The File System as a Logistical Medium

The file system in general is both material and immaterial. It is a calculated method and procedure, including logistics and a standardized classificatory system, that organizes the circulation, storage, and retrieval of documents. It is rooted in the larger imperative of the economy of efficiency and speed. The replacement of bound books and other horizontal document storage and retrieval media with vertical files early in the twentieth century enabled rapidly expanding business organizations to increase their capacity for holding and making practical use of more documents.Footnote ¹¹ The file system also entails a host of material things, including papers, documents, inscription tools and forms, the filing furniture, the index card boxes, the physical space for storage and for the pathways (dōsen) of document circulation, and workers. The system is therefore a dispositif, as defined by Michel Foucault as “a thoroughly heterogeneous ensemble consisting of discourses, institutions, architectural forms, regulatory decisions, laws, administrative measures, scientific statements, philosophical, moral and philanthropic propositions—in short, the said as much as the unsaid,” and it resides on “precisely the nature of the connection that can exist between these heterogeneous elements,” whose formation “has as its major function at a given historical moment that of responding to an urgent need.”Footnote ¹²

The central issue at stake here is how to foreground the emergent semiotic and medial capacity of the file system, which not only participated in but also materialized within the historically and politically charged context of the postwar PPO. The file system historically served as a linchpin by facilitating a “translation” between the economic rationality of efficiency and the political rationality of democracy, an interaction that ultimately ushered in a liberal mode of control and management.

Recent advancements in the anthropological study of bureaucracy have enriched both descriptive and analytical dimensions by elevating the role of material documents beyond their mere referential and semantic functions. This shift incorporates them into an ethnographic examination of bureaucratic practices. However, this development should not be construed as a mere expansion of ethnographic actors to include non-human elements, but as a significant analytical shift in understanding power and governance. Matthew Hull’sFootnote ¹³ inaugural ethnography of an Islamabad urban planning office is a case in point. While he delineates the material mediation of the everyday bureaucratic encounter between public servants through a variety of “graphic artifacts”—files, maps, lists, petitions, and tickets—his analysis posits government as the effect of forces and capacities of relations among the participating elements and resists the deduction of inscriptions and markings from what is postulated as a rational structure or system of bureaucracy. It effectively eschews a facile political functional critique (such as the question of whether or not a policy or practice is working effectively) and stays at the material and immanent level where graphic artifacts shape and participate in a particular configuration of people, ideas, and things as government.Footnote ¹⁴

A conceptually analogous trajectory can be found in the study of judicial bureaucracy, where the notion of “law” is construed as emergent through and as relations among humans and non-humans. Of particular salience is the circulation of legal documents, a process that not only literally “makes a case,” but also engenders evidence, constructs legal veracity, and safeguards judicial impartiality.Footnote ¹⁵ In Kyle McGee’sFootnote ¹⁶ lucid reading of Latour’s actor-network theoryFootnote ¹⁷ on jurisprudence, the force of law is understood as residing in, and contingent upon, its passage through variegated modes of material and technical mediation, as well as the indexical networks thereby generated.

As Alain Pottage observes, Latour’s material-compositionist approach to law shares an “old kinship” with Foucault’s conception of dispositif, law understood as an ensemble of heterogeneous elements assembled to produce a particular effect.Footnote ¹⁸ Likewise, in media studies, scholars of “cultural techniques”Footnote ¹⁹ seek to deconstruct super-organic entities such as “law” as an ensemble of material and technical mediations, or “ontic operations that precede concepts.”Footnote ²⁰ Studies of cultural techniques recognize the originary hybridity of people, things, and media in elemental forms of actions such as processing, cutting, plowing, opening, folding, enclosing, and transferring, since none of them can be achieved by humans alone.

Vismann explains that the concept of cultural techniques not only “un-blackboxes” law as concrete actions entangled with things and media, but also offers a compelling critique of the human monopoly on “sovereignty.” She contends, “The default positions of media and things that set cultural techniques into motion contradict a legally sanctioned, and thereby particularly widespread, notion: namely, the claim that only the subject can carry out actions and rule over things.”Footnote ²¹ The file system can thus be understood as a logistical medium. Vismann’s media archaeological analysis of “files” illuminates how the mythic origin of the power and authority of law in fact derives from the medial effect of files. In reference to the German translation of “files” (Akten), which does not distinguish between their form and content, Vismann points out the impossibility of separating the form and the content of the files and, therefore, of defining what it is as a discrete entity, and famously argues that “files process the separation of the law into authority and administration.”Footnote ²² Accordingly, she argues that the origin of law is not in the mythic past but in the present moment, on the surface and in the spatial milieu of mundane paper transactions.

Echoing Vismann’s media-technological approach to files—wherein she strategically collapses the distinction between their materiality and function as recording devicesFootnote ²³—Alexander Klose’s conceptualizes “containers,” both as material artifacts and epistemic concepts that function as powerful media-technological apparatuses.Footnote ²⁴ Serving as the linchpin of logistical systems—which are, in essence, the science of controlling the movement of goods and people—containers act as media to “transform the world into a moving warehouse and arrange it in the mode of standardized movable spatial units, switched processes, and clocked times.”Footnote ²⁵ Its operational principles of modularity, mobility, and standardization, which are the functions of cultural techniques, effectively “contain” logistical entropy. By extension, it also serves as a material mode of governance and control that profoundly impacts modern social life on multiple scales, encompassing everything from transportation and global supply chains to modular architecture, computer information systems, and beyond. The analytical lens of cultural techniques thus shifts away from content or governing rules, and instead focuses on their historically contingent, media-specific material operations. It is concerned with the semiotic operation of their indexical connectivity with and distinction from other entities, which then bring forth institutions, ideologies, social formation, and other normative underpinnings and epistemic structures of society that often dissimulate their material and technical origins.

In the process of implementing the file system, public prosecutors and clerks often conceptualized it in terms akin to large-scale infrastructure—railroads, pipelines, or underwater cables—that serves to connect and network both individuals and objects. One common metaphor was an irrigation system (Japanese irrigation relies on a complex system of canals). One officer from the Tokyo High PPO,Footnote ²⁶ for example, offered the following advice to his fellow participants in a workshop on how to divide filing tasks: “You need to figure out exactly who makes which document, when (in which order) to make them, and on what basis documents and cards can be produced. Then, you need to set forth the flow of work and think out from which part of the flow you should draw water.” The logistical trajectory of the file as an “irrigation system” was built into the office’s layout, and configured and channeled the workflow and the social relations in the PPO as part of its administrative infrastructure. John Durham Peters conceptualizes such channels as “logistical media”: “[T]he infrastructural role of media, which includes storage, transmission and processing systems. Logistical media are media that don’t necessarily have any content, but they coordinate or orient us in space and time. Logistical systems are also naturalized in that they do not appear political or contentious in any way….”Footnote ²⁷

Such a technical medium, a logistical form of power, anonymously orients and organizes people and things in an order without appearing to be political or contentious, and it thus underwrites a technology for “the conduct of conduct,”Footnote ²⁸ an originary form of liberal governmentality. Logistical media are thus a specific form of “the government of things,”Footnote ²⁹ whose power derives neither from rules nor disciplines, but from “disposition.” Government, for Foucault, “is not a matter of imposing a law on men, but of the disposition of things, that is to say, of employing tactics rather than laws, or, of as far as possible employing laws as tactics; arranging things so that this or that end may be achieved through a certain number of means.”Footnote ³⁰ The file system in the postwar PPO served as a device of control less over people than over the environment, as in the apt metaphor of an canal, a “milieu,” “ensure[s] a certain allocation of people in space, a canalization of their circulation, as well as the coding of their reciprocal relations.”Footnote ³¹

We will see that the engineering rationality of the file system that derived from the factory assembly line came to be implanted into the apparatus of the troubled postwar PPO, both as a government office and as a judicial authority disposing of cases. The mechanism of power that underlay the lofty authority of law and government was in fact made of paper.

The Postwar PPO

The Japanese public prosecutor, established in 1982, was modeled in part on the thirteenth-century French Procureur de roi. Footnote ³² Serving as a quasi-judicial career civil servant, the prosecutor represents the public interest, akin to the U.S. model of “United States v.…” or “The State v.…” Until 1947, Japan’s criminal procedure was non-adversarial, and the courts and the Public Prosecutor’s Office, then called Public Prosecutor’s Bureau, functioned as administrative branches under a unified Ministry of Justice. In court, the judge and prosecutor sat together in robes of slightly different hues looking down from a bench on the accused and defense attorney. Mirroring France’s Juge d’instruction, prewar Japanese courts oversaw investigations via pretrial hearings, and also wielded authority over the judicial police, who were the main actors in criminal investigations.

By the 1930s, Japan’s Public Prosecutor’s Bureau had amassed significant authority, partly to quell internal and external threats amid its colonial expansion. This period witnessed increasingly authoritarian practices, later deemed abuses of power. After 1945, the Allied Forces instigated a comprehensive governmental overhaul aimed at “reviving and strengthening democratic tendencies.” The Bureau, criticized for “secret inquisitions,” became a focal point of reform. Under the postwar judiciary restructuring, courts gained independence from the Ministry of Justice, culminating in a judiciary led by the Chief Justice of the Supreme Court.Footnote ³³ Consequently, the Bureau was reconstituted as the PPO, under a new law.Footnote ³⁴ Though the PPO was organizationally within and remained part of the Ministry of Justice, it was designated “independent.” Officially, the Minister of Justice oversaw the Prosecutor-General but, ostensibly, not individual prosecutors or cases. With an unmistakable trace of the U.S. Bill of Rights, Japan’s 1946 ConstitutionFootnote ³⁵ now foregrounded due process and procedural justice over substantive justice. Subsequent revisions to foundational laws like the Code of Criminal Procedure were aligned with these new constitutional principles, rooted in the Anglo-American adversarial system.

Efficiency as Democracy and Democracy As Efficiency

Postwar democratization of the justice system also meant aligning judicial process with the constitutional imperative of protecting civil rights through a speedy trial, which required expediting processing. This is where efficiency figured in as a constitutional issue in the postwar reconstruction of the justice system, and where scientific management of various sorts—including the humble file cabinet for papers—found its way into the administration of justice. The reformers’ view of the inescapable linkage between non-democratic practices and inefficient documentary practices was clear. The GHQ’s Public Administration Division (succeeded by the Civil Service Division) formulated and secured adoption of the Civil Service Law of 1947, which declared that as its purpose “to assure the people of democratic and efficient administration of public service by establishing basic standards….”Footnote ³⁶ In his 1946 memorandum addressed to Brigadier General Courtney Whitney (chief of the Government Section, GHQ), Lieutenant Milton J. Esman of the Public Administration Division in the Government Section (a former member on the United States Civil Service Commission) made a forceful recommendation that GHQ issue “guidance on the proper techniques to employ”Footnote ³⁷ in modernizing the Japanese bureaucracy, which, he recognized, was still impaired by “the major bulwarks of feudal and totalitarian Japan.” Existing Japanese bureaucrats and officials were “incompetent to manage a modern democratic society,” because “modern democratic government requires a democratic and efficient public service.”Footnote ³⁸

The PPO’s file system was not a mere carryover from prewar Taylorist principles.Footnote ³⁹ It is better understood as part of the postwar “re-introduction” of scientific management, initially through GHQ and its Civil Communication Section initiatives. Japan adopted Taylorism and scientific management principles early on, starting in the 1910s. However, as prewar Japanese proponents of Taylorism recounted, until the postwar era, excepting certain isolated practices, there was little robust engagement with these principles within the realm of office administration, particularly in white-collar and governmental sectors.Footnote ⁴⁰ Ueno Yōichi, often regarded as the pioneer of Japanese scientific management, observe, “As was the case in the States, in Japan too, initially, the focus was primarily on researching production efficiency. Office work was thought to be something that would somehow get sorted out if left to the office workers…. The issue of administrative efficiency in Japan began to be rigorously studied only after the end of the war.”Footnote ⁴¹

The American literature on office administration management, re-introduced to Japan in the postwar era, was inflected with the language of cybernetic principles such as “control, feedback loop, and communication.” They no longer constituted a mere toolkit for enhancing labor-time efficiency, but shifted to embodying an imaginary of an interconnected, self-regulating “system” (shisutemu). This marked a conceptual transition from viewing office work as a “necessary evil” (hitsuyō-aku) ancillary to business operations, to recognizing it as the “brain” or “central nerve” essential to organizational success.

Some were lucky enough to witness those advanced office management practices personally. In one of the famous study missions to the United States orchestrated by the Japan Productivity Center—established in 1955 as a collaborative venture between Japan’s private and public sectors—a delegation of thirteen eminent corporate administrators and management experts toured companies across seven major cities to scrutinize advanced office management practices. One participant noted how “office” in the American business organizations is “the focal point of the company, the center of communication, facilitating interdepartmental coordination through record processing to enhance operations.”Footnote ⁴² The mission’s report highlighted emergent technologies like the Punch Card System, Electronic Data Processing Systems, Information Retrieval, and the Management Information System, through which paperwork was transformed into new modes of existence as “data” and “information.” Most notably, they glimpsed the automated future of office administration that would become possible through such new technologies coordinating and communicating with each other.

Taylor’s maxim that “efficiency is the hope of democracy”Footnote ⁴³ was reinterpreted to cultivate “modern” social relations in white-collar environments, free from the personal authority of supervisors, prewar feudalism, abuse, and servitude. The “feudal” was to be supplanted by democratic techniques, giving way to liberal governance rooted in rationality and science. This new vision of political freedom was at the core of new designs for liberal capitalism, more “open” markets, and more “free” economic actors that the American reformers had in mind when envisioning Japanese democracy and its citizens. Coinciding with this was the notion of liberal—or distanced, indirect—rule. As economic reconstruction became intertwined with democratization, the term “scientific management,” marred by the connotations of the prewar factory operations, was supplanted by “productivity” (seisan-sei). Governmental and quasi-public entities like the Japan Productivity Center championed the idea that administration should be driven by organizational prowess rather than individual personalities,Footnote ⁴⁴ and that office work should shift “from people to documents, from documents to files.”Footnote ⁴⁵

Steel filing cabinets also glowed with the aura of modernity. Traditionally, records were organized in commercial ledger-like bound books, commonly called daifukuchō, literally “a book of great fortune and prosperity.” One entered a written notation in a bound book based on a chronological, alphabetical, or numerical order—the order allowed the case to be found again later. Replacing daifukuchō with the file signaled the “democratization of paperwork,” as stated on a banner popular at that time (see figure 1). The replacement of hefty bound books with the sleek file cabinet and index cards, and democratizing paperwork through the file system, were meant to liberate official information from the arcane and personal authority and quasi-ownership of the book, since personal ownership of information was often at the core of the pre-Occupation system. Files only belonged to no one person. The file system as scientific management was thus imbued with moral value, not unlike the vision of a truly democratic system invested in cybernetics during the Cold War era in Europe and the United States, in which the promise of “information” confounded control with democracy.Footnote ⁴⁶

Figure 1. “The Democratization of Office Work: From Daifukuchō to the Filing System” [daifukuchōshiki kara fairingu shisutemu e: jimu no minshuka] Nōritsudō 17, 11 (1950): 13–14.

Democratization of the PPO

What survived the sweeping reorganization of the PPO by the Allied Occupation was the so called “principle of the united body of prosecutors” (kensatsukan dōittai no gensoku) (see figure 2).Footnote ⁴⁷ Derived from the French system, “règle de l’indivisibilitè du parquet,” meaning “the rule of the indivisibility of public prosecutors,” this refers to the legal and political imaginary in which “the public prosecutors are organized in the nationally united and hierarchical collective, with regards to prosecutorial administration, they are to act as an indivisible whole.”Footnote ⁴⁸ The public prosecutor is thus both one and many, concrete yet abstract, acting as an individual professional, and at the same time embodying the abstraction of the entirety of the Prosecutor’s Office as one. It is an ultimate aporetic body politic.

Figure 2. The difference between the regular government office administration and the dokuninsei prosecutorial administration. Komura, Yasuhide, Kensatsuchō ni okeru jimu kanri no mondaiten, 29.

One side of the principle is called dokuninsei, which means that each and every prosecutor holds the full power of the PPO and ultimately represents the nation, with no requirement to consult with supervisors or even peers. At the same time, individual prosecutors are required to act as one body and intention within the office in its hierarchical command structure.Footnote ⁴⁹ This is why the PPO is referred to in the singular; only buildings and physical offices can be pluralized. So, the other side of “principle of the united body of the prosecutors” is its mandate of absolute obedience of the subordinate to the superior’s command (jōmei kafuku), a clearly vertical unity that is called “one taut thread.” In the mid-1950s, Kenshū, the PPO house organ, carried several prize-winning essays on the theme of the democratization of prosecution, authored by both public prosecutors and public prosecutors’ assistant officers. The essays commonly began with critical reflection upon the prewar PPO as “fascist” (kensatsu fassho, prosecutorial fascism), arbitrary, unconstrained, elitist, and cliquish—a hotbed of state abuse and authoritarianism. The authors argued that the remedy for rooting out any remaining vestiges of all this was (democratically) “rationalizing and modernizing” human relationships in the office.

Another essayFootnote ⁵⁰ pointed out the persistent, blatant hierarchy within PPO workplace organization and underscored how assistant prosecutors, in particular, had high expectations that scientific and rational office procedures would liberate them from archaic patron-client relationships, known in Japanese as “oyabun-kobun,” and hoped that superiors would learn how to talk “democratically” to their subordinates. Yet another essay attributed the PPO’s detachment from the rest of the society, which was rapidly democratizing, to its continued reliance on outdated office tools. In the 1950s the PPO was still using traditional implements like brushes and “koyori,” a type of hand-twisted paper string used for binding documents. The essay argued that modern office technologies—such as Japanese typewriters, contemporary accounting forms, card indexes, and filing systems—were not merely tools for efficiency but crucial for achieving a democratic workplace.Footnote ⁵¹ This emancipatory potential that the author attributed to technology intriguingly aligns with the democratic aspirations foundational to Norbert Wiener’s development of cybernetics.Footnote ⁵²

Aporia of the Postwar Prosecutorial Administration

The postwar reorganization of the PPO administrative structure rendered untenable the “principle of the united body of prosecutors.” By convention, prosecutorial administration itself came to be bureaucratized and was divided into two lines of work. One was the prosecution itself, consisting of investigations and trials (kensatsu-jimu). The other was prosecutorial administration (kenmu-jimu) to be carried out collectively, independent of any individual prosecutor’s responsibility: receiving cases, cataloging evidence, custody procedure, and so on. The principle of the unity of the state in executing prosecution (which necessarily entails the support functions), meant that support work could not be allowed to be seen as separate from the underlying legal fiction that the public prosecutor was exercising all his appointed powers and duties. As an office of “individual independence,” the Public Prosecutor’s Assistant Officers and the Public Prosecutor(s) were legally deemed one body with no division of authority between them.

And, yet, as a site of office work, the distinction between the line functions directly involved in the core prosecutorial activities and the support functions of the prosecutorial administration emerged quite clearly. Figure 3 shows a typical organizational structure of a mid-to-large PPO in which the organization is divided by the types of administrative work. Departments are headed by the senior public prosecutor, whereas the “bureau” (kyoku) is headed by an assistant officer, which means that there is no prosecutor directly supervising its day-to-day work. This clearly did not adhere to either the Public Prosecutor’s Law or the fiction of prosecution as a process embodied in one official. It not only completely separated prosecutorial administration and support, but the Bureau of Secretariat, which the law stipulated should fall under the authority of the Public Prosecutor, was instead under that of the assistant officer heading it.

Figure 3. A common organizational structure of the mid-to-large size public prosecutor’s office. Komura, Yasuhide, Kensatsuchō ni okeru jimu kanri no mondaiten, 2.

Komura Yasuhide,Footnote ⁵³ himself a senior public prosecutor, critiqued the prevailing office hierarchy, arguing that the Public Prosecutor’s “managing his or her staff” was less an expression of rational bureaucratic structure than a manifestation of retrograde elitism. This, he contended, resulted in an undemocratic status distinction between public prosecutors and assistant officers. Komura raised concerns about such an organizational framework and labeled it not only irrational but also socially and ethically problematic. Such an unsubtle display of the demarcation and stratification between public prosecutors and assistant officers, as well as between prosecutorial functions and what were deemed “supportive” roles, he argued, were corrosive to the democratic ethos and could demoralize assistant officers. KomuraFootnote ⁵⁴ stressed that it was imperative to forestall the opening of any irreparable fissures between public prosecutors and their assistant officers that would leave them permanently separated like immiscible “water and oil.” For Komura, figure 4 represents an ideal organizational structure of the PPO.

Figure 4. The ideal public prosecutor’s office organizational structure. Komura, Yasuhide, Kensatsuchō ni okeru jimu kanri no mondaiten, (1963). 11.

As will be demonstrated in the next section, the humble file system turned out to offer a tangible manifestation of Komura’s vision for reforming the prosecutorial office, even if it did not actually succeed in diminishing the entrenched social distinctions. Its circulatory architecture was designed to suture the organizational rift between prosecutors and assistant officers and to institute a non-hierarchical, flat network process that effaced traditional distinctions—both administrative, between line functions and support roles, and social ones between public prosecutors and assistant officers. All were pulled into the circulatory, not hierarchical, path of the files.

The Implementation of the File System

This section explores the design and practical implementation of the file system through a case study of the Niigata District Public Prosecutor’s Office. It focuses on the initial phase, where local law enforcement referred police dossiers to the district PPO. From there, a bundle of documents was transformed and stabilized into a “case” that was prosecutable, and was thus made “ripe” for use, as Latour puts it,Footnote ⁵⁵ as it traveled through a designated circulatory channel.

In transitioning to the file system, the previously used bound books were transformed into individual file cards,Footnote ⁵⁶ generally sorted into three types: (1) Case Card, (2) Receipt Card, and (3) Name Card (and a Statistics Card was made later in the process) (see figure 5) at the initial case intake process. The system also introduced a “Registry Sheet” (seiri genpyō) that summarized the entire record of a suspect, from initial receipt of the dossier to case closure or suspension. This eliminated the need to consult multiple offices for such information.

Figure 5. From bound books to cards.

Figure 6 illustrates the logistical trajectory of the dossier. The police dossier goes to the Case Office in the (Support) Prosecutorial Administration Section. Within the Case Office, there are four desks—the “Case Distribution Desk,” “Case Intake Desk,” “Evidentiary Materials Intake Desk,” and “Statistics Intake Desk”—where the initial paperwork is generated. The formal procedure of case-making begins with the creation of a Registry Sheet; two cards (a Name Card and a Receipt Card) that are mobile and move between offices with the Registry Sheet and the dossier; and the “Statistics Card,” a semi-mobile card that travels with the other documents part way through the circuit before returning. In addition, two stationary Cards (an Evidence Custody Receipt Card, and a Criminal Record Card) are created in the appropriate office and stored there.

Figure 6. The passage of the filing system turning the dossier to a case file at Niigata District PPO, 1949.

Within the Case Office, the dossier’s journey begins at the Evidentiary Materials Intake Desk, where the Evidence Custody Receipt Card is created. This card awaits the subsequent arrival of the Registry Sheet for completion. The dossier moves to the Case Distribution Desk and then to the Prosecution Review Officer. Next, it reaches the Case Intake Desk, where two mobile cards—the Name Card and the Receipt Card—are generated along with the Registry Sheet. These documents accompany the dossier to the Statistics Intake Desk, where a Statistics Card is created based on the Registry Sheet. This card is promptly sent to the Statistics Office. The dossier and the two mobile cards, and the Registry Sheet return to the Evidentiary Materials Intake Desk to complete the Evidence Custody Receipt Card, which then moves to the Evidentiary Materials Office for filing. The final leg involves the Criminal Record Card Office, where the Crime Card is generated and filed. The dossier and associated documents undergo a last inspection at the Prosecutorial Affairs Section, from where the Registry Sheet and the dossier go to the assigned lead prosecutor, and the mobile cards go to the Case Distribution Desk for archiving. The Registry Sheet is then archived at Registry Sheet Office, while the dossier stays with the assigned lead prosecutor.

Unlike traditional bound books, the file system’s sheets and cards embody principles of “theoretical kinematics,” functioning as a dynamic system in which the rearrangement of one element necessitates the reconfiguration of others.Footnote ⁵⁷ This mobility extends beyond mere bureaucratic shuffling; it constitutes a logistical medium that interacts with the administration of justice within the framework of “democratic” reform. As the dossier traverses the designated circuit in temporal sequence, and is inscribed at each office, it self-reflexively indexes and records its own path taken. This stabilizes the case into a legally robust object that counts for (now) unquestionable facts produced by the limitation of “facts” to those properly documented, inscribed, and circulated in a self-referential logistical channel.

The stationary cards—the Crime Card and the Evidence Custody Receipt Card—are additional material-indexical traces of the circuit navigated by the mobile cards and the Registry Sheet. For example, the Crime Card, generated on the basis of the dossier and the two mobile cards, not only triangulates the facts they contain, but also indexes their circulation. It records its own trace by having required information inscribed on it, and thus physically “proves” that the dossier, the two mobile cards, and the Registry Sheet traveled through the Crime Card Office; otherwise a Crime Card could not have been compiled. Similarly, the existence of the Evidence Custody Receipt Card, and thereby its legal validity, is contingent upon the return of the Registry Sheet to the Evidentiary Materials Intake Desk. In other words, this circulatory self-referentiality grounds the file system’s own authority as a universe of legal documents. In essence, this is how a “case” materializes. Each card, sheet, and dossier creates a system of circular causality, being mutually indexical and reinforcing in terms of traces, routes, times, and locations. This interweaving of office space, people, documents, and file cabinets essentially transforms the whole into “one giant index card box.”Footnote ⁵⁸

This is a semiotic process of recursivity.Footnote ⁵⁹ The file system stabilizes itself through the multiple, cross-referencing, triangulating iteration of the case and its facts. Recursivity literally materializes the process in due process; circulating papers and files were the material form of constitutional due process in the democratic reform of postwar Japan, and an example of Latour’s insight that “law is itself its own metalanguage.”Footnote ⁶⁰ The file system thus operates as the law’s logistical medium, self-recording its mundane due process in order to ensure the lofty due process famously guaranteed by the rule of law. It forms a self-referential and self-generating process with no external command and control, an essential semiotic condition of a system of administering criminal cases in particular, but also of “liberal government” in general.

The file system redefined what a “job” is. Just like the Taylorist assembly line, a “job” was turned into a task that was clearly divided, atomized, and linearized in a larger scheme of things. Each job position was designed to be autonomous from any personal ties of obligation or benevolence that might have survived from prewar social hierarchies at work. Instructions from the Judicial Affairs Office emphasized a rational division of labor to avoid any wasteful redundancy. To put it another way, as the representative from the Tokyo High Prosecutor’s Office did, each person must stop meddling with others’ work and mind their own, and deepen their respect for others, and others’ respect for them, through each person taking responsibility only for their own duly assigned tasks.Footnote ⁶¹

The diagram in figure 6 illustrates the shift from human-centric to document-centric control within organizational logistics. Gone are the days of relying on a “living dictionary” (ikijibiki), on individual staff members acting as human repositories of institutional knowledge. This transition also reconfigures work ethics, moving Weber’s concept of Lebensordnung,Footnote ⁶² or an ethical life order grounded in individual interiority, to a material documentation system. Such indirect control of people via the file system is a form of liberal governmentalityFootnote ⁶³ that intervenes in and arranges the milieu (the environment as the media), not workers’ minds or bodies. The figure 6 diagram is to be read as a map of a territory activated as, and transformed into, a milieu by the file’s logistical system.Footnote ⁶⁴ It represents not a form of worker consciousness but an unintended apparatus for the conduct of conduct, which aspires to govern people not through injunctions, rules, ethics, or morals, but rather the milieu activated by the logistical channel.

Obviously, the diagram is neither a representation of empirical reality nor a structural model. Contrary to expectations, it coexisted with hierarchical practices. It did not supplant the traditional “ringi” decision-making process—a bottom-up hierarchy for review and approval—nor did it threaten the existing structure of authority and social relations. One might consider how the notion of “flow”—the seemingly smooth transit of paper—obscures underlying tensions and power dynamics. Yet, the diagram must not be dismissed as a mere “rice cake drawn in a picture,” to invoke a Japanese idiom. In the postwar period, when the very essence of “democracy” was radically up for grabs, a diagram like this one was imbued with aspirations, and pointed toward diverse future visions.

To return to the Niigata case, this diagram represented the organization of work through a bird’s-eye view of document circulation (see figures 7 and 8).Footnote ⁶⁵ It allowed both prosecutors and assistant officers to see their own position and relationship with others in the organization less as a hierarchy than as a network. In stark contrast with the familiar Taylorist organizational chart that highlights hierarchies governed by “command and obey,” this diagram emphasized circulatory networks. Managers might be nominally higher, but in practice even Chief Prosecutors or the Section Chief of the Investigative Affairs Office were represented as mere nodes in a network of circulating documents, diagrammatically reducing the relevance of formal or personal status, akin to how an assembly line component was indifferent to the worker installing it. It presented PPO workers with a new way of seeing themselves and their relationship to other works situated in the circular network rather than in the hierarchical structure.

Figure 7. Niigata District PPO Organizational Chart. Hōmufu Kenmukyoku, Fairingu shisutemu ni tsuite, 243.

Figure 8. Niigata District PPO Organizational Chart from the Point of the Paper Flow. Hōmufu Kenmukyoku, Fairingu shisutemu ni tsuite, 236.

In implementing the file system, the workspaces of both the Niigata and Sendai District PPOs were redesigned to optimize document flow. Niigata physically divided its Case Reception Office into two distinct units: one for the initial receipt of dossiers and the other for their subsequent processing, thereby mitigating the risk of confusion and filing errors.Footnote ⁶⁶ Sendai was different: constrained by spatial limitations, it expanded its operations across multiple buildings and even hallways. This spatial configuration was intended to minimize delays in circulating documents from one office to another within the file circuitry.Footnote ⁶⁷

The file’s logistical channel Footnote ⁶⁸ bridged the organizational divide between the support for prosecutorial administration and that of the main, professional prosecutorial work. The circuitous mechanism physically spanned the PPO’s organizational aporia, enacting a “unified” judicial entity and a hierarchical administrative body. The Tokyo District PPO exemplified this dynamic. Just like its smaller-scale counterparts in Niigata and Sendai, Tokyo reconfigured its office layout to rationalize document flow. When the Public Prosecutor’s Law took effect in May 1947, the Case Office functioned as a subsection within the Support Prosecutorial Administration Section of the General Affairs Department, which encompassed all prosecutorial functions aside from trials and investigations. By July 1948, to align with the newly instituted “position-classification system,” the Case Office was elevated to an independent section with three designated subsections for reception, disposition, and appeals.Footnote ⁶⁹ That September, the General Affairs Department was rebranded as the Support Prosecutorial Administration Department,Footnote ⁷⁰ thereby achieving organizational parity with the departments responsible for investigation and trials. This restructuring formalized the division between core and support administration.

On 1 March 1949, in anticipation of the file system’s pilot launch scheduled for the following month, the Support Prosecutorial Administration Department was subdivided into a Case Section, a Disposition Section, and a newly established Registry Slip Section.Footnote ⁷¹ The latter was further divided into two subsections: one responsible for document production, and another for archiving and calls for retrieval. In January 1950, the Registry Slip Section was relocated to the Trials Department,Footnote ⁷² indicating its transition from a “support” function to a “main” prosecutorial role. Within the Trials Department, the Registry Section comprised three subsections: the first oversaw planning, statistics, and the entire filing process from the production of the file to its storage; the second managed file lending for trial-attending prosecutors and disposition of complete case records; and the third facilitated through coordination with the other two subsections.

The Registry Slip traversed two departments previously divided by function and official status: prosecutorial administration led by prosecutors, and “support” administration by assistant officers. The logistical pathway materialized the principle of prosecution indivisibility, which was not reflected in official “command-and-obey” vertical organizational charts. The flat file system’s diagram subtly renegotiated different kinds of hierarchies. It established a temporally ordered, mutually exclusive network of tasks, which sharply demarcated Person A’s job from Person B’s job. This clarity made it possible for assistant officers to voice concerns—albeit respectfully and without being impertinent—about prosecutors hampering file flow. At one file system workshop, some of the senior assistant officers reported that prosecutors misplaced files frequently. A Niigata officer, using honorifics, noted: “I fear that it might sound disrespectful for me to say this, but current prosecutors often lose (nakusaremasu, an honorific form of ‘to lose’) important records. Every time the record disappears, they ask us to search for it. It gives us a pain in the neck.”Footnote ⁷³

The file system also obviated any private ownership of documents across ranks, and rendered the entire operation transparent. This transition posed significant challenges particularly for public prosecutors habituated to keeping documents and files in their own desk drawers. A Yokohama District Office assistant officer attributed the difficulties in switching to the file system in part to the fact that “public prosecutors have an attachment to bound books in the daifukuchō system, and have emotional difficultly letting go of the old system.”Footnote ⁷⁴

The introduction of memos Footnote ⁷⁵ (derived directly from the English word), a writing genre that was new became a focal point of prosecutorial resistance to the new file system because it was time-consuming (see figure 9). In the prewar period, one public prosecutor normally handled a case from inception to conclusion, making private notes known as tebikae—for personal use. The postwar case influx, however, necessitated personnel flexibility and interchangeability for speedy trials. Consequently, a trial’s progress had to be communicated via memos, especially to those handling appeals. Authored by the prosecutor and inserted into the case file, these memos supplanted tebikae, transforming them into circulated and public (shared) documents visible “to” the file system. While the hierarchical “command and obey” structure within the PPO remained intact, the file system subtly integrated prosecutors into a more decentralized administrative network.

Figure 9. “Memo” Sheet (Sample) from Kyoto District PPO. Hōmufu Kenmukyoku, Fairingu shisutemu ni tsuite, 268.

Guilt by Indexicality and the Birth of the Delinquent

Once the file system transformed a bundle of papers into a case file, it entered into a dense network of indexical Footnote ⁷⁶ relationships of traces, a material-semiotic nexus, whose intensification promised to isolate and identify suspects in new ways. The PPO leaders’ aspiration was to build an ever-growing paper-based database. The primary indexical function of the Name Card was to locate the folder corresponding to the named suspect or accused. But the system acquired another utility, as theorized by the Judicial Affairs Office: individuals linked to prior cases could serve as potential suspects in subsequent investigations:

In the file system, paper evolved into a “virtual” medium endowed with a future tense, thus creating a nebulous universe of individuals linked to crimes or known criminals. The distinction between predictive analytics and the creation of a new “criminal-prone” category was, of course, tenuous. Unlike names stuck in bound books—essentially “dead letters”—Name Cards remained permanently alive for future use, ready to be activated to cross-index with other names and cases. Existing in a liminal space between past and future, Name Cards were both ever-present and virtual (potential), built into a growing network of indexical contingencies. Transcribing names—previous offenders, witnesses, and other associated individuals—from bound books into Name Cards to file in the system materially opened a universe of latent delinquency beyond a mere epithet; it became an actionable, dividing practice. These individuals were not recidivists or criminals yet, but they could potentially become so. And would not those in charge of public order be derelict in their duty to ignore or fail to act on such knowledge?Footnote ⁷⁸

The power at work here, while subtle, yielded concrete and unprecedented effects. Foucault notes that the use of individual index cards for criminal registers in nineteenth-century France materialized the techno-administrative expression of a form of power, which was “both individualizing and totalizing.”Footnote ⁷⁹ “Delinquency,” then, emerged as a perpetual state of “political observatory,” as individual index cards integrated into the existing card-index system allowed for their utilization by statisticians and sociologists, extending far beyond their initial use by the police. Foucault thus observed, “The appearance of the card-index system and the constitution of the human sciences are another invention that historians have taken little note of.”Footnote ⁸⁰

Such an alchemy of the file system marked a pivotal shift in the concept of “efficiency” from labor-time minimization to the maximization of surplus, that is, indexical traces that could aid present and future criminal investigations. Judicial Affairs Office leaders also posited that the “memo” could potentially lead to a treasure trove once it was fed into the file system automatically. In his address to the Deputy Chief Prosecutor’s conference on 16 May 1950, Takahashi advocated for producing concise memos for future investigations rather than wasting time producing non-evidentiary statements of mere archival value.Footnote ⁸¹ He thus told the Deputy Chief Prosecutors,

The new procedural law, modeled after the Anglo-American adversarial system that mandates objective evidence, provided an additional rationale for meticulously documenting the details of individuals involved in a case—be they suspects, witnesses, or associates—irrespective of the case’s outcome, for the purposes of the file system. Takeuchi Juhei, Prosecutor in the Supreme Public Prosecutor’s Office, highlighted this shift by citing a case in which the prosecution’s inability to corroborate the accused’s presence at the crime scene led to acquittal. Takeuchi reminded his audience that, unlike during the prewar era, when the absence of objective proof did not necessarily disadvantage the prosecutor, with the new procedure such an absence would benefit the accused. But with the new file system, evidence collected that did not result in conviction could serve as potent material for later cases. For instance, suppose one found a newspaper article which recounts that the accused told a reporter that he had left Tokyo and arrived at an alleged crime scene on the previous day. This would be valuable to the prosecution. In the old criminal system, evidence collected for each case was deemed to have no use after the case was over. Under the new law, which demanded objective evidence, materials from past cases had the potential to be important for new cases. News and magazine articles that seemed irrelevant to a case could become conclusive evidence. Takeuchi urged his audience to see the file system as a powerful method for organizing and storing such materials and their potential.Footnote ⁸³

The document’s expanded temporality—from encompassing only the past and the present to also incorporating the future—marked a substantive shift in the function of written records. These records were no longer only for assembling facts for a trial; they acquired an open-ended indexical value that might materialize as new data entered the system. When combined with stored information, this evolving corpus could generate new knowledge.

Flow: Like Water Flowing Downstream

Despite the GHQ’s belief that they had significantly weakened the power and authority of the PPO, its senior leaders recognized that the PPO had actually weathered the GHQ’s reforms. At least that is how they presented the PPO’s postwar standing to the file system workshop participants. Takeuchi assured the participants that the principle of prosecutorial indivisibility, the kernel of its organizational structure and ethos, had been intact since before the war. What had changed is that all the other state institutions had been unrecognizably decentralized, split into pieces leaving no trace of their prewar prototypes. Takeuchi said and that the PPO was “probably the only centralized organization left in Japan.” He continued:

Takeuchi underscored the continued relevance of the PPO’s centralized structure, while also recognizing the democratic exigencies of the new era, a reckoning that the use of overt force and authoritarian tactics was no longer sustainable. In describing this tension, between conforming to postwar democratic norms and aspiring to further fortify the PPO’s vertical organizational strength, Takeuchi asked, “Is there a way to truly utilize and operate such an organization?” He presented the file system as one way to do so. It eschewed force and instead resorted to the administrative technique of arranging and circulating documents. In a way, this transition from coercive methods to the logistical media that arranged and circulated documents aligned with Foucault’s concept of dispositif.

Diverging substantially from the GHQ’s original conception, and perhaps from the PPO workers’ expectations that the new system would encourage democratization, senior leaders envisaged a dynamic, national communication infrastructure wherein the Supreme PPO would automatically and continually receive data from regional offices, which would in turn regularly receive authoritative directives from the central office. In other words, senior leaders saw in the horizontal logistical channel of the file system a potent vertical network—a “center of calculation,” as Latour would put itFootnote ⁸⁴—capable of generating novel and powerful investigatory and prosecutorial knowledge, which would benefit the nation. This move, conveniently, also aligned with the principle of prosecutorial indivisibility, an ironic but calculated move on the part of senior leaders, who infused it with prewar vigor while simultaneously reimagining its postwar institutional ethos.

The “flow” in the assembly line analogy for the file system became recast as the (natural, law of gravity-given) flow over a waterfall. The Prosecutorial Administration Bureau Chief thus clearly recognized the power of the file system as a logistical medium, and emphasized: “It is about ‘keiro,’ the ‘channel, through which a document is produced, and is about its whole process until it goes into the filing cabinet. As is clear from the phrase, the nagare (‘flow’) of administration, to administer is to cause to flow, to keep flowing. It is as though the water flows from the high to the low; it must automatically flow. It must not stagnate or flow backward along the way.”Footnote ⁸⁵

Takeuchi Juhei, prosecutor in Supreme Public Prosecutor’s Office, explained to the file system workshop participants how the ostensibly horizontal work process of the file system could engender a “vertical,” centralized logistical channel running from the upper echelons to the lower tiers, and vice versa. Recognizing that this “vertical” work—comprising the transmission of files and progress reports in memo form to higher offices—constituted an additional layer of responsibilities for both prosecutors and assistant officers, Takeuchi argued that the file system automated this process, often unbeknownst to the workers themselves. To underscore his point, he posed a rhetorical question, implicitly answered by the file system itself: “It is crucial for all district and local prosecutor’s offices to promptly relay the progress of their cases to higher authorities. However, if an officer becomes too preoccupied with composing such progress reports, their primary responsibilities could suffer neglect. Although vital, these reports constitute ‘vertical’ tasks, in contrast to the main work. The question then arises: How can this ‘vertical’ work be seamlessly and automatically navigated through many hands just like water flows from the high to the low?”Footnote ⁸⁶ In his vision,

The label “division of labor -assembly line” (bungyōteki nagare sagyō) system was directly borrowed from the Taylorist shop floor management lexicon. The metaphor of nature’s uninterrupted water flow concealed the arduous nature of the labor involved in the system as well as the labor-management strife inherent in it. It also encoded a desire and will to re-establish the centralized authority and organizational robustness that characterized the prewar PPO and recommitted to the principle of the indivisibility of prosecutors.