Introduction

Police forces became the necessary by-product of nation-states around the globe, developed into highly bureaucratic and sometimes bulky organizations. The oversight of the police becomes a complicated, if not impossible, task, especially for members of the public without “professional” criminal justice training. The oversight of police misconduct could be complex in nature because of the intertwining of victims, police officers and complainants. Police misconduct is defined as the misbehaviour that results in the filing of a formal complaint, an internal investigation, or the imposition of disciplinary actions against an officer [1]. To resolve police complaints and allegations of police misconduct, some governments have deployed certain measures to independently investigate and resolve police complaints, while others have developed oversight agencies to monitor and oversee investigations by police.

This leads to our puzzle: what kind of institutional setting would better resolve the problem of police misconduct and complaints? Should police be checked by their own counterparts, for the sake of the expertise and knowledge they have, or should the police be checked by civil groups or other citizens, to better enhance the accountability of the force to the public? Scholars have built several models to comprehend this complex issue. They are the internal affairs model, the civilian control model and the civilian review model. These three models represent two extremes and a way in between. The internal affairs model looks for internal checks, in which a police complaint would be investigated by police counterparts, while the civilian control model leaves the duty to civilian groups or members of the community, maximizing independence from the police force. The civilian review model is a mixture of the two, with investigation and review processes shared by the two parties [17, 22, 24].

Different forms of oversight institutions and procedures in democratic regimes might be established, with the objective of making police accountable to the public interest. These also aim to strengthen the survival of the police force because scandals about the abuse of “authority” threaten its reputation and public trust [13]. The protection of the human rights of citizens is simultaneously a purpose of establishing the police complaint system, addressing public grievances and fostering better police-public cooperation.

Insight from the Asian experience

Prevailing studies on police complaints and oversight bodies are context-based and concentrated on the Anglo-Saxon perspective [20, 21]. They generally examined the transformation and mission of oversight agencies, their institutionalization, and the structural strengths and limitations of the mechanism in an individual country. This study follows scholarship from Western democracies but chooses Asian jurisdictions to examine the design and operation of the police oversight mechanism (for example, [3, 4, 19, 23]. Hong Kong is a decolonized, not fully democratic society inheriting the common law system from the British colonial legacy, while Japan and Taiwan both practice the continental law system under democracy. The different historical trajectories led to three different police oversight mechanisms, varying in terms of professionalism, representation and empowerment, from their Western counterparts.

We explore how their complaints and oversight bodies were initiated and evolved to the current form to check the power of the police. Our discussion starts by tracing the developmental trajectories of police oversight bodies, examining what, how, and why the previous incidents or strategic choices of governmental actors shaped the current institutional arrangements. We attempt to reflect the politics involved in the development or reformation of police oversight bodies across different regions, hoping to understand how and why different pathways took place in this Asian context. A comparative account of the key features of these countries’ contemporary police oversight mechanisms will follow.

Research methods

A wide range of substantive sources, such as parliamentary papers, legislative debate and policy documents, were obtained to explain the different trajectories. We contrast our study with two prevailing data collection strategies in scholarship on Asian policing. The quantitative studies on behavioural and perceptual orientation in different societies or the qualitative, issue-based investigation of policing phenomena arousing international concern.Footnote 1 We notice that both strategies have their limitations: they are either too descriptive, lacking in-depth analysis of the similarities and differences among the systems practised in different countries, or they are “thematic”, concentrating on the discussion of very specific dimensions in policing or criminal justice. The valuable outputs, however, would frustrate readers who are not competent with Asian languages but eager to obtain a clearer picture of the contextual characteristics and dynamics that are crucial to understanding the operations and controversies of policing in Asian jurisdictions.

Our strategy is to reveal the dynamics by revisiting the primary resources published by different branches of the three governments targeted. We examined and analysed both Chinese and English sources for Hong Kong, the archival data in Japanese, and the Chinese resources publicized by the Taiwan government. Our research team is composed of native-speaker scholars of the three places, and we hope this enables our analysis to accurately capture the internal dynamics of the development of police oversight mechanisms.

We conducted systematic searches for the publicly accessible materials of the three places to critically examine the official discourse and identify key dynamics that have influenced the evolution of the police oversight mechanism. Public documents from the executive branch were first examined, followed by the archives and proceedings of the legislative assembly, contrasting the official discourse with that of civil societies. We retrieved the police oversight-related electronic archives in the Hong Kong Legislative Council (LegCo), together with publications of the police oversight body (PCC/IPCC). In the context of Japan, we selected the Police White Books and official documents published by the National Police Agency (NPA), which provide the official discourse for the police oversight mechanism. The proceedings of the House of Representatives (the National Diet) were also examined to better articulate the dynamics between the executive and legislative branches. We recognized, in addition to the national-level discourse, the significance of the “bottom-up” discussion—the documents issued by the prefectural police forces, particularly the operational guidelines for frontline police members. For Taiwan, we concentrated our efforts on the annual statistics from the National Police Agency and the investigation reports of police misconduct issued by the Control Yuan (an ombudsman-type assembly empowered by the constitution to monitor the public service).

Policing and police in Hong Kong, Japan and Taiwan

We start by highlighting the key features of policing and police in the three jurisdictions. Hong Kong operates within the common law system, inherited from the colonial British legacy. The principle of due process, independent investigation, and checks and balances are explicitly stated in the laws of Hong Kong. In addition to statute laws (codes) and ordinances, the doctrine of precedent (i.e., judge-made laws) is emphasized in the common law system. Hence, judgements made related to police misconduct would be considered legal principles in lawsuits. The Hong Kong Police Force (HKPF) is a law enforcement agency, taking charge of internal security management, investigation and arrest of criminals and presenting cases to the Secretary of Justice (Attorney General before 1997), who makes public prosecution. With sovereignty transfer to China in 1997, the criminal justice system was basically maintained in Beijing’s promise of “One Country, Two Systems”.

As the largest disciplinary service under the Security Bureau of Hong Kong Government, the HKPF consists of five departments in six regions. As of 30 April 2020, the force has 28,818 disciplined police officers who are employed as civil servants and 4,289 civilian officers [8]. The force operates within the constabulary concept with a primary purpose to ensure a safe and stable society by maintaining law and order, upholding the rule of law, preventing crime, and safeguarding life and property [8]. The HKPF was established under the colonial policing model and gradually gained its reputation in the final decades of British rule since the 1970s, combating corruption and introducing community policing practices and professionalization reforms to position itself in the role of “serving the community” since the 1990s. However, it has become highly controversial in recent decades, with severe public confrontations, which have reduced the public confidence in the HKPF [6, 9].

Japan is commonly regarded as a member of the civil law system. However, it actually has an eclectic system incorporating some common law features in the current form. Modern Japan's legal system commenced in the Meiji era. Reforms of the legal system and modernization of laws were later introduced on the advice of Western legal scholars. Influence of France and German laws could be found in the foundation of civil and criminal laws, with common law being referred to in their amendments. After the defeat in WWII, Japan's legal system was substantially transformed under the influence of American law. The peace constitution, legal, administrative system and policing systems experienced a complete overhaul in terms of structure and operation. The Supreme Commander for the Allied Powers (SCAP) oversaw the process, emphasizing the strengthening of democratic ideas and institutions under the stipulation of the Potsdam declaration. Three main principles were stressed in the new constitution: popular sovereignty, respect for basic human rights, and pacifism, and the status of the emperor was transformed into a symbol of the people. The trial mechanism was given the power of unconstitutional review, and the special courts were barred. This ideal to strive for democratic rule can also be found in the drafting of a police oversight mechanism, which prioritized “democracy”, despite its controversial changes.

The police forces in Japan are composed of the National Police Agency (NPA) and 47 prefectural police forces. The NPA operates at the national level and supervises the prefectural police forces. It is responsible for formulating police systems, conducting nationwide police operations in cases of national public safety and emergency, administering and coordinating policing activities, including police education and training, communications, and crime identification [37]. The prefectural police forces are responsible for daily operations and the policing of their respective prefectures. For example, the Tokyo Metropolitan Police Department (TMPD) is one of the prefectural police forces responsible for law enforcement within the city of Tokyo. Among the 259,224 police officers across the country, officers below the rank of Assistant Commissioner are known as local public servants, while officers beyond this rank are known to be national public servants.

Although Taiwan also adopts a civil law system, it inherits some legal principles and practices from the United States, hence making Taiwan a hybrid civil law jurisdiction. Codes are still the main body of this civil law jurisdiction. After its establishment in 1912, the Nationalist government gradually codified the Criminal Code, the Civil Code, Administrative Laws, and the Constitution [2, p. 23]. After the Republic of China regime was re-established in Taiwan, these codes eventually became the law of Taiwan.

The Taiwan police originated from the policing model established by the Chinese National Party (Kuomintang, KMT) in mainland China. It also has a number of features inherited from Japan, which ruled the island for 50 years (1895–1945), and features from the German civil law system [33, p. 10, 33]. The National Police Agency of Taiwan is composed of 18 internal units that are responsible for the planning, supervision and evaluation of national police functions, together with 20 subordinate agencies (i.e., Criminal Investigation Bureau, Aviation Police Bureau) responsible for various specialized police tasks [38]. As of the end of 2019, the police agency had a total of 73,405 employees. Approximately 70,000 of them are employed as police officers, while the rest are general administration and technical staff [40]. The Taiwan police force is supervised under the Ministry of the Interior. It is “obligated to maintain public order, protect social security, prevent all hazards, and promote welfare according to applicable laws” [34]. The police headquarter at Taipei appoints the heads of other cities and counties, although these departments are considered agencies of their local governments. Thus, a solid chain of command is formed from centre to local departments.

Researching the police complaints and oversight systems

There has been continuous discussion among academics over the purposes of police oversight, and how we could assess the effectiveness of these mechanisms in different political contexts [3, 4, 22, 23, 29]. Smith [25] identified four primary functions of police complaint systems: managerial, liability, restorative, and accountability. One of the most important purposes of these systems is to enhance public trust in the police force and hence, the force could continue to maintain order in the community.

Existing studies on police oversight have identified different types of oversight models, with each having different structures, power arrangements, and degree of lay involvement. These models include: (1) internal affairs model, (2) civilian review model, and (3) civilian control model. These models vary in the extent of citizens’ participation, and how and where the power of investigation is being vested. With different institutional settings, the purposes it served could be vastly varied. If investigative power is tightly held by the police department internally, less accountability could be guaranteed but it might bring managerial advantage. On the other hand, if more openness is brought into the system, more citizens’ participation could bring more trust into the mechanism, yet it might take more time and effort. More institutional pressure could be felt by the police force as well.

Hence, the role of police in the system is a key question of practical and scholarly debates in Western democracies. Whether an independent or highly autonomous police oversight body should be established to receive citizens’ complaints to the police, take charge of investigation, and follow up on disciplinary sanctions is the centre of the controversy [22]. In the “internal affairs model”, complaints of police misconduct are internally managed by a specific unit within the police force. Specialized officers usually take up the role on a job rotation basis and handle all complaints from citizens. Acceptance of complaints, opening of a case file, launching investigation, concluding the allegation and taking follow-up actions are all operated in an “in-house” manner. No external party is involved in police oversight, except for the judiciary, when police officers are screened for misconduct, criminally liable and prosecuted by public litigation [22, 24].

The oversight agency in the “civilian control model” performs not just an advisory function but has delegated authority to check over the police. This agency is independent from the police unit and can receive and handle complaints against the police and recommend or even take action to sanction officers who are found to be guilty of misconduct. Usually, such agencies are composed of a board of civilians who investigate, review, and adjudicate police misconduct allegations. They have the complete responsibility for conducting independent investigations, proposing disciplinary decisions and prosecutions, and evaluating police internal corruption prevention strategies. Police were excluded from the investigation and review process. The underlying rationale of this model is to enhance public participation and confidence in the mechanism by preventing “self-investigation”. It also rests on the assumption that outsiders can provide meaningful inputs to the force to improve policing.

The “civilian review model” is a hybrid of the “internal affairs” and “civilian control” models and is characterized by the presence of both “internal” and “external” elements in the investigation and review processes. This model represents a compromise between the demand and is opposite to a fully independent system. Members of the oversight board are from different walks of life, not necessarily having a policing or law enforcement background, and they are appointed to review the conclusion submitted by the police on the complaints against them. The composition of the board varies from country to country. Some members are socio-political elites appointed by the executive in an individual capacity. In some countries, they assume this public service on a voluntary basis and are supported by a secretariat made up of a team of professionals. The statutory authority of the oversight board also depends on the source of its authority—members might be constitutionally empowered, delegated by a particular legislation, or directly accountable to the leader or legislature of government. In general, the oversight body in this model can suggest procedural changes to the complaints directed by the police, urge disciplinary decisions on officers who are substantiated for misconduct after in-house investigation, and call for review of controversial decisions on complaint cases. However, they do not have the authority to take proactive investigations or order sanctions against police officers.

A recent trend: from independent review to independent investigation

Prenzler argued that significant variation remains in the structure and responsibilities of oversight agencies, but there is an ongoing trend towards replacing an ombudsman model (civilian review model) with a public sector-wide commission (independent civilian oversight model). Recent debates on police oversight concentrate on the capacity and empowerment of the independent oversight agency: whether it can have the statutory authority to proactively launch the investigation upon the receipt of a complaint or even be equipped with the adjudication power to state the order for the police to take disciplinary action if the conclusion is that an officer was guilty of misconduct. Hopkins and Flemington [10] argued that investigations of allegations of misconduct, criminality and human rights abuses must be conducted by an agency that is not only institutionally independent of police but also practically, culturally and politically independent. MacPherson [28] also stressed that investigation steps can and should be taken to ensure that serious complaints against police officers are independently investigated. He claimed that the investigation of police officers by their own or another police service is widely regarded as unjust and does not inspire public confidence.

Between or beyond?

The recent trend and studies in police oversight mechanisms seem to suggest a reform choice, from internal affairs model / citizen review model, to citizen control model. The utmost importance of citizens’ participation reigns on the checklist of purposes in police oversight. However, institutional development was more complicated than that. Alternative options should be provided for incremental and gradual reform on police oversight mechanisms. Our study aims at providing, especially with the case of Japan, an alternative pathway in-between citizen review and citizen control model. And also, with the case of Taiwan, a constitutionally empowered model beyond the citizen control model, which further enhances the institutional check on police misconduct. In the following, we will show how the two cases acted as the alternative pathways in-between and beyond the citizens review and citizen control models.

Comparing police oversight: historical and institutional perspectives

To answer the question, we set this study professionalism versus democracy in police oversight in the Asian context, and we examine and compare the mechanisms of complaints against police and their oversight bodies in the three places, exploring their paths of development, prevailing mechanisms of institutionalization, and recent controversies regarding their operation.

Path dependence, institutional development and changes

Fioretos et al. [5] argued that institutional change is a result of a sudden shift or exogenous shock. The reformation of or significant changes in an institution occur when a critical juncture appears. A critical juncture characterizes the adoption of a particular institutional arrangement over other alternatives in an externally contingent situation. The junctures are “critical” in the sense that the choices made by important actors are decisive. Once a particular path is chosen, it is difficult to return to the initial point given the alternatives available [16]. Similarly, Krasner [14] proposed the theory of punctuated equilibrium. It posits that the institution remains in a prolonged period of stasis and is punctured by sudden shifts in society or government, which lead to a radical change in the institution.

Our review of the mechanisms from the three countries highlights their transformation after the Second World War. This was a watershed for the transformation of the governance system in the three places—the “liberation” and British resumption of Hong Kong governance; the Allied occupation of Japan; and the decolonization and relocation of the KMT government to Taiwan—all were the starting points at which the new regime values modified the policing system. We examine the key rationales, features and evolution of police complaints and oversight systems. Critical junctures for their transformation factors are also explored.

Hong Kong: government’s advocacy for a “professional” advisory body

Syndicated corruption and allegations of police misconduct were prevalent in Hong Kong before the 1980s. However, citizens were demotivated to report problems and make complaints against the police. In 1974, the Complaints Against Police Office (CAPO) was established within the Royal Hong Kong Police Force to investigate police misconduct. Prior to the establishment of the CAPO, complaints against the police were investigated by police chief inspectors, and investigations conducted were administratively reviewed by a district commander. Since then, the CAPO has monopolized the investigative duties in the police oversight mechanism. Indeed, there were a series of struggles and demands seeking to give checks or get rid of this investigative power from the internal organs of the police force.

In 1988, legislators moved to putting the CAPO and PCC under the jurisdiction of the newly formed Commissioner for Administrative Complaints (later the Ombudsman). With this very first attempt ended in failure, struggle continued by legislators injecting progressive ideas into the system. In 1993, a democrat legislator launched a motion urging the government to “set up an independent body to receive and reinvestigate complaints in relation to police officers”, per se a complete overhaul of the existing system. Notably, this “radical” move was in fact supported by the majority in the legislature, with 32 in favour and 20 against the motion. However, the administration stood firm on not making a complete change to the existing police oversight mechanism. This was on the grounds that it would be better for investigations to be carried out by the police themselves, for the sake of their expertise and for the concern of police morale during the transition period. As such, the government turned down the motion.

Instead, the administration dished out a more conservative reform. Only a series of minor reforms, taking place from 1994 to 1996, included renaming the PCC to the IPCC (adding the word “independent” before), granting members limited access to interview witnesses or video-recording interviews. Hence, the rather moderate path taken by the administration to reform the police oversight mechanism focused mainly on the PCC (later IPCC), the “specialized police oversight body”, in itself only a non-statutory body inherited from the task group formed by the Unofficial Members of the Executive and Legislative Councils (UMELCO).

The Professionalism-packaged reform continued, with power held by the IPCC kept literally intact, though the government spelt and promoted her function more explicitly. In 1996, the government tabled a bill, hoping to grant a “statutory status to the IPCC”; yet, the bill faced severe criticism in the LegCo for its conservativeness. The Security Panel chairman blamed the administration for keeping the IPCC as a “toothless tiger” [15]. The government finally withdrew the bill, leading to the stagnation in reform of the police oversight mechanism. Not until June 2009, in the post-colonial era, the IPCC finally became a statutory institution, stipulated by the IPCC Ordinance (Cap. 604, Laws of Hong Kong). Composed of seven parts and four divisions, it entails the IPCC to have functions in several categories, covering “commissioner’s categorization of complaints, investigation or interim investigation reports on reportable complaints, and other functions relating to reportable complaints etc.”; and powers relating to its affairs.

The “reform” in Hong Kong has not changed the fundamental power relation in the police oversight mechanism, heavily coloured by a rationale of professionalism, defending the CAPO and leaving the police force to self-check. Hence, the mechanism is best categorized by a CAPO, an internal police organ where monopolized most power, being the only police complaints body that investigate against police complaints, and a IPCC, officially entitled as the specialized police oversight/monitoring body, yet, is comparatively weak, without any power of investigation. Not to mention, her rather limited size and function to intervene sufficiently into the investigation process, e.g., limited rights to interview. The colonial nature of the police force and the police oversight mechanism has been largely kept intact, as there was no exogenous shock leading to any fundamental change. Additionally, the “executive-led” setting of the colonial government gave the administration the privilege to neglect the demand from the legislature. Therefore, the power of handling police complaints remained tightly seized by the internal investigative body, while the reforms made to the monitoring body provided only supplementary support, if not being decorative in nature.

Japan: democratic representation under the “national security” concern

The model in Japan is characterized by popular representation in the external oversight body in the absence of investigation power. The model is richly coloured by democracy and largely a result of the democratization project imposed by the United States in the post-war era. Since the Meiji restoration, the Japanese police have been centralized by the Police Affairs Bureau of the Home Ministry. However, the Japanese policing system was forced into a complete change after the Second World War. To satisfy the American, the design of the new police oversight mechanism heavily led towards “ensuring democratic governance” [39]. As a result, significant institutional reforms, including the abolition of the Home Ministry, were implemented. The Public Safety Commission was established to maintain “the neutrality of the police force and the democratic methods within police administration”, enacted by the new parliament under the new constitution in 1947 [36]. A new Police Act was drafted by the new Japanese cabinet, with General MacArthur’s request to establish a new police oversight mechanism. The Public Safety Commission was hence established under the decision made in the two Houses of Representatives (Diet) in December of the same year. The Police Act also reformed the traditionally centralized national police system into a two-tiered system, better reflecting the principles of democracy and local autonomy enshrined in the new Constitution.

Under these new institutions, democratic control was emphasized. The Public Safety Commission, composed by citizen representatives, was adopted, decentralizing the management of the police to the public. All cities, districts and villages had their own public safety commissions, a decision agreed by the Diet in December 1947 and subsequently put into practice. All public safety commission members were appointed by the Prime Minister or the local governor of the corresponding region with the approval of the National or Local Diet. With this mechanism, public safety commission members were theoretically “representatives of the people” [36]. With them in the administrative position over the police forces, the police force was “overseen by the people” from the official standpoint. Therefore, a system emphasizing representation was built in early post-war Japan.

In the aftermath of the Chinese civil war and the Korean war, the occupation policy, originally heavily advocating the principle of democracy, began to turn right rapidly at the background of an escalating Cold War. The Police Act was completely revised and promulgated in 1954 after Japan regained its independence and sovereignty with the treaty of San Francisco. A recentralization of the police organizations took place by abolishing the two-tiered system and establishing the National Police Agency and the prefectural police departments. As a part of the revised police system, the National Public Safety Commission (NPSC) enjoys a higher status than the NPA, meaning that it is currently in a leading position of police power, at least legally.

While the NPSC was set up to oversee the NPA, prefectural public safety commissions were also set up to oversee the frontline policing practices of the prefectural police forces. With the intention to “ensure the cabinet’s responsibility in public order” that was supposedly held by the NPSC, the reform specified the Minister of State as the mandatory chairman of the NPSC [37]. This means that a cabinet member will always be present in the most major police oversight body. City, district, and village public safety commissions were also removed due to the financial burdens they posed to the local governing bodies, leaving the monitoring of regional police forces in the hands of prefectural police forces [36]. The Public Safety Commission’s mechanism was not notably amended until 2000, when a series of police misconduct incidents across the country were reported. Public trust in the police force was considerably damaged, and improving the Public Safety Commission system became a part of the police reform campaign launched by the cabinet. The reform was later passed, introducing a complaint management system into prefectural public safety commissions and further clarifying the public safety commission’s duty and granting the power to oversee specific investigations when necessary [39].

Taiwan: the parallel existence of internal investigation and a constitutionally empowered “ombudsman”

The Internal Affairs Office of the National Police Agency (IAO), the Control Yuan, and the Civil Ethics Unit are the major components in Taiwan's police oversight mechanism, which is another democracy-leaning system. This system stands out by emphasizing representation, and even more, on empowerment. The Internal Affairs Office is an internal agency under the National Police Agency, handling complaints against police. Its establishment dates back to 1912. The National Police Headquarters under the Ministry of the Interior was established to supervise national police affairs. Starting from the capital, police stations in other provinces, counties and trading ports were then established [38]. Simultaneously, regulations were developed to require both central and municipal police departments to establish their own police oversight bodies, hence every police department deployed a certain number of personnel to supervise their own work. After the political turmoil in the 1930s–1940s and the KMT’s exile to Taiwan, these scattered police oversight bodies unified together and formed the IAO. As ratified by the National Police Agency Organization Act, the IAO is an internal, supporting agency to monitor and issue compliments to police officers. Moreover, it bears the responsibility to initiate investigations to prevent police officers from becoming involved in criminal offences or power abuse [30, 35].

The Control Yuan is a distinctive feature in this democracy-leaning system, and as one of the most influential constitutional structures, holding the ability to exercise the power of impeachment against government officials, including police. Its origin is from Dr. Sun Yat-Sen, founder of the Republic of China, who sought to prevent legislative bodies from being overpowered and the administrative branch from being undermined. As Taiwan was a new democratic regime, it was designed to have an independent body beyond the traditional separation of powers. As the supreme scrutiny authority, it was enabled “to achieve division of labor… [exercising] full power of impeachment…with all rankings of government officials and civil servants… [The Control Yuan] can exercise its power of urging other public agencies to correct their inappropriate policies” [41]. Thus, the Taiwan police force has an external, superior constitutional body to monitor its behaviour. Referring to the Control Yuan’s official website, the general public preserves the right to submit written complaints against the police in addition to the channel provided by IAO. Members of the Control Yuan are empowered to initiate investigations related to police complaints. If police officers involved are found to be in misconduct or guilty, orders are directed to their unit to correct their behaviours and practices [31].

The Civil Ethics Units (formally Government Employee Ethics Unit), founded in 1953, is an auxiliary agency in Taiwan’s police oversight mechanism. Originally formed to be a surveillance body serving the authoritarian regime, its function changed after democratization. According to the Ministry of Justice, this unit was established to handle “government employee ethics… excluding the Legislative Yuan/councils, military units and public schools of various levels” [18]. Unlike the members of the IAO, members of the ethics unit have a more diverse background. Other than prosecutors deployed by the Ministry of Justice, members from the Agency Against Corruption and frontline police officers may join the unit. The presumption is that frontline police officers should be familiar with police operating procedures as well as the police subculture [32]. The Civic Ethics Unit has various sub-branches throughout police stations in Taiwan, in particular in large police forces, such as those in TaoYuan and Kaohsiung. Coordinating with the IAO, it intends to create an integrated police oversight mechanism for Taiwan.

Current police oversight mechanisms

The previous section presented the historical trajectory of the police complaints and oversight bodies in various jurisdictions. This part offers a comparison of their current forms, particularly the distribution of powers in different organs within the systems.

In these three jurisdictions, the organs receiving complaints from the public vary. In Hong Kong, the Complaints Against Police Office (CAPO) is the sole complaint receiving agency. In both Japan and Taiwan, there is more than one agency channelling grievances from the public. Prefectural Public Safety Commissions (PPSCs) or the prefectural police can take complaints in Japan. In Taiwan, the power is distributed to the Internal Affairs Office (IAO) under the Taiwan Police, the Control Yuan, and the Civil Service Ethics Unit.

Investigation starts after formal complaints are filed. The CAPO in Hong Kong manages the investigation and resolution of all complaints of misconduct or allegations of crime against the police or any member attached to the force, excluding traffic complaints. In Japan, all investigations are undertaken by the police internally, yet direct supervision from members of the PPSC is theoretically possible. In Taiwan, the IAO is the key agency that handles complaints related to police misconduct except corruption-related cases. Its duties include receiving complaints and investigating the imposition of charges. However, the constitutionally empowered Control Yuan could also receive complaints from citizens and investigate. The powers of the Control Yuan reach far, without limitations related to the types of complaints, which are mainly handled by the Control Yuan’s Complaint Receipt Centre [31].

We can take a closer look at the procedure to show the power distribution. The complaints received are categorized by the CAPO. Minor complaints, such as rudeness or incivility, might be resolved initially before they turn into formal complaints. The majority of the complaints are not counted as “substantiated”. Many of them are “informally resolved”, “withdrawn” or “non-pursuable”. Only when a case is found “substantiated” is a full investigation initiated, with formal statements being taken. Only upon completion is an investigation report submitted to the IPCC Secretariat [12]. The IPCC cannot take part in the investigation process of the CAPO.

In Japan, whether complaints are filed to the police or to the PPSC, investigations are still completed by the police internally, though there is a mechanism for cross-checking. The police grant the final decision rights of measures and replies to the PPSC, while the PPSC grants the decision rights to the prefectural chief [38]. If the complaints are submitted to the PPSC, the prefectural police force is ordered to conduct investigations. Reports then pass back to the PPSC for approval. If a report is approved, it will be delivered to the complainant. If not, in case of unsatisfactory findings, the PPSCs are empowered to order a reinvestigation. If deemed necessary, they can also assign one of their own members or appoint overseers within the police system to directly oversee a particular investigation.

In Taiwan, police complaints are largely investigated by the IAO, with the public reporting to the nearest local police station. If police officers are found guilty, managerial, disciplinary or administrative measures can be taken. Alternatively, citizens in Taiwan can present grievances to the Control Yuan. With citizens’ written complaints, members can initiate investigations. Unlike the IAO, the Control Yuan can exercise its power of impeachment. Orders are issued to the relevant commanding units to take corrective measures.

An important structural difference between Hong Kong and Japan and Taiwan lies in the presence of a specialized oversight body (IPCC in HK), which distinguishes this professionalism-leaning system from the other two democracy-leaning systems. While the presence of an independent police oversight body might be common for Western democracies, this is not the case in Asia. Very different from the case of Hong Kong, where the body is dedicated to monitoring the investigations of complaints, there are no specialized oversight bodies in Japan and Taiwan. Even though the body is specialized in police oversight, the power vested into the institution is not comparable to the power of the Control Yuan in Taiwan and the NPSC/PPSC in Japan. The presence of a specialized police oversight body in Hong Kong does not guarantee an independent investigation of police misconduct. As a specialized body independent from the police force, the IPCC is supposed to monitor the investigations of reportable complaints. It is a statutory body with legal status. However, the functions of the IPCC defined by the administration are only to review and endorse the report prepared by the CAPO [11].

In contrast to the situation in Japan, although there are no independent and specialized bodies to oversee complaints against the police, as public safety commissions are still part of the entire police mechanism, it is also important to note that public safety commissions are institutionally superior to the police force. Additionally, the public representatives sitting in the NPSC or PPSC enhance the level of public mandate. This representation gives the institution privileges that Hong Kong’s agency does not have. Both the NPA and the Prefectural Police are required to report to the relevant public safety commissions, and if the results are unsatisfactory, the public safety commissions are empowered to command further investigations. More importantly, the PSCs can also appoint special overseers to oversee certain investigations. All results and punishment decisions must be reported to the Public Safety Commission, and its agreement is required for the final decision (NPA, Japan, 2005). In sum, the PSC is more powerful than the IPCC institutionally. The Public Safety Commissions are defined as a commanding unit of the police force by representation, enhancing the accountability and public control of the NPA and the prefectural police.

The Control Yuan in Taiwan is an ombudsman unit at the constitutional level that supervises and oversees all the misconduct of government officials, including the police force. Representativeness can be gained on nomination by the president with the consent of the Legislative Yuan, itself composed by election [31]. Therefore, independent investigation can be maintained without a specialized police oversight body but with the endorsement and empowerment of the public mandate.Footnote 2 We will further highlight their disparity by bringing in three concepts: professionalism, representation and empowerment (Figs. 1, 2, 3).

Fig. 1
figure 1

Complaint procedures in Hong Kong

Full size image
Fig. 2
figure 2

Complaint procedures in Japan

Full size image
Fig. 3
figure 3

Complaint procedures in Taiwan

Full size image

Three major controversies: professionalism, representation and empowerment

By reviewing the police complaints and oversight mechanisms of the three places, we found the key controversies in line with the debates in the prevailing literature: is the police force better checked by its own counterparts or public representatives? Among the scholarship, three major dimensions are intensively debated: the “professionalism” of the investigation and the investigators, the “representation” of the public mandate, and the “empowerment” of the external oversight body. “Professionalism” advocates argue that stakeholders with front-line policing experience must be involved in all police oversight processes for the sake of their knowledge in operational rules and complexities in daily policing practices, which enables them to better handle police complaints. “Representation” refers to the composition and formation of a police oversight board: whether the members are democratically elected or sufficiently represent the public. For “empowerment”, the critiques concentrate on the power of the oversight bodies: whether they are empowered to carry out independent investigation or just equipped with a passive role in monitoring the investigations conducted.

In this study, we found that the traditional categorization of police oversight mechanisms (“Internal Affairs”, “Civilian Review” or “Civilian Control”) could not sufficiently give us a full picture of police oversight in these Asian jurisdictions. Japan and Taiwan, both democratic countries, could not be easily categorized as Civilian Review or Civilian Control, as they developed distinguished police oversight mechanisms based on their respective historical trajectories. The Japan case is rather close to the civilian review model, with the NPSC and PPSC lacking investigative power, yet bringing in more public mandate by increasing the “representativeness” of the membership. The latter is similar but one-step beyond the civilian control model, highlighted by the independent power of the Control Yuan, which is empowered not only to launch investigations on any government official, but also constitutionally based. By emphasizing “empowerment”, we find that Taiwan’s model gives even more institutional power to police oversight bodies than other independent civilian-control police oversight bodies. The institution in Hong Kong does not only outstand itself as the internal affairs model, but rather defends the institution by emphasizing on “professionalism”. The authority keeps the specialized police oversight agency under-powered, leaving the process of police oversight largely in hands of the internal agency of the police force.

This offers the insight that the development of the police oversight mechanism is highly correlated to the regime value and the historical trajectory. As the regime values define what society ought to pursue, choosing between stability and accountability, etc., they determine the “purpose” of these police oversight mechanisms. As the discussion above revealed, the oversight system in Hong Kong seemed to include both internal and external stakeholders in the process, but the Hong Kong Police Force monopolized the investigation, revealing the regime's nature to be nondemocratic, if not authoritarian. The primary investigation power for complaints against police in Japan and Taiwan is still in the hands of the police. Like the CAPO of the Hong Kong Police, the prefectural police in Japan and the IAO of the Taiwan police receive, screen and follow up all the complaint cases. However, the Control Yuan of Taiwan also has the supreme constitutional power to investigate the police, even though the Control Yuan is not a “professional” and specialized body for police misconduct investigation.

A major disparity lies in the power to audit investigation results submitted by the police. While the oversight bodies in Taiwan (Control Yuan) and Japan (PPSC) have the statutory power to audit the investigation results completed by the police, with power of re-investigation to a certain levelFootnote 3 The IPCC in Hong Kong can only observe, monitor and review the investigations conducted by the CAPO and lacks the power of proactive investigation.

The presence of a “professional” and independent oversight body (IPCC) differentiates the oversight arrangements in Hong Kong from those in Japan and Taiwan. However, an independent police oversight body does not guarantee an impartial and professional investigation due to institutional limitations. The body might only be advisory in nature, insufficiently empowered, lacking resources and falling short of the policing and legal expertise needed to effectively monitor the process of complaints investigation (Table 1).

Table 1 Comparison of the roles of the police oversight bodies
Full size table

Conclusion: path dependence and regime values for professionalism and democracy

We choose to study these three places with varying levels of democracy, but share several commonalities: all of them have recorded good figures of public satisfaction to their police forces; citizens’ complaints against the police have all been low level; police were perceived as ‘little corrupted’ and ‘trustworthy’; and the overall crime rate is low as well. The policing oversight mechanisms of the three places were developed and evolved from the three mainstream models as identified by literature, but they all showed some distinctive features that are not exhibited in western democracies: the emphasis on professionalism and elitism in the consultative review process in Hong Kong; stress on civil society participation in Japan; and highlight on the substantial empowerment of external oversight bodies in Taiwan.

Paradoxically, the non-fully democratic Hong Kong SAR is the only case among the three places to have publicly accessible and comprehensive figures on the complaints against police, cases being externally reviewed and substantiated for disciplinary actions. The transparency also shows the low substantiation rate for the complaints against police, which indirectly explains the accusation of ‘toothless tiger’ of the IPCC (Fig. 4). The Control Yuan of Taiwan is constitutionally empowered for administrative oversight. However, it does not only perform the function of police watchdog, and simultaneously a watchman for other governmental departments. Hence, the statistics of Control Yuan does not specifically target the complaints against the police. Instead, their checking of the police could be revealed in the statistics under the category of the Ministry of the Interior, with the former placed under her command. A wide range of other governmental agencies including the Office of the President, the National Security Bureau, the Legislative Yuan and the Supreme Court were also checked by the Control Yuan. Henceforth, the statistics publicized by the Control Yuan could not provide a specific set of figures against police, like the practice in Hong Kong, that would enable us to directly assess the effectiveness of police oversight despite her transparency and accountability to the public. In Japan, the complaint figures and number of officers being disciplined were both available from the publications of various policing units- the NPSC, PPSC or NPA. However, they did not show the substantiation cases after investigation unlike the practice in Hong Kong (Figs. 5, 6).

Fig. 4
figure 4

Complaints against police reviewed by the IPCC, Hong Kong. Sources: Independent Police Complaints Council, Hong Kong 2009–2020

Full size image
Fig. 5
figure 5

Number of Police Officers being dismissed, suspended, salary reduced and warned annually. Sources: National Police Agency Japan, 2002–2009

Full size image
Fig. 6
figure 6

Complaints processing time. Sources: National Police Agency Japan, 2002–2009

Full size image

The incomplete set of data primarily disables us to provide a simple answer with statistics to weigh the efficacy of different policing oversight mechanisms. Nevertheless, we could see the institutional strength of the Control Yuan in light of several recent and important incidents. During the Sunflower Student Movement in 2014, with police being accused of force abuse, the investigation result of IAO found their colleagues exercising their power legitimately. A Huge outcry urged the Control Yuan to re-investigate. The Control Yuan lived up with their expectation and published their discordant investigation result from the IAO, urging the Ministry of the Interior to review their operation and bring the alleged police officers under public prosecution.Footnote 4 Another incident happened in 2015, in which the Control Yuan proactively investigated, against the police department under TaoYuan City. It was after a car accident made by local police constable Li Wen Yi, with his colleagues trying to cover up his guilt, the police department was later found responsible and criticized by the Control Yuan for seriously undermining the public trust towards police.Footnote 5

With the institutional privilege of the Taiwan and Japan police oversight mechanism on responding to public mandate, our study reveals the critical juncture of their development: what explains the critical differences in these police oversight bodies, leading to their contrasting institutional settings, one emphasizing insiders and professionals’ expertise and another filled with popularly elected representatives? The study sheds light on the debate between “professionalism” and “democracy” in police oversight to gain maximum trust from the public. We attempt to map and compare the evolving oversight mechanism in Hong Kong, Japan, and Taiwan and explain what has successfully brought reforms.

Providing more autonomy to the police oversight body is the direction of reforms in Western democracies [26]. A number of police complaint bodies that were limited to internal investigation have been replaced by external bodies with investigation powers to ensure independence or separation from the police. However, our study revealed that the process of reform in the Asian context was not always easy. The trend of reforming the oversight system for more autonomy has yet to spread to the countries in Asia. The study found that Hong Kong, Japan, and Taiwan are still relying on internal mechanisms to address the investigations of misconduct and public complaints. While it seems that the investigations of complaints are internally handled, in fact, in Japan and Taiwan, there are bodies external or superior to the force that are authorized to investigate public complaints or monitor police misconduct. We are happy to see that this trend partly echoes the tendency elsewhere around the world. These are not the primary investigating agencies, yet they retain the rights to engage in investigation, which is a similar approach that is used in Australia and New Zealand [3]. Furthermore, the external monitoring of investigations conducted by the police differs quite significantly across these three regions. While no specialized oversight bodies are established in Japan and Taiwan to review the investigations of complaints, there might be sufficient checking, institutionally, to impose constraints on the internal investigation body. In contrast, there is a specialized body to review the police investigations of complaints in Hong Kong, the IPCC, yet independent investigation is not guaranteed, for the lack of institutional mandate empowering it to make independent investigations. At the end of the day, if a democratically empowered organization is found not professional enough, it could be trained and enhanced by better selection of personnel. However, if the organization is not institutionally empowered, the professional membership could not save it from public skepticism in lack of sufficient checking.

The development of police oversight bodies are heavily constrained by respective historical trajectory and will of government officials. Reform, however, is triggered by critical events leading to critical junctures. For example, the United Kingdom replaced their Police Complaints Authority in 2003 with the Independent Police Complaints Commission due to allegations of a lack of independence [27]. While this paper has attempted to present the police oversight arrangements within historical and institutional contexts comparatively in three Asian locations, it has yet to fully measure their performance, as well as provide a prescription for eliminating police misconduct once-and-for-all. Instead, we show constraint and alternative possibilities in police oversight mechanism reform, contributing to future studies in police oversight.

This study finds that historical conjunctures and regime leadership are both decisive in transforming police oversight. The lack of democracy in government in general constrained the empowerment and enhancement of representation in the police oversight mechanism, institutionally paralyzed the administration to restore confidence of the police force in face of crisis. Solely relying on professionalism, could not sustain the system’s reputation in the longer term. The comparison primarily finds that competence of personnel and transparency of the investigation process appear not decisive in determining public trust towards the oversight mechanism. Representation and empowerment instead save the system.