The Constructed Truth

Abstract

Witnesses rarely testify at trial in China and the courts routinely rely on investigative dossiers to determine the guilt or innocence of the defendant. Shielded from external scrutiny, relatively little is known about how these investigative dossiers are constructed and whether they are truly reliable. To understand the construction process for police cases, ethnography, semi-structured interviews and content analysis of the dossiers have been conducted to explore the formation of evidence during the police investigation. This article reveals that the constructed evidence is subject to manipulation and distortion designed to enhance the incrimination of the accused. With a lack of the functional equivalence in defence construction required to challenge the facts presented in these dossiers, the current criminal justice system in China is structurally weak and fails to function as a truth-finding process.

Keywords

Case construction China confession criminal justice system criminal procedure law 2012 investigation investigative dossiers statements the police witnesses

Introduction

One striking feature of criminal trials in China is the absence of witnesses. Various surveys indicate that witnesses testify in court in less than 5% of cases (Chen, 2007; Shi, 2002; Shi, 2013).¹ China has claimed to have departed from its socialist tradition and reformed towards an adversarial system beginning in 1996. However, the absence of witnesses in the Chinese court contrasts with this reformed adversarial format. With the majority of trials having no live witnesses, the court hearing is about reading out the statements compiled in the investigative dossier (Zhencha juanzong) fully or selectively. This feature has long been criticized by academics. Over the last two decades, a large body of literature has emerged to identify the reasons that lead witnesses to refrain from giving oral testimonies at trial (e.g. Chen, 2001; McConville et al, 2011: 246; Zuo and Ma, 2005; Zhang, 2014). For instance, witnesses’ fear of revenge and the lack of protection have long been acknowledged as prominent reasons. Other organizational issues, such as judges’ stereotypical dossier-reading working model, which naturally expels the principle of using oral evidence, have been suggested as a contributing factor. Studies have also alluded to the fact that judges and prosecutors have the mentality of wanting to block live testimony in order to control the result of the trial.

While the cause of the low attendance of witnesses is worthy of analysis, given the fact that the courts rely almost entirely on written dossiers to determine the guilt or innocence of defendants,² it is crucial to examine the integrity of the dossiers. How are these dossiers created and are they truly reliable? Despite previous research reflecting on police files (McConville et al, 2011, Chapter 4; He, 2014, Chapter 6), there has been no scholarly investigation exploring these critical issues. Drawing upon empirical data, this article will investigate the way that investigative dossiers are created by the police, specifically focusing on the integrity of the written evidence and its implications within the Chinese criminal justice system. It aims to understand whether it is safe to rely on police case to make judicial decisions.

Social construction theory argues that facts are created through human agreement rather than being phenomena intrinsic to nature (Goldman, 1986; Searle, 1995). Opposing the idea of the existence of absolute truth, constructivism stipulates that it is the observer who constructs the reality and any statement of the observer about his observation is purely her own interpretation (Herting and Stein, 2007). Extending this notion into the criminal justice sphere, Sanders (1987) and McConville et al. (1991) argue that criminal cases should not be perceived as ‘discovered’ or ‘unearthed’ objective entities that exist independently of legal actors. What happens at every stage of the criminal process is subject to rejection, addition, selection and reformulation. This construction process involves not only the interpretation of evidence but also the very creation of evidence itself (McConville et al, 1991: 12; Sanders, 1987). They argue that while facts, values and rules are interpreted separately in legal rhetoric, in the world of case construction, they are fully integrated.

Accepting so-called truth or reality is socially constructed, assessments of what happened are always creator-dependant ascriptions. Applying this understanding to the Chinese criminal process, this article examines the truth presented by the police in the case dossiers. Methodology session sets out an account of the data collection for this study. Then the following session discusses the role of the police within the context of China, including the value systems and ideologies of the police which provide a backdrop to the analysis of the construction of police dossiers. Section ‘Formulating the official version of the truth’ describes the way that the official version of the truth is constructed, which creates the basis upon which police cases are framed and developed. By referring to accounts and details provided by legal actors, in relation to the way that the statement was made, section ‘Aligning the Evidence With the Official Version of Truth’ examines the reliability of police dossiers. Finally, the section ‘The Construction of Police Dossiers Within the Context of the Chinese Criminal Process’ places the construction of police cases in the context of the Chinese criminal justice system and considers the implications that this unreliability could have for miscarriages of justice in China.

Methodology

This research aims to explore the construction process for police cases, focusing on the interaction between the police and other legal actors. With no prior knowledge of legal practice within the legal institutions, I believed the knowledge of complex criminal processes can be best informed by observing a ‘real-life’ setting of the legal institution and experiencing the interactive situations. Therefore, the method of ethnography was adopted to understand the legal culture of the institution by ‘piercing and progressively reducing the ‘‘otherness” that separates the researcher from among who she seeks new knowledge’ (Fox, 2004: 73). Such a position allows the researcher to be an interpreter of the data yielded by the participation and a ‘knower’ who understands the shared ideology, values and dilemmas of the legal personnel. However, since every source of police practice in China is treated as sensitive and police stations are closed to external researchers, I had to modify my research plan.

With the help of an informal contact, I approached a local procuratorate and was permitted to observe inside a prosecutor’s office, which was responsible for reviewing police cases and deciding whether or not to proceed with prosecution. With over 630,000 residents in its jurisdiction, the procuratorate (site A) was located in a large city in Western China. It deals with over 2000 cases from 13 local police stations every year. Fieldwork was carried out over a 6-month period, comprising of 18 weeks in 2012 and a 6-week follow-up visit in 2013. I was given unmediated access to the investigative dossiers pending trial and had the opportunity to observe prosecutorial interrogation, in which the written statements in the dossier were verified using suspects’ accounts. I was able to speak to the police officers who sent new cases to the office and requiring legal opinions from prosecutors. There was bureaucratic trust between the police and prosecutors, meaning that their casual conversations provided a rich reservoir of information with which to understand the inner workings of Chinese legal institutions.

As the observation was carried out in a natural setting, the researcher’s control over the research process was limited. Despite the fact that the legal actors were formally informed, occasionally, it was not possible to ensure that all the people I observed were notified: in a busy prosecutor’s office with spontaneous interactions constantly taking place, it was sometimes physically impossible to seek consent from every one.³ Nonetheless, the main participants had been informed at the beginning of the observational period of my role as a researcher. I had also explained my research subject, so that people involved knew the direction of my study. During the course of my fieldwork, all the institutional and personal details involved were recorded anonymously so that people who had participated could not be recognized. The major cases that I followed were also modified to some extent so that they could not be identified.

Due to the difficulty of negotiating access to other sites, observation in this study was confined to the procuratorate in site A, where drug trafficking, dangerous driving, the sale of (illicit) receipts and theft comprised 80% of cases. During the course of the ethnography, 240 evidence case dossiers⁴ were reviewed. Among these dossiers, 64 criminal cases (eight of which involved co-defendants in one case) were monitored and recorded from the point that the case was transferred to the prosecution, to the outcome of the trial. These evidence dossiers used by the legal institutions for decision-making were paper-based files. Inside the evidence dossier, important official documents usually include the registration form of the criminal case (baoan cailiao), the record of the arrest of the suspect (zhuabu jingguo), interrogation records (xunwen bilu), the victim’s statement(s) (beihairen bilu), witnesses’ statement(s) (zhengren zhengyan), crime scene inspection (xianchang kanyanbilu), the list of seized items (kouyawuping qingdan), experts’ evaluation form(s) (jianding jielun) and the record of identification (bianren bilu). Content analysis was carried out to scrutinize the use and the frequency of keywords, narrative styles, the sequence of listing items, the use of ambiguous and unambiguous words and the repetition of language format within and between the case dossiers. To compare what I had observed with practices in other parts of China, at the end of the observational period, 28 semi-structured interviews with different legal actors (police, prosecutors, judges and defence lawyers) were conducted in 10 geographic areas across China. It is worth noting that the omnipresent repression of the liberty of speech in China made the negotiation of interviews extremely difficult. Hence, I am indebted to the interviewees who imparted valuable information to me in this study.

It has been acknowledged that a research interview is ‘not an open and dominance-free dialogue between egalitarian partners, but a specific hierarchical and instrumental form of conversation, where the interviewer sets the stage and scripts in accord with her research interests’ (Kvale, 2007: 49). Hence, the interview is orchestrated and constructed by both interviewer and the interviewee to elicit information important to understand the investigation practice. After my observation, the interview conversations were focused upon issues arising during the course of ethnography, seeking verification of the data extracted earlier in the fieldwork. As the interviewees were comprised of different strands of the legal professional, the ethnographic data could be cross-checked by having conversations with those legal actors. In light of this comparison, I discovered that police practice in site A was certainly not restricted to the particular institutional culture or area I observed. Nevertheless, it remains the case that this research may not represent overall police practice across China, given its vast landscape and cultural diversity. In this regard, further studies with a more comprehensive scale are required in future.

The Role of the Police in the Criminal Justice System

The Chinese police force, or the Public Security Bureau (PSB), is the most powerful institution within the criminal justice system. The police have extensive powers in relation to criminal investigation. They can make decisions on issues such as granting bail, imposing residential surveillance of up to 6 months and detaining a suspect for up to 37 days, as well as searching persons or premises and seizing material evidence. Their power also extends to a wide range of intrusive administrative penalties, which enable them to punish and dispose of minor offences directly. The vast array of policing powers is intended to allow them to maintain social justice and the security of the State. According to Article 50 of Criminal Procedure Law 2012 (CPL, 2012), the police, as well as prosecutors and judges, have a legal duty to obtain ‘both exculpating and inculpating evidence to prove the case and to be loyal to the facts’. Their task is to gather evidence objectively and put defendants against whom there is sufficient evidence of guilt before the courts.

Despite the role of impartial investigator portrayed by this legal rhetoric, the police have rarely been quasi-judicial in practice (Chen, 2002; Xu, 2010). As a vital part of state apparatus, their chief function is to act proactively to curb rising crime rates and to maintain the stability of the political regime (Wang, 2014). Police in China are responsible for ‘everyday forms of social management’, preventing and suppressing various scales of protests and revolts (Wang, 2014). This in many ways is in congruence with Herbert Packer’s model of crime control (Packer, 1968). In contrast with a system that upholds a value in protecting the citizen from being unjustifiably penalized (the due process), Chinese criminal justice is very much in keeping with the model that prioritizes the efficient repression of crime. This is evident in a series of ‘Hard Strike’ (yanda) anti-crime campaigns launched in the early 1980s, with ongoing sequences through to the current time. Politically motivated, Hard Strike is a movement to swiftly and harshly combat crime waves (McConville et al., 2011; Tanner, 1999; Trevaskes, 2007). The processing of cases during the campaigns was handled in an extremely rapid manner, with investigation and prosecution merging and the judiciary becoming a rubber stamp on the procuratorate’s decision on the conviction of the accused (He, 2008). The police have been constantly demanded to combat crime harshly and swiftly, with the rest of the criminal process operating on administrative rather than judicial principles. The crime control model exemplified by Chinese criminal justice policy is arguably not merely a crime repressing mechanism geared towards social security. According to Trevaskes (2007), efficiently apprehending and punishing criminals have been seen as a political statement, which demonstrates the power and legitimacy of the political authority.

Until 2015, the police’s crime control performance was evaluated by the appraisal system (jixiao kaohe zhidi). This is a managerial framework devised to incentivise legal institutions to engineer the success of prosecutions (Li E, 2010; Zhu, 2009). The police were allocated an annual target number of suspects to arrest and prosecute. An accomplished task would bring advantages in the form of increase of salary and promotions; failing to fulfil the target would result in sanctions (Interview BPO-1). The appraisal system has effectively shaped the role of the police and has unequivocally defined them as an accusing party, whose actions are underpinned by the presumption of guilt (Wang, 2004; Yuan, 2009). While it has now been abolished, the appraisal system has had a long-lasting impact on the ideology of police operation, which has cultivated a particular mindset for evidence collection. As one police officer informed me:

We don’t put the wrong version into the dossier – it only causes confusion. We just include the one which we believe is true. At the end of the day, we have to try our best to make all evidence look consistent, so that the prosecution decision or final conviction can be secured. (Interview EPO-1)

This was confirmed by a prosecutor, who was familiar with the practice of police investigations. He told me that contradictory facts were often processed by screening statements that were in the accused’s favour:

The police will not let witnesses say something different from the suspect’s guilty confession! If some witnesses’ statements were far away from the facts they [the police] believe to be true, they will just throw them away. (Field note APU-1)

It is also in accordance with the case dossiers that I reviewed. Aside from 65% of police cases that suspects were arrested red-handed, other crime types (such as assault, robbery, theft, rape) the pattern of language in the suspect’s confessions was overwhelmingly similar to the testimonies of the witnesses. In a sample of 64 cases, the keywords used by the witnesses repeatedly reappeared in the suspect’s confessions. In 42 of these cases, the critical details recorded between different statements in the case dossier were so highly matched that the narrative pattern and keywords overlapped. On one occasion, a prosecutor admonished two police officers:

I have to warn you that if you keep making all the details of different accounts exactly the same, the truthfulness of these statements will be doubted. In many circumstances, witnesses cannot remember the exact time. […] So if you just keep the details roughly the same and leave some reasonable discrepancies, the evidence is more convincing. If you try to match everything in the dossier, the effect may be contrary. (Field note APU-55)

This practice partially is directed by the fact that police in China are under enormous pressure. In many parts of China, investigating officers only represent a small proportion of the total workforce within a station. In order to accomplish the given task, shortcuts were sometimes justified. As a police officer explained:

The pressure (of the appraisal system) is immense. […] There are only two investigating officers in the whole station. Based on the crime rate in our area, the target is reasonable. But because we are short of staff, the pressure and workload assigned on each policeman’s shoulders are too heavy. I have family to feed and I work 14 h per day on average. I haven’t celebrated the Spring Festival for many years. I understand that you are interested in law, but following the law is not always practical. If you are in my shoes, you understand what I mean. (Interview DPO-1)

According to Article 195 of CPL, 2012, to convict the defendant, the corpus delicti must be clear and the incriminating evidence should be reliable and sufficient. This means that there must be a chain of inculpatory evidence, with all pieces of evidence relating to each other and pointing to a single fact without reasonable doubt (Article 195 of CPL, 2012; Article 104 of Judicial Explanation, 2012). This is the so-called corroboration rule (yinzheng yuanze) used by the courts to decide the admissibility and weight of evidence. All the statements and written documents in the dossier are required to respond to one another to make the prosecution case unequivocally inculpatory. Thus, it is comprehensible why the practice to ensure the coincidence of details among different documentary evidence in the dossier prevails. This has clear resonance with McConville et al.’s construction thesis. Based upon the available data and taking into account that the researcher had no access to a police station to witness the construction of police dossier, the following sections endeavour to make a sketch of the general principle that guides the process and evaluate whether it is safe to rely on the documentary evidence to determine the guilt or innocence of the accused.

Formulating the Official Version of the Truth

The fact that criminal cases are constructed by, and contingent upon, legal actors, does not imply that the pursuit of objective knowledge is ‘misconceived’ or ‘quixotic’ (Damaska, 1997: 297). Sound criminal justice systems provide rational mechanisms to achieve decisional rectitude. Adversarial systems rely upon opposing parties presenting competing versions of the case, challenging each other’s accuracy and thereby ultimately bringing about a composite picture of the truth. The inquisitorial system entrusts a neutral officer with the power to gather relevant evidence and prove the facts (Hodgson, 2002; Weigend, 2003). Chinese criminal justice does not appear to follow either of the above systems. Relying heavily upon the written evidence produced by the police outside the court and with witnesses rarely being cross-examined, court trials in China are essentially paper hearings. The written documents in the dossier are pronounced by prosecutors at trial. With no access to witnesses, defence lawyers have little means to test the veracity of these statements. Their main role at trial is to advance simple pleas in mitigation (McConville et al., 2011, Chapter 11–12; Field note APU-40, 41 and 43). In common law countries, this type of written evidence would be generally categorized as hearsay evidence, the reliability of which has been profoundly doubted. For instance, in Teper v. R. [1952] AC 480, the Privy Council in England has famously remarked:

[Hearsay evidence] is not the best evidence and it is not delivered on oath. The truthfulness and accuracy of the person whose words are spoken to by another witness cannot be tested by cross-examination, and the light which his demeanour would throw on his testimony is lost.

The existence of the hearsay rule can be justified by a number of factors based on human fallibility, such as defects in perception, memory, sincerity or ability to narrate clearly, of the maker of the statement (Choo, 1996). The guiding objective of this principle is the rectitude of outcome. Since the admission of unreliable prosecution evidence could result in the wrongful conviction of an innocent person, such risk should be reduced by exclusion of hearsay evidence. The confrontation between accused and accuser is regarded as ‘essential to a fair trial in a criminal prosecution’, which has a symbolic meaning that deeply rests within human nature (Summers, 2007). The accused’s right to question the witness is believed to promote openness and guard against coercion, ensuring that the accused can confront any adverse witness testimony and restrain the capricious use of governmental power (Summers, 2007).

In China, written statements obtained by the police are serving as the principal vehicle of proof. Paradoxically Chinese criminal procedure law appears to be specifically dedicated to re-establishing the factual truth of cases. Truthfulness is used as a rule to test the admissibility of evidence, and state officials are obliged to ‘be loyal to factual truth’ (Article 51 of CPL, 2012). Although the accused has the right to avoid self-incrimination, the law requires her to ‘answer truthfully during police questioning’ (Article 118 of CPL, 2012). Contradictory to construction theory, these rules imply a belief that the truth is ‘out there’ to be excavated. Truth is believed to be known by legal actors from the outset of the investigation and guides state officials’ search for evidence. Affirming the official version of truth has been the very first step of constructing the police case in many instances.

McConville et al. (2011) found that the police in China ‘play a relatively subsidiary role in the discovery of crime’; in their statistical survey, 74% of the criminal cases were reported by victims, victim’s family and other civilians. This is in line with the finding of this study. Victims or civilians’ report of crime has often been the earliest recorded account compiled in the case dossiers. Being the initial account of the police case, the statement provided by the victim or other informants has played a significant role in formulating the case fact in the dossier. Of all the investigative dossiers examined (240 cases), the keywords used by the victim⁵ and the way that the crime was depicted have repeatedly reappeared in other documentary evidence, most especially the suspect’s confessions. In certain instances, the language used in the accused’s statements was so starkly similar to the victim’s testimony that only the names and dates involved were different (APU-5). Considering that there were various ways to delineate an incident, it is highly unusual that the suspect and the victim, sometimes from very different social backgrounds, chose to use identical language to report the case. Thus, on one occasion, I asked the police officers by showing the case dossiers:

Researcher:

Why are the languages in those different statements so similar?

Police:

We have to determine the investigative direction in the beginning. The initial account reported by the victim and informant sometimes is the cornerstone of the investigation […] later evidence must reflect on the victim’s account. (Field note APU-13)

It appeared that the police used the information that reached them first, via the victim or other witnesses, as a guideline for interrogation which substantially influenced the formation of the official version of truth. This was subsequently confirmed by an experienced prosecutor, who explained the widespread practice in police investigation:

In practice, police officers always tailor the suspect’s statement to the victim’s account. They force the suspect to say the same thing as the victim said to them. They will ask the victim first and then interrogate the suspect in such a way that the suspect’s statements can dovetail with the victim’s statement. (Interview BPS-1)

Although the initial account may truly reflect the reported crime given the fresh memory of the witness, there are a host of reasons (such as defects in perception, sincerity or ability to narrate clearly) why credibility should not be attached to the statement given by the victim or other informants lightly. Without any close scrutiny and cross-examination of these statements, human errors might well be compounded. Despite the evident risk of this practice, it is the simplest and most straightforward way to secure a case. With all the evidence for trial in written form, the required link between the evidence is embodied in the particular reoccurring words and sentences to achieve the corroboration rule. As the witnesses’ statement has been settled at the beginning of the investigation, ensuring the suspect’s statements is given in a way that dovetails with the witnesses’ statements has become a manageable option to corroborate the evidence and establish an undisputable fact.

However, the witnesses’ accounts are not always available. In cases such as murder, the formulation of the facts is sometimes reliant on a range of factors, including the probability of conviction. In those situations, the crimes eventually presented by the police are likely to be selected for evidential considerations. As such, the crimes presented will often be the most winnable rather than the most serious. For example, in one grievous bodily harm case (guyi shanghai) involving sexual assault, the victim was in an unconscious state and could not provide any valuable testimony. According to his report during the prosecutorial interrogation, the suspect was encouraged by the police to confess an intention of rape rather than wounding, because taking advantage of the unconscious state of the victim could easily satisfy the mens rea of rape pursuant to the criminal law:

The police officer tried to persuade me to confess that I wanted to have sex with the girl. I said no. I would rather confess to a crime carrying a heavier penalty than an offence with a light one that I never committed.⁶ (Field note APU-30)

According to the prosecutor handling this case, the lack of evidence for an assault charge had led the police to reconsider the direction of the investigation, thereby prosecuting the suspect under the crime of rape has a higher chance of success. Applying the same principle, the police in site A chose to investigate cases as a drug possession rather than drug trafficking (CASEA 25, 37 and 38), sheltering prostitution rather than fraud (CASEA 15, 17, 38 and 41) through balancing the strength of the evidence at hand. The official version of truth was selected not because it is loyal to the facts, but it is based upon the probability of conviction.

Recently, reported miscarriages of justice indicate that the version of truth presented by the police may relate to stereotyping, intuition or imagination in specific circumstances.⁷ The facts contemplated thus provide some sort of link to explain the real evidence found relating to the crime in question. Suspects’ statements are seen as malleable and are shaped to fit into the version of truth. The quashed murder convictions of Zhang Gaoping and Zhang Hui are a telling example.⁸ Despite the DNA evidence suggesting that the perpetrator might be someone else, the police insisted on interrogating the suspects to confess, as their minds were made up that Zhang Gaoping and Zhao Hui were the murderers. In this sense, the police case is not objective and independent of legal personnel but rather is a construction in the interest of expediency.

Aligning the Evidence with the Official Version of Truth

Once the official version of facts has been formulated, the next step of the investigation is to confirm the pre-established truth by procuring a confession in writing. There is no requirement of voluntariness within Chinese law. Instead, suspects have a duty to answer police questioning truthfully (Article 118 of CPL, 2012). Despite prevalent reports of torture (see for example Human Rights Watch, 2015), police interrogation remains largely shielded from external scrutiny. Since 2013, video recording has been introduced into the police interrogation. However, this mainly applies to serious crimes where defendants might be sentenced to the death penalty or life imprisonment and its efficacy should not be overstated.⁹ For the vast majority of ordinary cases, interrogation was entirely controlled by the police, with no legal counsels allowed to be present.

Given the importance of investigative dossiers once a case is taken to court, the way that the record is generated is extremely significant. Police interrogation records follow a standardized format. They start with the duration and the location of the interrogation, the name of the interrogator(s) and transcriber and the notification of the suspect’s rights. The law requires that these details must be valid; otherwise, the interrogation record is inadmissible (Article 82 of the Supreme People’s Court Explanation, 2012). Despite this, in practice, the interrogation record was recognized as being subject to change in order to fulfil the legal requirements. In CASEA 23, for example, a prosecutor returned a case dossier back to the police, as the interrogation record had a duration time lasting greater than 12 h, which was forbidden by law. When the case was resubmitted to the procuratorate, the document remained intact, except that the time had been modified (Field note APU-13).

Prosecutors in China used to subject to the Appraisal System. One of their main tasks was to maintain a high conviction rate (Gao, 2013; Zhu, 2009). Hence, they are naturally allied with the police in working to strengthen the persuasiveness of the evidence in the dossier and enhance the possibility of conviction (McConville et al, 2011:387). Police officers would openly consult prosecutors for legal advice and prosecutors would assist with covering up the illegality of police procedure, helping recycle and legitimize tainted evidence (APU-17, 18, 19 and 45). In one instance, a police officer explained to the prosecutor why a specific interrogation date was chosen even though it did not match the time given by other suspects in the same case.

The main problem is that if we write the time that we took them [the suspects] back to the police station, we will have violated the law. […] We have to interrogate the suspects within twenty-four hours after they are detained. However, the first time I interrogated the other three suspects was three days after they were taken back to the police station. […] So I have to put the right date in the statement. (Field note APU-18)

According to an experienced police officer, critical elements in the statements, such as the duration of the interrogation, are usually left blank until the end of the investigation process to ensure that the details in the official record conform to legal requirements (Field note APU-15). As no authentication process is available to ensure the reliability of official records, the police are at liberty to select information that is compatible with the requirements of law.

In McConville et al.’s (1991) research into the construction of prosecution cases in England and Wales, they note that informal interviews might take place ‘off the record’ in order to obtain confessions from suspects. This practice has also been observed to have taken place during the fieldwork, where the police have discretion in filtering and adding accounts to conform to perceived facts. According to one defence lawyer, conversations preceding interrogation would not appear on the written record:

Yesterday I talked to my client, who told me that when he confessed his crime the first few times, the investigating officers did not even write his confession down. That’s because they believed that what he said did not conform to the direction in which the investigation was going. So they did not even record it! (Interview BDL-1)

It is worth noting that the allegations made by a number of suspects were coincidentally similar. During my observational period, 17 suspects claimed that certain interrogations recorded in the dossier did not even take place (APU-30, 32 and 40). In several prosecutorial interrogations, the suspects averred that some interrogations documented in the dossier were merely regurgitated versions of previous interrogations, with certain details changed (APU-40 and 41). For instance, during the prosecutorial interrogation, one suspect claimed that written evidence was fabricated by attributing admissions to him which he did not make:

I did not hide anything from the police but I know the police fabricated my statement. They told me ‘don’t worry, it is the same. Just sign your name on this paperwork and you can leave.’ I glanced at the statement. It was not what I said. Then they asked me to thumbprint it. They asked me to admit that I had taken the bag with me. I did not. I couldn’t admit to something I never did. Why on earth did they do that to me? (Field note APU-30)

Likewise, in another case, the suspect alleged that she was forced to sign a statement which did not reflect the interrogation:

The police officer said that they were going to have lunch and they wanted to finish it off. I expected to read the statement, but they told me that it was what I said and was reluctant to allow me to read. I read two pages and found the details were very different. They put a lot of new things in. They were anxious and just pressurised me to sign. Officer, the document is not what I said! (Field note APU-14)

These assertions were also echoed by the official dossiers examined in this study. In these statements, interaction between the police and suspects was conducted in the format of questions and answers. These accounts were often set out as if they were straightforward narratives, providing a neutral image that the record was faithfully transcribed. Yet when analysing the answers given by suspects, some of them were so implausible that they were obviously constructed in a collaborative effort between the police and the suspect. In one fraud case, for example, one of the interrogating records concerned the personal information of the victims. During the first interrogation, the suspect confessed the names of 18 victims, their telephone numbers, bank account numbers and the names of 20 hotels where the fraud took place. The fraud activities occurred over a 5-year period with the record giving precise times and dates of the fraudulent transactions (APU-6). It is certainly doubtful that the suspect could memorise all the information, including 18 of these 16-digit bank card numbers and 11-digit telephone numbers without a single mistake. In another theft case, the suspect confessed the serial reference numbers of 209 stolen goods, when interrogated by the police for the first time:

An Excerpt of the Record of Interrogation (first time) for a Theft Case:

Police:

What did you steal from this shop?

Suspect:

I stole 49 items of clothing; their serial numbers are: KVI-34-P90, JRV-54-V20, RKI-89-99 V, RSK-39K-90 K, REK-49-PE3, KVQ-90S-VI3, ISIP-39L-GIS, EKS-E30-SLK, KO0-LD-3L8 […]. I also stole 59 pieces of jewellery; their serial numbers are: KOS-39S-SKJ, K39-SKI-SKP, QID-JI3-VIP, SIS-39K-SI3, DI9-3KS-S9 S, ISI-SI9-3SI, DI0-D83-SKI, LA9-S9D-3KD, IDK-3K9-S39, […] I also stole 101 bags; their serial numbers are: IP0-SKI-SKE, KI3-ISS-90 S, SI9-SE9-SIV, SIS-DI8-S83 […]

Police:

Did you confess the truth?

Suspect:

Yes, I did. (Field note APU-6)

In the same case, when the suspect was interrogated for a second time, these large chunks of serial numbers appeared again in exactly the same sequence. When interviewed, some defence lawyers suggested that one of the strategies employed by the police was to replicate the suspect’s statements a couple of times, so that it appeared that the suspect had confessed consistently (BDL-1 and CDL-1, 3). They believed that these repeated confession records are manufactured by copying and pasting the suspect’s previous statements, as inaccuracies and spelling mistakes which occurred in the former account remained unchanged. One defence lawyer told me what he knew about the suspect’s statement:

Many interrogation records are pre-typed before the interrogation. Since computers are widely used today, interrogation statements are worse than they used to be […]. Now we see the same spelling mistakes and the same paragraph styles being repeated in different statements – they just use copy and paste! I read many of the suspect’s statements in the dossier. In the final four statements, the spelling mistakes and paragraphs were exactly the same! (BDL-1)

According to Article 120 of CPL, 2012, suspects are entitled to check the record of statements before they sign. On every page, they are requested to sign their names, which are superimposed with their thumbprints. They are also required to declare in writing at the end of the statement that they have read the content and agree that it is a true reflection. While these measures make the statement seem genuine, there has been strong suspicion that these statements are not conducted verbatim. In nearly one-third of the cases (n = 21) which I monitored, the suspect had complained to the prosecutors that the police distorted, exaggerated, misunderstood or even falsified their statements. A number of suspects (n = 27) claimed that they were not given the time to read the interrogation record before they signed. The conversations below illustrate this point: Prosecutor:

During police questioning, you said that you wanted to keep them [the drugs] for yourself because you had a quarrel with your ex-boyfriend who had these drugs.

Suspect:

No, it is not true. I did not say anything like that. It was made up by the policeman. […] I was only asked to sign. (Field note APU-21)

In a drug dealing case, a prosecutor asked the suspect whether he signed the record of interrogation: Prosecutor:

Did you sign the record?

Suspect:

They did not let me read what they wrote. They forced me to sign my name. (Field note APU-21)

The frequency of such repeated allegations casts doubt on the reliability of statements and there is no safeguard to guarantee authenticity. During my interviews, all the police officers claimed that the use of torture has diminished substantially in the past few years. Nonetheless, suspects occasionally reported that violence and psychological compulsion were employed to force them to sign the interrogation records. Aside from physical violence (n = 4), a greater number of suspects (n = 13) alleged that their signatures on the interrogation records were extorted by threatening the safety of their families (APU-22). These suspects subsequently told prosecutors and judges that they were compelled to give certain answers desired by the police:

They [the police] threatened me and asked me to sign the statement. If I did not sign, my family would be in danger. I did this to protect my family. (Field note APU-47)

The law requires the defence to provide details of the alleged abuse before the matter will be investigated and the illegally obtained confession excluded. However, since the police control the investigation process, gathering the evidence required to prove the unlawfulness of interrogation is extremely difficult. As the suspect is invariably interrogated multiple times, the exclusion of one interrogation record hardly makes any difference (Human Rights Watch, 2015; McConville et al, 2011:339). As a result, in most instances, the confession in question is admitted by the court and allegations dismissed.

As a general feature, all the confessions disputed by the suspect have been aligned with the contemplated official version of truth. Confession is an extraordinarily malleable form of evidence that can be shaped in a way to enumerate and be consistent with other evidence contained in the dossier. Compared to the suspect’s confessions, less is known to what extent other documentary evidence, such as witnesses’ statements, is recorded in a reliable manner. As the prosecutors in site A only interview victims in rape and assault cases, the number of cases in which witnesses’ written accounts were verified during my observation was very small (n = 7). Nonetheless, three witnesses disputed the accuracy of their statements as assembled in the dossier, asserting that the police did not record their testimonies faithfully. One of the cases concerned the sexual assault of a child. The mother of the victim told the prosecutor that the suspect had sexually assaulted her daughter twice and had stolen 2000 yuan from their apartment. However, the mother’s statement in the dossier only recorded one account of sexual assault and did not mention any stolen money. When asked whether she made the same report in the police station, the mother replied:

I don’t know why the police did not write them down, but I told them exactly the same thing: the same dates and the missing money. […] They just asked me to sign and I did not check the statement.

According to the prosecutor, since the suspect in this case did not confess the second account of the rape and theft, the police decided to alter the fact given by mother so that it could be aligned with the confession. Another type of evidence contained in the dossier, which carries a great deal of weight in adjudication, is crime scene evidence (xianchang kanyan). The most common form of crime scene evidence is a record of crime scene identification (xianchang zhiren), which usually consists of one or two photographs, invariably featuring the suspect pointing to a location or an object, with an annotation (such as ‘the suspect indicates the crime scene’) added to illustrate its meaning (APU-14, 36 and 43). This evidence was included in all dossiers I examined and is routinely accepted by the courts. In Western jurisdictions, crime scene evidence is the collection of real evidence or forensic analysis of evidence retrieved from a crime scene. However, in China, crime scene evidence is observed to be a by-product of confessions and is derived from police interrogation. During the interview, one defence lawyer, who was also an ex-police officer, explained to me why this type of evidence is not trustworthy:

The crime scene identification is a bit … [worrisome]. The police will take the suspect to a nominated place and take a photograph of him. This is the so-called identification evidence. In truth, the suspect has not identified anything at all. The suspect might not even know where the crime scene was until the police tell him where it was. Some migratory offenders are not familiar with a place as big as site C. They have no idea where the crime scene was. (Interview CDL-1)

Real, tangible evidence, which often carries considerable weight in proving the facts of an issue, is often not attached to the case dossier, and so is not passed on and presented in court. Hence, it cannot be cross-examined by the defence at trial.

The Construction of Police Dossiers Within the Context of the Chinese Criminal Process

The way that police cases have been constructed should be recognized as an entrenched practice which is consistent with the operation of the bureaucratic institution. Although the police are accorded a vast arrange of powers to deal with crimes, the decision-making process is hierarchical and most operational decisions must be reported to the Chief Officer for permission, leaving little discretionary power to the individual officer (Interview BPO-3). The lack of discretional power and heavy caseloads reduce enthusiasm, turning the investigation into a routine paper-based exercise (Interview HPO-1). In a substantial survey conducted by Scoggins and O’Brien (2015), they found that police in China are frustrated with the heavy caseload and administrative drudgery, which has led to low morale and low productivity. The limited wages and discontent with their work, as well as pressure imposed by various crime control tasks, have made the police inclined to process similar cases in a cost-efficient fashion (Field note APU-17).

It should also be noted that there is very little option available for the police to determine how the case should be presented. The law demands congruence of the evidence of the case, which requires the details of evidence must be in accordance with one another. Embodied in the case dossier, this legal prerequisite is integrated into the standard paragraphs and formatted language. Over the routinization of investigative activity, the police have habitually developed the ability to typify the situations of each individual case. With the assistance of the legal department within the police station, the fact details are polished into recognizable terms; idiosyncrasies of the cases are diminished to similar stories (Interview BPO-2 and 3). After different levels of editing throughout the hierarchy of the station, these dossiers are ready to despatch to the procuratorate for further examination.

It has been two decades since China initially introduced the reform of criminal justice in the direction of an adversarial system. However, this feature remains weak and the structure of the system is intact compared to its predecessor of socialist criminal justice: the criminal process is unilaterally dominated by the activities and decisions of three core institutions. The concept of the iron triangle, a coalition of the police, the procuratorate and the judiciary still defines the criminal process in China, leaving the defence with little standing, status or influence within the system (Fu, 2007 ; Li H, 2010; McConville et al, 2011; Tanner, 1999). The interrelationships between the three institutions are centrally embodied in the processing of case dossiers, which has been a focal point at every stage of the criminal procedure. After the police investigation has been completed and the case dossier has been built, the case will be transferred to the procuratorate for review, deciding issues in relation to prosecution. Although prosecutors in China officially have a supervisory role and are responsible for overseeing the legality of police investigations, in reality, the prosecutor and the police are closely aligned and collaborate with one another (McConville et al, 2011; APU-18; BPS-1). As the data revealed earlier in this article, it has been observed that the police are not only given a free hand in constructing cases against the accused but also are consistently assisted by prosecutors to ensure that illegally obtained evidence appears legitimate.

Being part of the iron triangle, the relationship between the courts and the procuratorate is intensified and institutionalized, sharing core values and strategically allied (McConville et al, 2011; APU-42, 43, 45 and 46). Since the case dossiers are constructed in such a way that the defendant’s confessions are always cemented by reference to other evidence available in the case dossier, the courts are unlikely to re-establish the credence of evidence contained in the dossier, which could potentially forfeit all the previous efforts made by their allies and undermine their relationship. Hence, there is no surprise that judges have rarely advocated the idea that witnesses should testify at court. As a matter of fact, judges’ preference for written dossiers over oral testimonies has been one of the key factors that have prevented witnesses from entering the courtroom (Mao and Yuan, 2015; Zuo and Ma, 2005). For instance, one judge commented that all judicial decisions should be based on case dossiers:

We are responsible for the truth that is embodied by the police dossier. (Interview ATJ-1)

Although defence lawyers partake in constructing defence cases, their strategies are highly restricted. Marginalized by legal institutions, they are often forced to take great risks when engaging in active defence preparation (Fu, 2007; Hou and Keith, 2011; Liu and Halliday, 2008, 2011; McConville et al, 2011). The direct danger emanates from Article 306 of the Criminal Law of 1997. This outlines a perjury offence which can be committed by the defence counsel and has frequently been utilized by the police and the prosecution to silence active defence lawyers. According to this article, a defender will be found guilty if he is involved in ‘destroying or forging evidence, helping any parties to destroy or forge evidence, coercing the witness or enticing him into changing his testimony in defiance of the facts or giving false testimony’. The great irony of this law resides in the fact that although the police and prosecutors are more likely to be involved in perjury-related activities, the law is formulated against the legal actor who might reveal this falsification. Fu Hualing suggests that when exculpatory evidence is sought by defence lawyers against the prosecution case, prosecutions are ‘immediately switched to Article 306, abandoning the former charge’ (Fu, 2007). Since 90% of cases against defence lawyers prosecuted under Article 306 result in acquittal (Hou and Keith, 2011), it is suspected that these prosecutions are malicious (Fu, 2007).

Faced by public outcry to abolish this article (see Li H, 2010; Zhang, 2005), CPL 2012 attempted to attenuate direct antagonism by debarring the PSB who investigated the offence represented by the defence lawyer from being involved in the investigation of the suspected perjury case. However, reports from social media indicate that evidence acquisition remains ‘the high-voltage zone’ for defenders (Beijing Shangquan Law firm, 2014). As a result, the vast majority of defence lawyers have adopted skills of self-censorship, avoiding the use of zealous defence strategies. According to Beijing Shangquan Law firm’s annual survey of 2014, 65.1% of defence lawyers believe that they should not collect evidence during the pretrial phase. Their case construction is mainly limited to the investigative dossier they can request from state officials.

There are many notorious cases in which innocent defendants have been falsely convicted. These cases (such as the She Xianglin and Teng Xinshan cases) share a similar pattern. After suspicions are raised against a certain person, an investigation is begun, based on an official version of the truth formulated by the police. With no reliable witness testimonies or forensic analysis, the formulated truth is reinforced by extorting confessions from the accused. In most reported instances, torture was used to obtain coerced confessions. Prosecutors in these cases failed to check the reliability of the evidence they provided. At trial, no witnesses were cross-examined and the defence counsels were unable to offer meaningful assistance to help defendants escape conviction for crimes that they did not commit. These cases have thoroughly exposed the structural weakness of the system that epitomizes an extraordinary confidence in the ability of the police to discover truth. However, there has been no agenda in the legal reform to address the critical issues. On the contrary, centrality of investigative dossiers is further consolidated and continues to frame criminal justice in China.

Concluding Remarks

In China, investigative dossiers are commonly used by courts to determine the guilt or innocence of defendants in lieu of witnesses’ oral testimony. In examining how these investigative dossiers are constructed by the police, my empirical data have suggested that integrity of the documentary evidence is suspect. Within the criminal justice framework, no legitimate process is currently available to ensure the authenticity of the transcripts taken from pretrial interrogations or other acts of investigation. The way that the facts of a case are presented illustrates that the Chinese criminal justice system is not specifically designed in a way that scientific evidence can be meticulously relied upon and witnesses’ testimonies can be skilfully cross-examined in order to discover truth. The dominant role played by the police and the prosecution in shaping evidence means that there is a lack of the functional equivalence of defence construction required to formulate a competing version of facts during the pretrial stage. In this sense, the Chinese criminal justice system is structurally weak and fails to function as a truth-finding process.

Footnotes

Acknowledgment

I would like to thank the Journal’s two anonymous reviewers for their valuable comments on an earlier draft of this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Beijing Shangquan Law firm (2014) The Annual Report of the Implementation Effect of the New Criminal Procedure Law 2013. Available at: http://www.sqxb.cn/content/details16_1644.html (accessed 9 March 2014).

Chen

(2001) Survey Report on the Problems in the Implementation of the Criminal Procedure Law. Beijing: Zhongguo Zhengfa University Press.

Chen

(2002) Restrictions and separation of power: Police power in criminal justice. Journal of Northwest Institute of Political Science and Law 1: 1–3.

Chen

(2007) Let witnesses go to the court: The study of witnesses’ testimony. Shangdong Police College Review 19: 2.

Chen

(2009) The truth in criminal process and legal reform in the standard of proof. Politics and Law Review 2: 1–17.

China Law Year book Editorial Committee (2009–2011) China Law Year Book. Beijing: Press of Law Year book of China.

Choo

(1996) Hearsay and Confrontation in Criminal Trials. Oxford: Clarendon Press Oxford.

Damaska

(1997) Truth in adjudication. Hastings Law Journal 49: 289–308.

Fox

(2004) Observations and reflections of a perpetual fieldworker. The Annals of the American Academy of Political and Social Science 595(1): 309–326.

10.

(2007) When lawyers are prosecuted: the struggle of a profession in transition. Social Science Research Network. http://papers.ssrn.com/sol13/papers.cfm?abstract_id=956500 (accessed 3 September 2013).

11.

Gao

(2013) The acquittal and its resolve proceedings: Analysis on the low acquittal rate. Law and social development 19(4): 65–80.

12.

Goldman

(1986) Epistemology and Cognition. Cambridge/London: Harvard University Press.

13.

(2008) The Construction of the People’s Court: 1978-2005. Beijing, China: China Social Science Press.

14.

(2014) Chinese Criminal Trials: A Comprehensive Empirical Inquiry. New York: Springer.

15.

Herting

Stein

(2007) The evolution of Luhmann’s systems theory with focus on the constructivist influence. International Journal of General Systems 36: 1–17.

16.

Hodgson

(2002) Constructing the pre-trial role of the defence in French criminal procedure: An adversarial outsider in an inquisitorial process? International Journal of Evidence and Proof 6: 4–8.

17.

Hou

Keith

(2011) The defence lawyer in the scales of Chinese criminal justice. Journal of Contemporary China 20(70): 379–389.

18.

Human Rights Watch (2015) Tiger stool and The head prisoner: The torture used by the Chinese police against the suspect. Human Rights Watch. 13 May 2015. Available at: https://www.hrw.org/node/278012 (accessed 7 July 2015).

19.

Kvale

(2007) Dominance through interviews and dialogues. Qualitative Inqiuiry 12(3): 480–500.

20.

(2010) The predicament caused by Article 306 of criminal law and how to resolve it. Procuratorate Forum 3: 65–66.

21.

(2010) Li Zhuang case: Examining the challenges facing criminal defense lawyers in China. Columbian Journal of Asian Law 24: 129.

22.

Liu

Halliday

(2008) Dancing handcuffed in the minefield: Survival strategies of defense lawyers in China’s criminal justice system’. Center on Law and Globalization Research Paper (08-04). Available at: http://ssrn.com/abstract=1269536 (accessed 9 September 2014).

23.

Mao

(2014) Why video recording has become alienated? Business Review (14). Available at: http://magazine.caijing.com.cn/2014-05-11/114171343.html (accessed 11 May 2014).

24.

Mao

Yuan

(2015) Observation of the efficacy of witnesses attending at trial from the perspective of new criminal procedure law. Journal of Jiangxi Police Institute 182: 107–111.

25.

McConville

Choongh

Wan

. (2011) Criminal justice in China: An empirical enquiry. Cheltenham/Massachusetts: Edward Elgar.

26.

McConville

Sanders

Leng

(1991) The Case for the Prosecution: Police Suspects and the Construction of Criminality. London: Routledge.

27.

Sanders

(1987) Constructing the case for the prosecution. Journal of Law and Society 14(2): 229–253.

28.

Packer

(1968) The Limits of the Criminal Sanction. Stanford: Stanford University Press.

29.

Scoggins

O’Brien

(2015) China’s unhappy police. Asian Survey Available at: SSRN http://ssrn.com/abstract=2563304 (accessed 9 September 2015).

30.

Searle

(1995) The Construction of Social Reality. London: Simon and Schuster.

31.

Shi

(2002) The study of witnesses’ attending to courts. Legal Review of National Prosecutor’s College 10: 60–68.

32.

Shi

(2013) Reforms of witnesses’ appearance at trial and its comments. Shandong Social Science 4(212): 18.

33.

Summers

(2007) Fair Trials: The European Criminal Procedural Tradition and the European Court of Human Rights. Oxford: Hart Publishing.

34.

Tanner

(1999) Strike Hard: Anti-crime Campaigns and Chinese Criminal Justice, 1979-1985. Ithaca, New York: East Asia Program-Cornell University.

35.

Trevaskes

(2007) Severe and swift justice in China. British Journal of Criminology 47: 23–41.

36.

Wang

(2004) The conviction rate and the appraisal system. The People’s Procuratorate 9: 49–50.

37.

Wang

(2014) Empowering the police: How the Chinese communist party manages its coercive leaders. The China Quarterly 219: 625–648.

38.

Weigend

(2003) Is the criminal process about truth: A German perspective. Harvard Journal of Law & Public Policy 26: 157–170.

39.

(2010) Taming Police Discretion Under the Reformation of Policing. Beijing: China Law press.

40.

Yuan

(2009) The relationship between the appraisal system and the legal institutions: From the perspective of acquittal. Kangding Normal Higher College Review 18: 56–59.

41.

Zhang

(2014) Witnesses’ appearance at trial and its implications. Hebei Law Review 19(3): 150–151.

42.

Zhang

(2005) On the abnormal occupational hazards for Chinese defence lawyers. Journal of Shandong Police College 1: 3–5.

43.

Zhu

(2009) The appraisal system in the criminal justice system. Law and Social Science 1: 5–15.

44.

Zuo

(2005) A theoretical clarification on witness’s presence in criminal trial. Chinese Legal Science 6: 14–18.