officers” in the 1930s.¹⁰⁴

The idea of the “emperor's police” continued to transform during the socio-political turbulence of the 1930s. By 1937, 1 year after the Tokyo Metropolitan Police assisted with quelling an attempted coup d’état by young officers of the imperial army and the same year that Japan invaded China, Matsui put forth the notion of “imperial-subject police” (kōmin keisatsu), explaining that “to be his majesty's police is at the same time to be the police for all the nation.”¹⁰⁵ And since the nation (kokumin) – or earlier masses (taishū) – were always-already imperial subjects, this meant that to be his majesty's police was to serve his majesty's imperial subjects. Matsui elaborated on this new conception of the police in a lecture in 1939, in which he drew upon imperial ideology to invest regular police functions with new importance.¹⁰⁶ First, Matsui posited that the police “serve the throne [kōun o fuyoku] through their duty of securing the peace,” thereby connected social order with the emperor-system. Furthermore, based “on the divine principle of coeval with heaven and earth [tenjō mukyū shinchoku]” the police unify politics and morals by “following the dictates of both the constitution and the imperial rescript on education” to “realize the unification of politics and morals.” He then turned to the police's function to administer society, noting that “the Japanese police is attentive to the nation's psychology and is consistently in touch with the conditions of social activity.” In the immediate wartime conditions, Matsui argued that the police work tirelessly “to accomplish the national-defense state and the new order in East Asia,” and strive “to complete the police of the Orient [tōyō keisatsu no kansei].” Ultimately, Matsui presented such campaigns as a vehicle for the police to “practice the police's way of the imperial subject” so that they can both “promote police-consciousness (imperial-subject police) to the nation” and internationally “to clarify an expanded imperial-way police [hiroku kōdō keisatsu]” which Matsui called “the police of eight corners under one roof” (hakkō ichiu no keisatsu).¹⁰⁷ Such a conception was propagated through the Police Training School where it was urged that regular officers study the esoteric teachings of ideological tracts such as the Ministry of Education's Kokutai no hongi (1937) in order to fulfill their police duties.¹⁰⁸

Conclusion: the police idea in total war

With Japan's invasion of China in 1937 and the start of the Pacific War in 1941, the explicitly ideological conception of the police that Matsui and others had developed in the mid-1930s became anchored to the police's task to secure social and economic resources for the war effort. Indeed, by the 1940s officials saw war mobilization – both material and spiritual – as the police's primary mode of mediation between society and state. For their part, the Special Higher Police (Tokkō) presented their goal as no longer simply to suppress political threats to the state but now as to actively “disseminate the imperial way” (kōdō no senpu) and the “spirit of our national founding” (chōkoku no goseishin). Parallel to the Tokkō's ideological efforts, the newly created “Economic Police” (keizai keisatsu) in 1938 were to mobilize material resources for the war effort, further intensifying the police's interventions into social life.¹⁰⁹ Pamphlets and books were published to instruct police officers on the new “economic” policing functions.¹¹⁰ For example, the 1938 Economic Police Reader explained that the Economic Police was a force tasked with “advancing the development of the state economy and national economy” by limiting needless consumption and expanding production of necessary resources for war.¹¹¹ Thereafter, updated Economic Police readers were published in response to the changing circumstances of the war effort, complimented by a continuous series of pamphlets that outlined specific information related to such things as price controls, rationing resources, etc.¹¹² And, if Matsui portrayed the Japanese police as ideologically transforming into “the police of eight corners under one roof,” the Economic Police's specific interventions were explained as contributing to the material construction of the Greater East Asia Co-Prosperity Sphere (Dai tō-a kyōeiken).¹¹³ Such messages were communicated directly to the population as well. In forums such as the Osaka Economic Police's Decisive War Exhibition in 1944, for example, the police conveyed that the “the activities of economic police are to prepare and strengthen the resolve on the national home front” in order to “win the holy war (seisen).”¹¹⁴ We see here how an earlier method to propagate the “national police” to the masses in the 1920s – i.e., public exhibitions – was used once again to communicate the police's wartime mandate to assist in mobilizing material and spiritual resources.

This new formulation of the police idea after 1938 then brings our analysis full-circle, and prompts us to ask: how far do the police's wartime mandates depart from how officials had earlier understood the police's essential function? We often think of total war mobilization as an exceptional state, distinct from conditions before and after. And in many ways, it was.¹¹⁵ However, although the police's social interventions were invested with greater importance after 1937, I would argue that this was not qualitatively different from their original mandate, but only a matter of degree. As we saw, in the 1870s Kawaji Toshiyoshi understood police power as a way to shore up the “vitality” of a state and to convey its strength to foreign powers. Once the socio-economic conditions of Japan had become complex by the 1910s, police officials responded by proposing to delegate their powers to civilian groups under the banner of the “national police” in the 1920s, with mixed results. By the 1930s, this notion of dispersing police power into society was yoked again to imperial ideology – symbolized in the “Emperor's Police” – to be used to mobilize the population for war by the end of the decade. In each formulation, the underlying function of police power continued to be the mediation between state and society, only with particular emphases shifting in response to changing socio-political conditions. And thus it was only natural that by 1937 the police became one of the central mechanisms for the state to prepare the empire for war. If anything, total war conditions brought the police mediations between state and society into sharper relief.