CN/EN

All
  • All
  • Product Management
  • News and Information
  • Introduction content
  • Business outlets
  • Frequently Asked Questions
  • Corporate Video
  • Corporate Portfolio

In 2002, Professor He Jiahong of Renmin University of China predicted: “Perhaps we are about to enter a new era of judicial proof—the era of electronic evidence.”[1]


 

In 2016, Professor Fan Chongyi of China University of Political Science and Law stated: “Technological revolution alone cannot automatically usher in a new era of judicial proof. Only by fully recognizing the significance of electronic evidence, undertaking institutional reforms and legal refinements, and establishing a scientifically sound framework for regulating electronic evidence—ensuring that such evidence is used lawfully, reasonably, and comprehensively—can we enable it to fulfill its intended functions and roles. Only then can we confidently assert that we have entered the springtime of a new era of judicial proof—the age of electronic evidence!”[2]


 

The two arguments above—one appears “radical,” the other leans toward “composure.” Which is right and which is wrong? Readers, please form your own judgments. As a legal professional who has been engaged in the study of electronic evidence for more than 25 years, I am not inclined to offer abstract prognostications; rather, I prefer to take a concrete, albeit limited, look at the matter.


 

I contend that the digitization of traditional evidence and the emergence of novel forms of electronic evidence constitute the fundamental trend shaping the transformation of judicial proof in the global digital age. Since the advent of the “big data era”[3], electronic evidence has entered a new phase marked by generational shifts, giving rise to a distinctive new paradigm underpinned by vast volumes of data. Regardless of how one views the age of electronic evidence, the legal community must undertake a comprehensive re‑engineering of its practices, theories, and institutional frameworks.


 

Moreover, establishing a mindset of robust evidentiary reasoning is, in fact, the precursor to conscious and deliberate action!


 

“Big‑Evidence Thinking” — this key concept is both the central insight behind my book, “Essentials of Electronic Evidence Review,” and the secret code that opens the door to readers’ study of “Electronic Evidence Investigation Skills.” It comprises five essential principles, which I will now elaborate.


 

I. Believing in the true principle of “evidence is king”


 

The essence of “grand‑evidence thinking” lies first and foremost in “grand” evidence—namely, genuinely valuing evidence, or, as it is often put, “placing evidence above all else.”


 

In the eyes of investigators, there are no facts—only evidence. Once a case arises, the truth can only be established through evidence. In the 18th century, the British master of evidence law, Jeremy Bentham, observed that evidence is the foundation of justice.[4] In the 1920s, the American scholar Benjamin Cardozo cautioned: “What most often lies at the heart of controversy is not law, but fact.”[5] And in the 1950s, the renowned Soviet jurist Vyshinsky remarked: “The art of adjudication is, in essence, nothing more than the art of utilizing evidence.”[6] As a Chinese proverb puts it, “Litigation is all about presenting evidence”; and both Chinese and foreign legal maxims alike affirm: “Facts speak louder than eloquence.”[7] All these expressions point to the same fundamental principle: the paramount importance of evidence in judicial proceedings.


 

The quintessential principle underlying this approach is “evidentiary adjudication,” which is regarded as the “emperor’s rule” of modern litigation. The origins of this doctrine have been lost to the mists of history. According to research by Professor Chen Guangzhong and other scholars, in 1876 Japan enacted the “Law on Judging Crimes Based on Evidence,” revising the earlier provision in the “Revised Statutes” that stipulated, “All criminal cases shall be concluded on the basis of oral confessions,” to read, “All criminal cases shall be decided on the basis of evidence.”[8] This groundbreaking provision has endured to the present day and has exerted influence across numerous countries and regions. For instance, Article 317 of Japan’s Code of Criminal Procedure provides: “In ascertaining the facts, one must rely on evidence.”[9] Similarly, Article 307, Paragraph 1 of South Korea’s Code of Criminal Procedure states: “In ascertaining the facts, one must base it on evidence”;[10] and Article 154, Paragraph 2 of Taiwan’s Criminal Procedure Law likewise stipulates: “Factual findings regarding a crime must be based on evidence; without evidence, no criminal fact may be established.”[11] What these provisions share is their emphasis on the principle that “the ascertainment of facts must be grounded in evidence.”


 

The development of China’s evidentiary legal system has been marked by repeated setbacks, and the naturally emerging formulation has been “based on facts.” This principle was first enshrined in the Criminal Procedure Law of the People’s Republic of China, promulgated in 1979. Subsequently, it was reaffirmed in numerous revised versions of the Criminal Procedure Law, as well as in the Judges Law, the Prosecutors Law, the Supervision Law, and other statutes. Later, the legal formulation was refined to read: “In ascertaining the facts of a case, evidence must be the sole basis.”[12] The highest judicial authority has observed: “The ascertainment of the facts of a case is neither the personal judgment nor the individual decision of any judicial officer, nor the judgment or decision of any single judicial organ—be it the public security organs, the procuratorates, or the people’s courts—or of any case-handling body or organizational unit within the public security, procuratorial, or judicial systems. Nor is it a ruling based on the exercise of leadership authority; rather, it must be a determination reached solely on the basis of evidence.”[13] “Before the evidence, there is no room for arbitrary power.”


 

However, the creed of “evidence is king” and the statutory principle of “evidentiary adjudication” have failed to escape the fate of being little more than “famous paintings hanging on the living-room wall.” In China, judicial practice rarely reflects the “evidence is king” ethos, let alone adheres to the statutory mandate of “evidentiary adjudication.” As I wrote this book, revisiting one concrete case after another involving electronic evidence, I came to deeply appreciate how the dual‑edged framework of “evidence is king” and “evidentiary adjudication” is routinely ignored—or even discarded—entirely. This has given rise to a host of anomalies in judicial practice.


 


 

One of the phenomena is the “dominance of confessions.” In ancient China, practices such as “torture before interrogation” and “confession under duress” were once regarded as sacrosanct, while in the West too, it was believed that “a judgment must be based on complete evidence; without such evidence, no verdict may be rendered… a defendant’s confession, whether obtained through torture or not, constitutes half‑proof at best…” Consequently, confessions unilaterally came to be seen as the most reliable form of evidence. Following the decade‑long devastation of the Cultural Revolution, China undertook a thorough self‑critique of such harmful legacies as “extracting confessions from defendants even in the absence of evidence,” “truth emerges under the heavy stick,” and “one person’s confession is heard, two people’s are trusted, three people’s seal the case”[14], and resolutely opposed the doctrine of “confession supremacy.” The 1979 Criminal Procedure Law, in Article 35, explicitly stipulated: “In adjudicating all cases, priority must be given to evidence and thorough investigation; confessions alone shall not be relied upon. Where there is only the defendant’s statement and no other evidence, the defendant may not be found guilty nor sentenced; where, in the absence of the defendant’s statement, the evidence is sufficient and conclusive, the defendant may be found guilty and sentenced.” Nevertheless, in practice, the old habit of “confession supremacy” has made a comeback. Beyond the persistent yet officially prohibited use of torture to extract confessions, arbitrary fragmentation of cases and the abusive application of plea bargaining have also emerged as sinister tools for undermining the principle of evidence‑based adjudication.


 

Joint criminal cases are, in principle, tried as a single consolidated proceeding, with separate trials reserved as an exception. Disjointed case handling refers to the practice, in criminal proceedings, of splitting multiple suspects or distinct criminal facts that were originally treated as a single case into several independent cases for separate adjudication. In China, the approach to separate trial of joint offenses has evolved from a passive to an active mode. The 2021 Interpretation of the Criminal Procedure Law for the first time articulated general norms governing the separate trial of joint offenses; however, it suffers from shortcomings such as insufficient clarity regarding its underlying objectives, an unclear scope of applicability, and inadequate procedural details. Consequently, it fails to address persistent issues in practice, including arbitrary expansion of the scope of separate trials, administrative overtones in decisions to split cases, inadequate safeguards for the right to confrontation, and the undue extension of res judicata effects from prior proceedings.[15]


 

As in Case No. [16], involving the crime of providing programs and tools for intruding into and illegally controlling computer information systems—specifically, the cases of XX Company, Yang XX, and others—outlined in Chapters 8 and 15 of this book, the prosecution’s most critical evidence was an expert opinion assessing the functionality of the software at issue: “The submitted program utilizes WeChat servers to enable WeChat login and possesses the capabilities to retrieve WeChat‑related information, upload device information, and obtain users’ account credentials. This functionality circumvents the security safeguards of the computer information system, thereby obtaining unauthorized access to sensitive system data, and thus qualifies as a program designed to intrude into the WeChat system.” However, courtroom cross‑examination revealed serious flaws in this expert assessment, including an unclear source of the samples examined and a procedurally unreliable appraisal process. In its first‑instance judgment, the court further acknowledged: “Upon examination, it is evident that the forensic report contains several deficiencies, such as verifying hash values only for a portion of the samples, failing to disclose details about certain tools used in the analysis, leaving the origins of some samples uncertain, and rendering the conclusions ambiguous. Accordingly, this forensic report should not serve as the basis for a final verdict; therefore, we accept the defense counsel’s relevant arguments.” Logically, once such pivotal evidence collapses, the court ought to render a not‑guilty verdict in accordance with the principle of evidentiary adjudication. Yet, surprisingly, the first‑instance court divided the case—concerning one corporate entity and six defendants appearing together—into two separate judgments, each imposing relatively lenient sentences. Moreover, both judgments repeatedly relied on confessions from co‑defendants to establish the facts, for instance stating: “Upon investigation, the testimonies of defendants Yang XX and Huang XX, along with the written statements of witnesses Lü X and Song X, confirm…” In reality, the so‑called witnesses Lü X and Song X were, in fact, defendants being tried in another case, while the statements of Yang XX and Huang XX were treated in the other judgment as witness testimony pertaining to a separately adjudicated matter. Given such questionable procedural maneuvers, it is hardly surprising that, on appeal, the defendants contended: “The verbal evidence presented by Huang XX and Yang XX in this case, as well as that of Song X and Lü X in the other case, all constitute defendant confessions. The original judgment’s reliance on these confessions to establish the criminal facts violates the Criminal Procedure Law’s evidentiary rule prohibiting the use of defendant confessions alone as the sole basis for conviction.”


 

From an evidentiary perspective, the reform of the leniency system for guilty pleas and acceptance of punishment seeks to clarify the evidentiary value of confessions, construct an accusation‑centered evidentiary framework anchored in oral statements, and thereby fulfill the task of criminal proof.[17] The system establishes a confession‑centric evidentiary regime, with voluntariness serving as a crucial safeguard for the authenticity of such statements. On this basis, at least according to official pronouncements, the leniency system has not relaxed its adherence to rigorous evidentiary standards. From a causal standpoint, the plea of guilt is both the rationale for procedural simplification and the primary reason for appropriately lowering the evidentiary threshold in summary proceedings.[18] Some scholars further contend that the proof of the defendant’s voluntary confession must meet the “beyond a reasonable doubt” standard to ensure its reliability, while the proof of the factual elements of the offense and sentencing factors may be reduced to a “general conviction,” thus streamlining the scope and methods of evidence collection.[19] Moreover, by imposing varying degrees of stringency on the evidentiary investigation procedures for cases involving guilty pleas and acceptance of punishment, the reform effectively results in an implicit lowering of the evidentiary standard despite the relaxation of procedural requirements.[20] This tendency is even more pronounced when the leniency system is abused.


 

For example, in Case No. [21] concerning the crime of illegally obtaining data from a computer information system—specifically, the ×× Network Company case—discussed in Chapters 4 and 15 of this book, the defendant company pleaded guilty and accepted punishment, and all five individual defendants also pleaded guilty and accepted punishment (with two of them being decided not to prosecute during the review-and-prosecution stage). Each admitted to the unlawful act of illegally obtaining data from a computer information system. However, after my legal team and I intervened, we conducted, for the first time in this case, a thorough examination of the electronic evidence. The factual findings established through court proceedings were as follows: (1) Third-party companies and individuals fabricated evidence to seek improper gains, including both the motive and the actual conduct of fabricating materials to file police reports; objectively, they engaged in falsifying evidence regarding property losses, forging forensic examination records prepared by public security authorities, and fabricating data obtained from Alibaba Cloud. (2) Virtually all objective evidence was found to be false and thus inadmissible as the basis for conviction, including Public Security (Police Branch) Forensic Examination Record No. [2019] 032901 and the associated computer hard drives, as well as the “×× Examination.ZIP” file, which lacked legality and authenticity; likewise, Public Security Maritime (Dual) Evidence‑Request Notice No. [2019] 000015 and the information provided by Alibaba Cloud likewise lacked legality, authenticity, and relevance; furthermore, the five image files corresponding to four IP addresses likewise lacked legality, authenticity, and relevance; and finally, all forensic appraisal reports issued by the designated appraisal institutions could not be used as evidence to substantiate the charge that the defendant company had illegally obtained data. In addition, the supplementary “Work Explanation” submitted by the public security authorities likewise lacked legality and authenticity. Ultimately, after three years of litigation, the case concluded with the procuratorial organ withdrawing the prosecution and deciding not to prosecute.


 

Looking back, this case arose because the public security organs and the procuratorial organs failed to conduct the requisite, proper evidentiary review of this plea‑bargaining matter in advance, nor did they arrange for the key witnesses and police officers to appear in court to testify afterward. As a result, the prosecution essentially amounted to an accusation based solely on the defendants’ confessions, with no other corroborating evidence. Moreover, the statements of the five defendants exhibited repetitive admissions and suggestive questioning—indeed, their third and fourth statements were virtually identical to their first—rendering them unreliable, inadmissible, and lacking both authenticity and legal validity. Consequently, such statements must not be admitted as evidence.


 

The second phenomenon is the proliferation of “situation reports.” “Situation reports” refer, in criminal proceedings, to a collective term for work‑related explanations, status updates, and clarifications issued—under the official seal of an institution—by investigative organs or the investigation divisions of the procuratorial organs, addressing issues that have arisen in a criminal case or require resolution. [22] China’s Criminal Procedure Law does not list “situation reports” among its eight statutorily recognized forms of evidence; nor do any of its provisions formally recognize such materials. Even from a substantive perspective, existing legal provisions primarily permit investigators and others to “rectify” or “rationally explain” the legality of evidence collection by offering “explanations,” [23] with such explanations, in principle, being required to be presented “in court to clarify the circumstances.” [24] The Supreme People’s Court’s more permissive interpretation of these materials has further exacerbated their overuse in practice. Article 135, paragraph 3, stipulates: “Explanatory materials submitted by the prosecutor to demonstrate the legality of the evidence‑gathering process must bear the signatures of the relevant investigators and interrogators and be affixed with the official seal of the institution. Materials lacking such signatures or seals shall not be admissible as evidence. Moreover, such explanatory materials may not, by themselves, serve as the basis for establishing the legality of the evidence‑gathering process.” If we liken litigation to a game of poker, it becomes readily apparent that today’s myriad “situation reports” have effectively become “wild cards”—capable of standing in for virtually any card. This development is difficult to comprehend.


 

In the case of Yan XX’s fraud conviction, various “statements of circumstances” proliferated in court. One type concerned the legality of computer forensics—for example, a “statement of circumstances” dated April 27, 2024, signed by two police officers, Zhang XX and another, purporting to establish the provenance of file No. 11.xlsx on the defendant Ma XX’s computer. It was subsequently discovered that Officer Zhang XX was not an investigating officer and that his signature was forged. A second type consisted of “statements of circumstances” substituting for official records of electronic‑evidence examination and inspection—such as the “Electronic Data Extraction Statement” prepared by the public security authorities to document their extraction of WeChat chat logs between witness Tang XX and the defendant Yan XX. The resulting 133-page transcript of those chats constituted the most critical evidence of the charge; however, it was later revealed that the purported on‑site examination of the mobile phone at the detention center was entirely fictitious, that the signatories Yan XX and Zhang XX were not duly qualified investigators, and that the witness Wang XX was a “professional” signer who had never been present at the scene. A third type comprised “statements of circumstances” bearing the official business seal of the witness’s organization, yet devoid of any individual signatures, with content pertaining to the provision of access to archives of goods‑sales records. A fourth type involved the first and second supplementary explanations appended to an audit report stamped with the official seal of an accounting firm; these failed to meet both the formal requirements for a “statement of circumstances” and, moreover, the formal requirements for a “report issued by a person with specialized knowledge on issues of a specialized nature in the case.”


 

Most importantly, the investigation and prosecution in this case have been cloaked by a series of “explanatory statements” that conceal the unlawful core of evidence‑gathering. These “explanatory statements” constitute substantive illegal evidence rather than merely defective evidence; they amount to materially distorted evidence, not evidence that can be reasonably explained. In this case, the “Explanation on the Extraction of Electronic Data” and the screenshots of chat records contained therein directly affect the determination of guilt and sentencing. The extractor performed operations directly on the original media, which neither ensures the authenticity of the data on that medium nor allows for any remedy once the data has been distorted. Even more seriously, the process of extracting electronic data involved deliberate falsification. As Lincoln once observed, “Calling a tail a leg does not make it a leg.”[25] By analogy, no matter how they are handled, “explanatory statements” cannot substitute for legally prescribed evidence.


 

On June 13, 2024, the Supreme People’s Procuratorate issued the guiding case “Intentional Injury Case Involving Zeng Moujia and Others: Rectification of Omitted Co‑defendants in the Investigation,” which explicitly stipulates that procuratorial organs must strengthen substantive review of “situation reports,” differentiate among various circumstances to determine appropriate responses, and promptly put forward corrective opinions in accordance with the law. The guiding significance of this case lies in reaffirming that “a ‘situation report’ cannot substitute for statutory evidence,” and clarifying that “when reviewing a case, the procuratorial organ shall examine the content of the ‘situation report’ submitted by the public security organ. If, upon review, it is found that the public security organ has used a ‘situation report’ in lieu of statutory evidence to prove the facts of the case, it shall urge the public security organ to re‑collect and secure such evidence in accordance with prescribed procedures and forms; and if the public security organ has neglected to perform its investigative duties by merely issuing a ‘situation report,’ it shall, in accordance with the law, submit supervisory and corrective opinions.” The rule-of-law spirit underlying this guiding case also holds significant value for standardizing the examination of electronic evidence.


 

Vision No. 3: “Electronic evidence is treated as if it were invisible.” Since its formal recognition in law in 2012, electronic evidence has become a central component of case files across various types of litigation. It typically takes the form of digital devices—such as computers, mobile phones, servers, and optical discs—along with the data they contain, which are handed over by investigative authorities to the procuratorial organs. Nevertheless, both the prosecution and the defense often fail to engage in thorough adversarial scrutiny of these novel forms of evidence, and the courts rarely provide opportunities for such contestation. In some cases, once proceedings reach the procuratorate or the court, when criminal defense lawyers request copies of the data‑bearing optical discs, case handlers frequently resort to evasive delays or outright refusals. In short, in most criminal justice proceedings, the actual substance of electronic evidence remains conspicuously absent from the courtroom.


 

In the first chapter of this book, I cited complaints from criminal defense lawyers that judicial personnel treat electronic evidence as “a malleable doll at their whim.” An undeniable objective reality is that the legal community involved in case handling—prosecutors, lawyers, judges, and others—either fail to “see” or simply do not “look” at electronic evidence. Their attention remains fixed on relevant forensic expert opinions, printed screen captures, and the transcripts derived from them, lingering there. I have attended several criminal trials and, at times, could keenly sense a certain disdain on the part of defendants with technical backgrounds—particularly software programmers: “What exactly are you ‘legal professionals’ talking about? Isn’t discussing the conviction and sentencing in so‑called cybercrime cases just talking past one another?” Even in the adjudication of cybercrime cases, which are widely regarded as most dependent on electronic evidence, such evidence has never been accorded genuine scrutiny—this has become an established fact. This constitutes a paradoxical phenomenon: “On the one hand, cybercrime is increasingly emblematic of the alienation and misuse of advanced technology; on the other hand, the courtroom proceedings in cybercrime cases are virtually indistinguishable from those in traditional cases, devoid of any high‑tech elements or components.”[26]


 

As in the case of KuaiBo Company’s crime of profiting from the dissemination of obscene materials, discussed in Chapters 1, 6, 8, and 11 of this book [27], the prosecution presented a wide array of evidence, including witness testimonies, documentary evidence, work reports from various entities, administrative authorities’ decisions to register and preserve items in advance, lists of items seized during administrative enforcement, records of administrative inspections, interrogation transcripts, lists of evidence obtained, lists of accepted evidentiary materials, as well as numerous expert opinions on the examination and identification of obscene materials and forensic reports on electronic data. The defense offered objections and cross‑examination regarding these pieces of evidence. However, in this case, the crucial evidence—the 21,251 obscene videos and the four servers storing them—was indispensable for proving the defendant’s conduct of knowingly permitting the dissemination of such materials, establishing the resulting harm, and indirectly demonstrating the defendant’s subjective awareness. Regrettably, throughout the proceedings, the defense never examined the “21,251 obscene videos and the four servers,” nor did the prosecution present this evidence in court. Ultimately, the case turned into a “Waterloo” for the prosecution, largely because they failed to conduct a thorough review of these electronic evidences and their associated media. Moreover, the defense’s cross‑examination in this matter left much room for reflection: they neglected to directly scrutinize the electronic evidence and its related media, instead confining themselves to perfunctory challenges to the relevant documentary evidence, interrogation records, and expert opinions. This should have been a landmark case involving electronic evidence, yet, for various reasons, it fell short of fully demonstrating the power of such evidence.


 

Indeed, China has also made efforts to emphasize electronic evidence in the handling of cybercrime cases. In 2020, the Supreme People’s Procuratorate initiated the drafting of the “Rules of the People’s Procuratorates on Handling Cybercrime Cases.” The draft’s Article 3 set forth requirements for the “guidance of evidence collection and the examination and application of electronic data,” stipulating that “the capacity for examining and utilizing electronic data should be enhanced, with a focus on building an evidentiary framework for cybercrime centered on electronic data.” The introduction of the term “core” was particularly striking. However, when the official version of the “Regulations of the People’s Procuratorates on Handling Cybercrime Cases” was issued in early 2021, that provision had completely disappeared. It was not until June of the same year, at a professional symposium I organized, that the principal drafter revealed the behind-the-scenes story: during the deliberations on the draft, there were numerous dissenting views regarding the original wording, and ultimately, the formulation was revised in accordance with the majority decision. The principal drafter also argued that the procuratorial system’s approach to handling cybercrime should pursue innovative development, “with a strong emphasis on constructing a criminal procedural evidentiary system centered on electronic data.” This notion of a “center” likewise conveyed a clear signal of prioritization. Nevertheless, despite this apparent progress in the judicial interpretation, it ultimately failed to break through.


 


 

Vision Four: “The Black Box of Expertise and Auditing.” In modern litigation, an increasing number of specialized facts are emerging—facts that “involve specialized knowledge and technical content, requiring the application of relevant scientific disciplines, skills, experience, or training for proper understanding.”[28] The disputes arising between the prosecution and the defense over such specialized facts constitute what are known as “specialized issues.” These issues are characterized by the following features: “they generally do not constitute legal questions”; “their resolution depends on specialized expertise, including scientific knowledge, professional skills, and field‑specific work experience”; and “they encompass both natural science and the humanities and social sciences, with natural science issues being particularly representative.”[29] With respect to these specialized facts or issues, expert opinions have long been regarded as the “king of evidence,” often decisive in determining the outcome of a case. However, due to institutional design in China, criminal forensic examinations are initiated solely by public authorities, leaving defense counsel and the defendant without the right to initiate such proceedings. Moreover, in judicial practice, experts frequently fail to appear in court to undergo cross‑examination; evidentiary materials submitted for examination are typically withheld from the defense; the methods employed lack mandatory safeguards and are often circumvented; and the examination process itself remains opaque and unclear. These shortcomings inevitably render expert opinions a “black box of evidence,” cloaked in the guise of scientific proof.


 

In the narrow sense, expert opinions in China today refer to the professional judgments rendered by the “four major categories” of forensic examinations. These include forensic medical examinations, physical evidence examinations, audio‑visual material examinations, and environmental damage assessments—each requiring entry‑level registration and subject to unified registration and oversight by the judicial administrative authorities. Electronic‑evidence forensic examinations are, in practice, subsumed under audio‑visual material examinations for registration and management purposes. A key manifestation of the growing “black‑box” nature of electronic‑evidence forensic examinations, beyond the aforementioned factors, is the emergence of distinctive patterns such as the coexistence of multiple expert opinions issued by the same corporate testing institution, and the frequent occurrence of non‑standard examinations conducted by internal units of higher‑level investigative agencies. For instance, in cases like Yu XX’s fraud conviction, one of the principal factual disputes concerns the amount of illicit proceeds. The first‑instance judgment stated that the illicit proceeds totaled RMB 3,456,243,623; however, the case file did not clearly set out the methodology used to calculate this figure. Upon review of the relevant evidence, it was found that a well‑known electronic‑data forensic institution based in Fujian had prepared four concurrent expert reports, each arriving at different conclusions. The samples examined, the methods employed, and the examiners were identical across all reports; only the findings differed. The discrepancies primarily stemmed from variations in the statistical procedures—presumably reflecting the client’s or the institution’s view that earlier methods were inaccurate, prompting repeated supplementary analyses. All four reports were admitted as evidence, yet none excluded irrelevant amounts, such as test sums or routine loan transactions. In situations where multiple expert opinions conflict, the examiners failed to appear in court to clarify their positions. Under such circumstances, the determination of the criminal facts—including the amount involved—is bound to be muddled; who can say which opinion is correct? Or, put another way, perhaps none of them is correct at all. Another example arises in the case of Zhang XX’s crime of operating a casino, where the defendant was accused of knowingly providing fund‑payment and settlement services to a gambling website, thereby facilitating the collection of gambling proceeds, with particularly serious consequences. As to whether the website in question—the “XX Sports” online platform—possessed gambling functions, the determination was made by the Public Security Bureau’s Administration of Public Order Division of XX City, which issued the “Report on the Determination of Gambling Functions of the ‘XX Sports’ Online Platform,” concluding that the platform featured betting and win‑loss mechanisms and lacked any risk‑prevention measures to prohibit gambling activities. Regardless of the fact that this “determination report” lacks the examiner’s signature and detailed methodology, its very compliance—both in terms of the standardization of the examination methods and the procedural rigor of the assessment—inevitably invites scrutiny from the prosecution. Judicial expertise, once hailed as a cornerstone of scientific evidence, has long since been reduced to a mere arbiter of right and wrong.


 

In a broad sense, expert opinions also encompass the professional judgments rendered through “four‑category‑outside” appraisal or examination activities. A common example is accounting and auditing reports—often referred to simply as “expert‑accounting‑as‑evidence.” Such reports are typically prepared by accounting firms commissioned by investigative authorities to conduct accounting and auditing examinations of paper‑based or electronic accounting and audit records belonging to entities implicated in a case. They have become standard evidentiary staples in handling mass‑impact criminal cases; however, they harbor critical shortcomings, including time‑consuming and costly procedures, as well as questionable accuracy. For instance, in Chapters Four and Five of this book, in the case of Wang et al. for the crime of organizing and leading pyramid‑scheme activities [30], an accounting firm conducted statistical analyses based on transaction‑flow records obtained from platforms such as Alipay, concluding that the total amount involved was RMB 178,965,438.00, while the aggregate sums transferred or disbursed as commissions amounted to RMB 95,626,640.19. Both sets of figures were subjected to vigorous challenges during trial, with the main grounds for skepticism centering on the fact that the investigating authorities, the accounting firm, and the Alipay platform—all parties concerned—do not vouch for the authenticity of the transaction records, and that significant doubts remain regarding the provenance and veracity of these records. Despite such objections, these evidentiary challenges are often difficult to sustain before the court, as the certified public accountants and other professionals responsible for the audit typically fail to appear in court to undergo cross‑examination. Strictly speaking, “expert‑accounting‑as‑evidence” or audit reports do not constitute legally prescribed forms of evidence under China’s Criminal Procedure Law. The widespread reliance on such reports in practice may be characterized as a “super black hole” within the judicial proof‑gathering process.


 

Vision Five: “Comprehensive Determination as a Black Hole.” This is a dynamic concept corresponding to the static “evidence black box” of forensic opinions and audit reports. Originally, comprehensive determination was a traditional tool in China’s judicial proof‑making process for assessing the authenticity of evidence, referring to the practice of “reviewing all case‑related evidence in its entirety.”[31] Beginning in 2014, this method was introduced into the realm of proving cybercrime and has been applied primarily to electronic evidence and related corroborating materials. For instance, the 2014 Joint Opinions of the Supreme People’s Court, the Supreme People’s Procuratorate, and the Ministry of Public Security on Several Issues Concerning the Application of Criminal Procedure in Handling Cybercrime Cases stipulated that, in cases involving cybercrimes directed at or organized, instigated, or assisted by an indeterminate number of individuals, where objective constraints prevent the individual collection of relevant testimonial evidence, the method of comprehensive determination may be employed. Subsequently, numerous judicial interpretations and similar normative documents have reiterated the use of this approach for comprehensive assessment and judgment. Drawing on the handling of telecommunications‑based cybercrimes as an example, some scholars have parsed “comprehensive determination” into three distinct modes: first, a plea‑and‑punishment model that takes the suspect’s self‑identification as its foundation and supplements it with corroborating evidence; second, a sampling‑based evidentiary approach followed by holistic case‑wide comprehensive determination; and third, a limited punitive model that, constrained by traditional notions of criminal proof, seeks during investigation to establish the lower threshold for the amount involved in the relevant offense under the applicable criminal statute, thereby basing conviction and sentencing on that threshold.[32]


 

In practice, the comprehensive assessment of electronic evidence has become increasingly widespread. Typically, prosecutors appearing in court no longer diligently present electronic evidence as they once did—often omitting it altogether—and they rarely address the defense’s challenges to such evidence. Instead, they simply leave the matter to the court for what is euphemistically called a “comprehensive determination.” Some judges seek verification from public security authorities after the hearing; others calculate the amount of crime themselves based on backend data; and still others rely outright on verbal testimony or explanatory statements. In short, as long as a few pieces of corroborating material can be assembled, judges are all too willing to render verdicts on the basis of subjective judgment. This has led to a serious distortion in the application of electronic evidence.


 

For example, in Case No. [33] of Chapter 7 of this book—Dong et al. for the crime of infringing upon citizens’ personal information—the primary evidence relied upon for prosecution consisted of electronic data, records of site inspections and examinations, and expert opinions. The court’s judgment explicitly excluded the evidentiary value of the inspection and examination records, on the grounds that “the public security authorities failed to verify the authenticity and originality of the server logs provided by Ziroom, nor did they offer a reasonable explanation.” It likewise excluded the evidentiary value of the server‑log expert opinion, reasoning that “the legality of the data obtained during the inspection and examination could not be confirmed, and neither the authenticity of the logs themselves nor the relationships among different logs could be ascertained or analyzed.” In fact, the court also rejected the evidentiary value of the server logs and other electronic data submitted by Ziroom, as their authenticity and originality remained unverified. Nevertheless, the court’s “findings” section stated: “The evidence presented by the victimized entity, together with the defendant’s statements and the contents of the mobile‑phone forensic report, corroborate the methods and means employed in this case to infringe upon citizens’ personal information, the sensitivity of the specific personal information obtained, and the sheer volume of such information acquired. In light of the specific criteria for conviction and sentencing, this court finds that the circumstances of this case constitute an exceptionally serious instance of infringement upon citizens’ personal information.” This clearly amounts to establishing the facts of the case—and thereby creating a baffling “black hole of findings”—on the basis of verbal evidence that is imprecise, subjective, internally contradictory, and poorly reliable.


 

Vision Six: “The supplementary case file is used as a shield.” In China, criminal case files maintained by judicial authorities are divided into primary and supplementary (internal) files. The supplementary file primarily contains internal submission documents and confidential materials that are not intended for public disclosure. For instance, the main contents of the supplementary file kept by the court include: (1) the case file cover; (2) the table of contents within the file; (3) records of case review; (4) the reviewing report prepared by the case handler; (5) materials or minutes documenting internal exchanges of views between the handler and relevant departments; (6) internal requests for instructions and corresponding replies pertaining to the case; (7) minutes of collegial panel deliberations on the case; (8) records of the trial bench’s study and reporting on the case; (9) minutes of discussions held by the adjudication committee; (10) original and duplicate copies of the comprehensive case report; (11) the original judgment or ruling; (12) the trial supervision form or the opinion letter remanding the case for retrial; (13) other materials deemed unsuitable for public release; (14) the preparatory notes; and (15) the bottom of the file. [34] Generally, the supplementary file is not transferred along with the case but is instead kept by the archives departments of the public security, procuratorial, and judicial organs. In practice, however, there have been instances where supplementary files were nevertheless transferred and inadvertently made public.


 

In practice, there have been instances where electronic evidence is filed in the auxiliary case file, thereby undermining the very purposes of presenting and challenging such evidence. For example, in Chapter 13 of this book, in the case involving Luo XX’s crimes of bribery of an influential person and misappropriation of funds [35], the key incriminating evidence consisted of two image files—“IMG-2493.JPG” and “IMG-2494.JPG”—extracted by investigators, which were documented in the “Electronic Data Examination Record ‘× Public (Economic) Inspection [2019] No. 1’.” Upon verification, it was found that the extraction process for these two image files did not comply with the standards set forth in the Ministry of Public Security’s “Rules on Electronic Data Collection in Criminal Cases by Public Security Organs (2019),” nor did the images originate from the original storage medium or a forensic image; instead, they were derived from a compressed file named “×× Apple Phone_html.zip.” In this case, the simplest way to ascertain the truth would have been to directly examine the compressed file “×× Apple Phone_html.zip.” However, the response provided by the investigating authorities was merely a “Statement of Circumstances,” indicating that the “×× Apple Phone_html.zip” had already been placed in the auxiliary case file. The statement specifically read: “The mobile phone submitted by Du XX to the public security organs on September 14, 2018, has had its electronic data extracted, resulting in the evidence‑gathering report ‘×× Apple Phone_html.’ This evidence‑gathering report (electronic data) has been filed in our bureau’s investigative case file for record‑keeping purposes; the evidence‑gathering report on Du XX’s mobile phone, ‘×× Apple Phone_html,’ has likewise been filed in our bureau’s investigative case file for record‑keeping purposes; and the record of the electronic data extraction from Du XX’s mobile phone has also been filed in our bureau’s investigative case file for record‑keeping purposes.” This practice of filing evidence in the “auxiliary case file” effectively denies the defense counsel the opportunity to verify and cross‑examine the evidence, thus obstructing the path to uncovering the truth. It is hardly surprising that some scholars have criticized the existence of the auxiliary case‑file system in people’s courts as constituting a “malady” of the judiciary, one that “has had a negative impact on the progress of judicial reform” [36].


 

Vision Seven: “Endless Supplementary Investigations.” Article 274, Paragraph 1 of the Supreme People’s Court’s Interpretations on Several Issues Concerning the Application of the Criminal Procedure Law of the People’s Republic of China provides: “During the trial, if the prosecutor finds that the case requires supplementary investigation and proposes a postponement of the proceedings, the collegial panel may grant such a request; however, the number of proposed postponements shall not exceed two.” This provision reflects a spirit of compromise. Nevertheless, in practice, an endless pattern often emerges—conducting further investigations only after the two rounds of supplementary inquiries have concluded. Such a practice entails grave adverse consequences, including “violating the principle of adversarial balance,” “undermining the protection of the defendant’s human rights,” and “resulting in the waste of judicial resources and a decline in litigation efficiency.” With respect to the examination of electronic evidence, this approach can also undermine the probative value of the electronic evidence itself and its related corroborating evidence.


 

As in the aforementioned case involving Wang et al. for the crime of organizing and leading pyramid-scheme activities, the procuratorial organ conducted two rounds of supplementary investigations; however, all electronic evidence and related supporting materials obtained were subsequently rejected. Thereafter, the investigative authorities undertook another round of supplementary investigation, exceeding the statutory time limit to obtain purported financial transaction records compiled by third-party institutions such as Alipay, and commissioned an accounting firm to prepare a special audit report. This “Special Audit Report,” along with its accompanying evidentiary materials, was deemed critical to determining guilt and sentencing; yet neither the public security organs, nor the procuratorial organs, nor the trial court complied with the provisions of the Criminal Procedure Law by reading the report to the defendant or soliciting his views on whether to request a re‑examination or additional expert appraisal. In response, the defense counsel raised objections, arguing that the formation of this evidence occurred after the court had returned the case for supplementary investigation. Nevertheless, the court declined to accept this defense argument.[37]


 

In response to this phenomenon, some Chinese scholars have called for the “prompt abolition of the supplementary investigation system during the trial phase.” They argue that both the prosecution and the defense should be granted equal rights to request evidence, with out-of-court investigations conducted by the people’s courts, focusing primarily on evidence favorable to the defendant. As for witnesses who retract their testimony during the trial, they should not be unilaterally questioned and verified by the investigative (investigation) authorities; instead, they ought to be summoned to appear in court to testify.[38]


 

II. Embracing the “Massive Evidence” That Has Already Arrived


 

The essence of the “grand evidence‑based thinking” lies in squarely confronting the “datafication” and “massification” of evidence—namely, the sheer volume of evidence.


 

Since data have come to be measured in gigabytes, terabytes, and petabytes, electronic evidence in individual cases has increasingly taken the form of an inexhaustible reservoir of digital material. Such evidence typically manifests as vast volumes of office document files, photographic evidence, emails, communication and chat logs, telecommunications data, financial transaction records, web‑page data, public‑opinion data, logistics data, location‑tracking data, and log files. In cases I have personally handled, some involve over 6 billion financial transaction records; in others, a single smartphone’s WeChat chat history exceeds two million entries; in still others, computer log data number in the hundreds of millions; and in yet other cases, the sheer volume of video files surpasses ten million. This diverse array of massive data has long been embedded in the investigative process—this is an undeniable reality. Accordingly, investigative authorities have already adopted innovative practices such as Blu-ray disc burning and the use of Blu-ray discs, as well as blockchain‑based evidence preservation and big‑data analytics platforms for evidence examination.


 

Meanwhile, investigative authorities still widely rely on the traditional practice of circulating paper-based case files. My research reveals that a single case can generate thousands of volumes, with records routinely surpassing 2,000, 5,000, 8,000, or even 12,000 pages. Yet the process of locating, collecting, analyzing, and integrating evidentiary information within these paper files is exceedingly slow and inefficient, consuming substantial judicial resources at every stage—preliminary evidence‑gathering, mid‑process evidence duplication, and later-stage evidence utilization. In one case involving the crime of illegally absorbing public deposits, in which I served as counsel, although the case file comprised only about 200 volumes, an electronic version was not prepared for my review until near the end of the procuratorate’s examination‑and‑prosecution phase. Upon inquiry, the primary reason was that the procuratorate’s case‑management department prioritized digitizing paper files related to cases under review for arrest. This mechanism, which relies almost exclusively on paper-based materials, warrants serious reconsideration. Given that one of the key pieces of evidence in this case was a “Special Audit Report,” whose conclusions raised significant doubts, I repeatedly requested that the procuratorate prioritize providing both the report itself and the underlying raw data, so that a thorough re‑examination could be conducted. Ultimately, I managed to obtain a PDF copy of the Special Audit Report, but the provision of the original data remains indefinitely delayed. This situation underscores investigators’ entrenched habit of working with paper documents, reflecting both an unconscious reluctance and an inability to embrace the vast quantities of digital data now available.


 

From the perspective of electronic evidence, human justice has entered an era of massive volumes of evidence. So how can case-handlers effectively leverage such vast amounts of evidence in their work? In short, they must confront this deluge of data head-on and, by harnessing algorithmic and data‑driven models, adopt a “machine‑plus‑human” two‑step approach to transform both their mindset and skill set. Chapter 4 of this book, “Intelligent Evidence Review,” offers a focused discussion on this topic. Here, we supplement that analysis by examining four cutting‑edge explorations currently underway in practical application.


 

One form of investigation is “financial analysis as evidence.” This refers to the systematic sorting and analysis, conducted by investigative and supervisory authorities in accordance with the law, of financial data related to a case. Such analysis examines the source and destination accounts involved, the amounts transacted, the timing of transactions, and the correlations and commonalities among counterparties, thereby establishing relationships between accounts and tracing the flow of funds, and ultimately producing a specialized report that supports fact-finding and legal assessment. In contrast, “forensic accounting as evidence” encompasses both: (1) specialized expert opinions issued by forensic‑data appraisal institutions commissioned by the investigating agency to examine backend server data of the entities implicated; and (2) specialized analytical reports prepared by accounting firms engaged to conduct accounting audits—based on paper or electronic accounting and auditing records—of those same entities. Comparing the two, “financial analysis as evidence” and “forensic accounting as evidence” exhibit an inherent continuity, yet the former clearly enjoys certain advantages. First, “financial analysis as evidence” pertains to the actual monetary sums involved in the case, whereas “forensic accounting as evidence” reflects the book‑keeping figures recorded in official ledgers. Put simply, the former corresponds to real‑world amounts, while the latter relates to nominal, ledger‑based figures. Second, “financial analysis as evidence” relies on persistent electronic data, whereas “forensic accounting as evidence” often depends on electronic records that are prone to loss or alteration. Consequently, the former can generate evidence that endures indefinitely, while the latter may become difficult to verify over time. Third, the findings of “financial analysis as evidence” are readily subject to judicial scrutiny, whereas the results of “forensic accounting as evidence” tend to be more challenging to authenticate in court. This highlights a critical issue in the era of big data: whether expert judgments grounded in such evidence can withstand judicial review. It is thus evident that, at present, investigative authorities should not shy away from—or dismiss—“financial analysis as evidence.”


 

This necessitates that investigating authorities actively expand access to financial data originating from commercial banks, foreign exchange regulatory agencies, and third-party payment institutions, and develop and deploy sophisticated data‑analytics models capable of tracing the origins of funds, tracking their destinations, and clarifying transaction structures. On the basis of machine‑generated, rapid insights into the fund‑flow patterns associated with designated accounts and their counterparty accounts within a specified scope, human reviewers should conduct manual verification and issue professional reports. For instance, in the aforementioned case involving Wang XX and others for the crime of organizing and leading pyramid‑scheme activities, investigators must discard substandard “forensic‑accounting‑based evidence” masquerading as “Jinxi Weizheng,” retrieve compliant financial data for cleansing, import it into the Jinxi Weizheng platform for expert examination and analysis, and thereby arrive at findings on illicit proceeds that stand up to historical scrutiny.


 

The second line of inquiry is corroborated by digital traces. Behind big data, trajectory analysis typically involves conducting in-depth scrutiny of an individual’s location or movement data to reconstruct their activities at a specific moment or over a given time period, thereby identifying their whereabouts or travel patterns. Historically employed as a lead in criminal investigations, it now faces the challenge of being further formalized as admissible evidence.


 

The objects of analysis under the “trace‑as‑evidence” paradigm are various types of location data or trajectory data. Location data primarily refer to geolocation information derived from positioning technologies or base‑station data; trajectory data, by contrast, consist of time‑ordered sequences of location records over a given period and encompass such forms as vehicle/ride trajectories, flight paths, accommodation histories, mobile‑phone movement patterns, and online‑behavior footprints, among others.[39] The analytical approaches employed in this context may range from straightforward big‑data‑driven methods—such as investigators reconstructing discrete waypoints by parsing location metadata embedded in relevant electronic evidence—to more specialized analyses conducted on dedicated big‑data platforms—for example, tracing trajectories through the examination of RFID‑based movement records, public‑security‑agency RFID big‑data systems, RFID‑big‑data application registration logs, big‑data trajectory query reports, and vehicle‑trajectory information aggregated via intelligent transportation‑integration platforms. Adopting trace‑as‑evidence as a standard practice should become a new skill for investigators seeking to manage and interpret vast volumes of digital evidence.


 

For example, in Chapter 10 of this book, the case involving the crime of fraud committed through the fictitious financial‑management websites “Zao Gu Fang” and “You Cai Bao”[40], the suspect operated from abroad but was apprehended within China. By examining the locations of key base stations associated with the mobile phone numbers used by the suspect in China, investigators were able to identify each suspect and establish a chain of evidence tracing their movements prior to出境. Similarly, in Chapter 10, the case of Li ×× for the crime of fraud[41], to determine whether the defendant had impersonated a particular virtual identity to obtain property, investigators could consult the WeChat account login logs for Li ×× obtained from Tencent. These logs would reveal whether the trajectory indicated during the period of the offense corresponded to the defendant’s actual location, and whether the post‑offense login activity conflicted with the fact that the defendant was in custody.


 

The third dimension of exploration is relational evidence. Big data can capture a wide array of social relationships. Based on the collection and analysis of social media data, investigators can employ social network analysis to map out the target’s social network or structural diagram, thereby uncovering leads about the actors involved in criminal activities. Our research indicates that disciplinary inspection and supervision authorities frequently rely on methods such as profiling specific associates to break deadlocks in anti-corruption cases, while public security organs often use profiling of shared relationships to narrow down suspects in joint‑crime investigations. This approach also calls for specialized algorithmic models, including time‑specific profiling, anomaly‑duration profiling, sensitive‑time profiling, and companion‑relationship profiling. In theory, relational profiling can generate new forms of evidence; however, this remains to be validated through practical application and further refined through accumulated experience.


 

As illustrated in Chapter 10 and Chapter 11 of this book, in the case of Cui × for fraud [42], the court of first instance, relying on verbal evidence such as confessions, found that Cui ×, the chairman of the platform company involved, was the principal offender who organized the female members of the platform to commit fraud. By the time the case reached the second-instance review, the defense counsel, by analyzing the communication records between the defendant Cui ×, as well as Fang × and Tian ×× (both senior executives of the company), and various female livestreamers, uncovered that Fang × and Tian ×× had secretly established another company to profit from organizing the female members. Consequently, it was no longer appropriate to extend liability for complicity to the chairman, Cui ×. The second-instance court accordingly revised the verdict, finding Cui × to be an accomplice in the scheme to defraud the platform’s female members. The successful outcome of the defense in this case underscores the necessity of turning “connections as evidence” into a practical reality.


 

The fourth aspect of exploration is that patterns serve as evidence. Once individual data points are aggregated to a sufficient scale, the underlying regularities they reveal can be used to reconstruct the facts of a case. Central to this process is the systematic understanding distilled from vast amounts of data—manifested in specialized analytical reports. This is precisely what distinguishes and endows big data with its unique appeal as evidence. However, whether such patterns can indeed function as proof remains an unresolved theoretical question, one that has become particularly salient in the era of big data.


 

For instance, in a tax‑related case, investigators obtained the complete set of financial data—comprising over one hundred million transaction records and tens of thousands of accounts—and engaged technical experts to “develop a funds‑characteristics analysis model… which, in a very short time, enabled the profiling of suspicious fund networks. Using relationship‑visualization techniques, they clearly mapped the origins and destinations of these funds and automatically assigned category labels to both account numbers and entities.”[43] Such analytical reports resemble the opinions of forensic experts or other specialists, yet they differ markedly from them; their level of specialization has already surpassed that of human experts. More precisely, they are judgments generated through machine‑learning algorithms.


 

III. Examining the Endogenous “Evidence Structure”


 

The essence of the “grand evidence‑based thinking” lies in three key points, the first of which is grounding itself in the “structuring” of evidence—namely, placing greater emphasis on the endogenous “evidence structure.”


 

The renowned American scholar of evidence, McCormick, famously observed: “A single brick does not make a wall.”[44] To elaborate, whether a particular piece of evidence or a set of evidence is relevant to the case is one question, whereas whether the totality of the evidence presented by a party has been adequately laid before the jury is another. The standards for adjudicating these two questions differ, with the former being less stringent than the latter. He thus highlighted the distinction between viewing evidence in isolation and considering it within a systematic framework—this represents structural thinking in the field of evidence law.


 

So‑called structural thinking is a mode and method of reasoning that, from the perspectives of systems, structure, elements, and hierarchy, treats complex and multifaceted phenomena as an integrated whole. It emphasizes understanding objective reality through the lens of systemic structure and seeks to identify the optimal configuration in order to achieve the highest possible system performance. [45] This approach constitutes a scientific way of analyzing matters from a structural standpoint and holds a deeply entrenched position within the field of judicial proof. When handling cases, practitioners employing structural thinking primarily aim to construct various types of evidence combinations, evidence systems, or evidence structures. These three terms are largely synonymous, all reflecting the principle that “a single piece of evidence is insufficient to establish a case.” To illustrate with the example of an evidence combination: it is “not merely a simple juxtaposition or accumulation, but rather the process of assembling parts into a whole, integrating elements into a system, and organizing individuals into an entity,” encompassing both “the organization of components into a unified whole and the integration of their internal relationships.” [46] Traditionally, such evidence combinations have been constructed by investigators through logical argumentation. Generally speaking, when multiple pieces of evidence—such as testimonial evidence, physical evidence, documentary evidence, and hybrid electronic evidence—pertain to the same factual scenario, and when these evidences point in the same direction, corroborate one another, and exhibit no irreconcilable contradictions, they may be assembled into an evidence combination. Overall, this represents an external structural arrangement.


 

With the emergence of electronic evidence, the structural framework of judicial proof has been enriched, giving rise to a wide array of novel evidence combinations, evidentiary systems, and evidentiary structures. These developments represent both opportunities for theoretical innovation and avenues for practical breakthroughs.


 

One of the new evidence‑combining frameworks is the triadic structure of electronic evidence—“content–attributes–traces.” Electronic evidence exists within the information space. No individual can directly access the information space to generate evidence; rather, such evidence is produced by machines composed of software, hardware, and other components. This gives rise to the intrinsic architecture of electronic evidence: “content–attributes–traces.” Here, “content” refers to those portions of electronic data that can be readily displayed on a computer screen or printed as documents—for example, the middle section of a Word document when viewed in hexadecimal format. “Attributes” encompass the standard metadata—such as creation date, duration, author, software name, version number, and file size—that are typically presented through document properties; they may also include type‑specific attributes, such as the aperture value, exposure time, ISO speed, focal length, maximum aperture, and metering mode embedded in JPEG files via EXIF data, which are usually manifested in the header and footer sections when displayed in hexadecimal. “Traces,” meanwhile, comprise both the various trace files generated by information systems—such as Windows log files—and the trace characteristics that electronic evidence leaves at different levels of the information system, including printed outputs and online activity records. In cases where electronic evidence serves as the basis for a verdict, it must conform to this triadic structure to effectively fulfill its probative function.


 

For example, in the case of Li’s crime of infringing upon trade secrets, as discussed in Chapters 1, 3, 6, and 13 of this book [47], one of the central points of contention was whether the disputed file, ××.xls, had been fabricated and subsequently transplanted into the computer after the fact. In such circumstances, to render a professional judgment, investigators may examine whether the content of the ××.xls file—including sensitive “trade secret” information such as a list of hundreds of thousands of customers—was obtained or controlled by the defendant, and whether it was known to the victim or other parties. However, this alone is insufficient; investigators must also clarify the file’s metadata, including standard “property information” fields such as “creation time,” “last modified time,” and “last accessed time,” as well as “extended information” items like “time of creation,” “date of last save,” “time of last print,” and “total editing time.” Furthermore, they need to scrutinize the file’s forensic traces, encompassing system files, system logs, user history records, temporary log files, chat logs, session logs, application logs, user authentication logs, database logs, cookies, and more. In this case, the most pivotal electronic evidence was a log trace labeled $J, which clearly demonstrated that the ××.xls file had been implanted after someone adjusted the computer’s clock to August 6, 2015, on or after January 30, 2019. Nevertheless, it is readily apparent that what ultimately served as the evidentiary foundation in this case was the comprehensive framework comprising the file’s content, metadata, and forensic traces.


 

The second new evidence combination is the “Appraisal—Data (Media)—Collection” trinity. Electronic evidence must be converted with technical means before it can be read, and it increasingly appears in massive volumes, so investigators usually cannot directly inspect the electronic evidence or its media to determine the facts of a case. In practice, electronic evidence (abbreviated as “data (media)”) needs to be transformed into an expert appraisal opinion (abbreviated as “appraisal”), for example by requesting an appraisal to “locate electronic evidence related to the case” and having experts produce an appraisal report. This forms an initial “Appraisal—Data (Media)” framework. At the same time, to show how the electronic evidence used as appraisal material was obtained, investigators must produce legal documents such as on-site inspection records, remote inspection records, and network online extraction records (abbreviated as “collection”), corroborated by written records + audio/video recordings + attachment CDs, etc. This forms an initial “Data (Media)—Collection” framework. Combined, they create the complete “Appraisal—Data (Media)—Collection” trinity. It is a coherent, endogenous combination of evidence.


 

For example, in the aforementioned case against ×× Company and Yang×× for providing programs and tools for intrusion and illegal control of computer information systems, the key evidence proving the main fact—that the software developed and provided by the defendant constituted programs/tools for intrusion and illegal control—was an expert appraisal report whose opinion stated that “the submitted program has intrusion functions and is an intrusion program.” How should one examine whether that appraisal opinion is valid? Investigators cannot judge an appraisal by the appraisal alone; they must clarify the appraisal materials used. That led to finding a source code file from suspect Zhou×Jie's computer hard drive, files from a witness’s feature phone, files produced by investigative experiments conducted by the authorities, and a special compressed document provided by the victim company. These various electronic files were burned onto different discs, and to determine how these files and discs were obtained, investigators also had to produce an electronic evidence inventory, investigative experiment records, and so on. After such sorting, the case quickly reached the basic conclusion that the appraisal opinion was unreliable. There was no special clairvoyance involved; the key was thinking in terms of the “Appraisal—Data (Media)—Collection” trinity.


 

The third new evidence combination is system review of multi-version and continuum evidence. As electronic evidence shifts toward massive scale, multi-version and continuum phenomena follow. Multi-version electronic evidence means that electronic evidence and its related evidence appear in multiple versions, such as original files and image files, original data and backup data, primary data and secondary data, or successive copies of inspection records and forensic appraisal reports. The continuum of electronic evidence means that electronic evidence and related evidence appear in continuous forms, such as a WORD document that during creation produces a single file, a compressed archive, sent email, forwarded email, and archived file, or a series of documents like inspection records, seizure inventories, appraisal commission agreements, appraisal opinions, and appraisal notices. These constitute the horizontal and vertical structures in the lifecycle of electronic evidence, providing a natural foundation for comparative review of electronic evidence.


 

For example, in the aforementioned case against Dong×× and others for infringing citizens’ personal information, the investigative agency extracted two database files provided by the victim company on two occasions and submitted them to the same appraisal institution for four appraisals. This created multi-version and continuum electronic evidence. The former includes on-site inspection records 公(网安)勘【2019】040101号, 【2019】050601号, and the Beijing ×× forensic audiovisual opinion documents 京××【2019】声像鉴字第339号、第360号、第376号(旧)、第376号(新)《司法鉴定意见书》; the latter mainly includes work notes produced by the investigative agency, data materials provided by the parties, inspection records, forensic appraisal opinions, and interrogation records. By comparing according to the two inspection and four appraisal time points, I found falsehoods and illegalities in the on-site inspection and appraisal processes.


 

The fourth new evidence combination is clusters of linked evidence across different spaces and different nodes. Electronic evidence is primarily evidence in the information space, but it also involves physical space in the course of the crime and the investigation. This produces combinations of electronic information-space evidence and related physical-space evidence, analogous to different parts of a human body forming a connected relationship. Generally, if they are consistent in content and are located in separate spaces, they constitute a cluster of linked evidence located in different spaces. Electronic evidence also commonly coexists across multiple information systems and multiple network nodes, more closely resembling different body parts forming a linked relationship. Likewise, if an operating system and an application file system’s data series were produced by the same machine behavior, or if data series on different network systems were produced by the same network interaction, they constitute a cluster of linked evidence located at different nodes.


 

For example, in the case of Wu×× organizing and leading pyramid selling activities described in Chapters 8 and 9 of this book [48], the crucial evidence was claimed to be data extracted remotely from a U.S. server by an appraiser from the ×× Electronic Data Forensic Center; the appraisal report indicated the appraiser started the electronic evidence workstation, simultaneously started video recording, opened the server URL involved in the case, entered account credentials, and downloaded data. This created a linkage among the data download action, the data fixation action, and the video recording action; the downloaded data and the video file produced by that appraisal formed a linked evidence cluster. Through linked-space review across the two spaces it was relatively easy to discover various doubts about the false appraisal. Similarly, in the Chapter 9 dispute between Company A and Companies B and C over sales [49], the important evidence was an email allegedly from an employee of Company C guaranteeing something, and the email’s authenticity was hotly disputed in court. Abnormalities included possible double-impressioning of the stamp text, incomplete EXIF information in attached images, and possible replacement during attachment forwarding. In such cases, investigators should also search the other endpoint computer used to send/receive the email, the two endpoint servers, and any computers that may have forwarded the email, locate the corresponding versions of the email to form a linkage, and perform a complete review of the sending process.


 

The fifth new evidence combination is evidence chains from different angles. The evidence chain is the basic theory for deciding a case based on indirect evidence. Article 140 of the Supreme People’s Court’s Interpretation on the Application of the Criminal Procedure Law of the People’s Republic of China provides that indirect evidence can only be used to convict if “the evidence in the whole case forms a complete chain of evidence.” The term “evidence chain” is a vivid way to describe a closed loop built from different pieces of indirect evidence. Although electronic evidence is not entirely indirect evidence, it has a natural advantage in building evidence chains. Chapter 10 of this book explains that electronic evidence has unique advantages for constructing evidence chains. This is particularly evident in its ability to naturally or deliberately form multidimensional evidence chains, thereby promoting specialized, chain-style high-level review techniques.


 

Elementary techniques include building chains of evidence based on different stages and the same subject. For example, in Chapter 10 of this book, the fraud case against Guo ×× and 28 others [50] occurred abroad. Investigators both followed the main steps of the fraud—“computer push of voice packs,” “first line impersonating Hong Kong Immigration officers,” “second line impersonating public security officers,” “third line impersonating prosecutors”—to sort out, step by step, items such as the “voice packs,” “call logs,” “household registration information, arrest warrants, indictments, website login traces,” and the “fake Supreme People’s Procuratorate website, Skype chat records,” and also assembled electronic evidence chains for different victims by linking victims’ identity information, transfer records, victim statements, and chat records. Advanced techniques include building evidence chains based on means extraction, relationship characterization, identity identification, trajectory determination, and spatiotemporal bounding. For example, in the cases of Cui × (fraud), Liao ×× et al. (fraud), and Li ×× (fraud), investigators built “behavior/means chains” via keyword searches, constructed “relationship chains” through communications data analysis, established “trajectory chains” by querying mobile service records, and built “time-space relationships” from mobile login logs… The complexity and diversity of these evidence chains align with the intrinsic chain-forming characteristics of electronic evidence.


 

All of these endogenous evidence structures, in fact, form a highly imaginative, large-scale system of electronic evidence. This far exceeds traditional evidence systems. When investigators conduct electronic evidence review, they should be guided in this direction.


 

4. Expanding human-like "proof models"


 

The fourth characteristic of “big evidence thinking” is analyzing evidence with a human-like cognitive model, that is, a human-like “proof model.”


 

How do investigators understand a case? Generally speaking, this group has a distinctive way of thinking. Conan Doyle, through Sherlock Holmes in The Adventures of Sherlock Holmes, wrote: “To a logician, life is like a vast chain: on seeing one link, one can infer the condition of the whole chain.” [51] This evaluates investigators’ intellectual appeal from the perspective of formal logic. A more widespread metaphor is the “magic crystal ball.” La Belkin said: “An investigator’s ‘magic crystal ball’ … has many facets; through them the investigator can see the ‘misty distance’ of the case. One facet is the technical arsenal, another is logic, and a third is the various principles of psychology, especially criminal psychology.” [52] This indicates that investigators must blend technical knowledge, logical reasoning, and psychological understanding in their thinking. People often say investigators are eclectic knowledge carriers, which highlights the importance of knowledge integration.


 

In the age of massive data, how can investigators use machine intelligence to confront huge volumes of information? This requires mechanizing and intelligently encoding human investigative experience—teaching machines to think like investigators. The trick here is to expand human-like “proof models.” A proof model here can be understood as an abstraction of investigators’ judicial proof methods into a framework. A large part of it can be digitized and implemented intelligently. It is a key to investigative automation. One could say that as various investigative knowledges become modeled and even intelligent, the new-quality productive power generated by assisted investigative tools seems to emerge continuously from beneath the surface. This process is happening and will continue to occur.


 

One major category of human-like “proof models” is the data collision model. Data collision refers to selecting one or several data items from different databases as identifying data and comparing them to discover supervisory clues present across batches of cases. A data collision model automatically compares and analyzes two or more datasets to find similarities and differences, thereby proving case facts. It is the large-scale, mechanized version of investigators’ comparison and review. Common software today: Word can automatically compare differences between two documents, Excel can compare and reconcile data across different columns or sheets, and some professional data-collision software or platforms can achieve horizontal linking and vertical filtering across domains. This lays the foundation for efficiently selecting and applying data collision models during evidence review in individual cases.


 

For example, in the aforementioned Wu ×× case on organizing and leading pyramid selling, the most important inculpatory evidence was an appraisal stating that the disputed platform had 62,818 members and formed 46 promotion tiers. To verify whether this professional judgment met the legal threshold for conviction (“30 people, three levels”), the defense team performed data collision analysis on members’ login IP addresses in the involved database (including users’ registration IPs and last-login IPs). They concluded that there were only 116 distinct last-login IPs and only 60 IPs for users who placed orders, and they performed multidimensional deduplication using ID numbers. On this basis they further showed that the appraisal’s claim of “46 promotion tiers” was not equivalent to the rebate/return tiers. The data-collision analysis method used here is abstractly a data collision model.


 

The second major category of human-like “proof models” is the data profiling model. Data profiling means identifying regular patterns from individual cases, inductively analyzing anomalous features and elements of cases, and constructing a profile that can be supervised. A data profiling model automatically analyzes multiple datasets across multiple dimensions to build a characteristic profile of an actor or behavior, thereby proving case facts. Data profiling is widely used in commercial marketing—dynamic pricing for frequent customers is a typical example. Its use in judicial investigations is generally thought to have originated in profiling corrupt officials, typically involving characterization of identity information, family background, work history, social relationships, activity trajectories, asset conditions, and other anomalies. Today data profiling has spread to many scenes and corners of judicial work. In short, whenever it is necessary to understand a party or other participant’s relationship to the case facts, data profiling is useful.


 

For instance, in Chapter 11 of this book, the Chen × gambling case [53], the key dispute was whether local authorities had grabbed jurisdiction to conduct “offshore fishing.” The defense needed to depict how local members had registered on and used the platform. This can be filtered by local IP addresses and phone numbers, and matched members can be profiled using platform records. The result showed that dozens of police officers, including the police who accepted the reports, were registered members of the alleged gambling platform and had used the platform; and two complainants’ reported information did not match their platform records but did match the platform records of the reporting police… This discovery can be understood as a data profiling of local members, abstracted into a data profiling model.


 

The third major category of human-like “proof models” is the data mining model. Data mining refers to using algorithms to extract a few direct crime-indicative data points from massive datasets under certain conditions. A data mining model automatically performs microscopic analysis of a dataset to uncover one or more concealed pieces of evidence that point to criminality, thereby proving case facts. Big data is like an ocean and mining is like panning for treasure. This requires transferring investigators’ discerning insight to data models. Generally, investigators identify anomalies based on specific evidence features they discover, and data models need to incorporate those experiences to optimize analytical logic. The most thorough data mining is often considered to be conducted at the hexadecimal level, searching for crime-related, byte-level “trace evidence.”


 

For example, in Chapter 6 of this book, the case of a DDoS‑attack‑related crime [54], the suspect performed data processing on the hard drive (virtual machine) involved. To uncover traces left by the suspect’s DDoS attack, investigators conducted low‑level hexadecimal‑based data mining on the suspect’s virtual‑machine data, identifying evidence of data exchanges with the victim organization’s servers, traces of VPN usage, and fragments of data used to extort the victim. The key to successfully solving this case lay in searching for specific keywords at the hexadecimal level, a technique that can be further developed into a series of data‑mining models.


 

The fourth major category of human-like “proof models” is the data‑penetration class of models. Data penetration refers to looking beyond the apparent legality of a case to conduct an in-depth examination of the underlying anomalies concealed within related matters. A data-penetration model performs automated, granular analysis on an unspecified dataset, uncovering the surface-level facts of a case and elucidating the intrinsic relationships—such as financial flows, equity structures, social-media connections, logistics patterns, locational trends, and communication dynamics—to substantiate the factual basis of the case. Compared with the aforementioned types of data models, data-penetration models place even greater emphasis on large‑model thinking.


 

For example, in the case of Feng et al. for the crime of illegally absorbing public deposits, the audit report concluded that the platform involved had illegally absorbed 14.5 billion yuan in public deposits, including 2.21 billion yuan attributable to duplicate investments. To verify the accuracy of this conclusion, the defense counsel team requested a copy of the source data used by the accounting firm in preparing the audit report. The review revealed that the firm had based its assessment solely on the investment‑submission forms embedded in the platform’s backend data, lacking both a proper extraction record and the essential steps required to authenticate and validate the information. A particularly glaring error was that the accounting firm had double‑counted massive investments totaling over several billion yuan and omitted more than one billion yuan in funds that had already been repaid. On this basis, the defense team sought expert assistance to scrutinize the backend data and, relying on objectively verifiable cash‑flow records, arrive at an accurate determination of the true amount of illegally raised funds.


 

The introduction of the aforementioned data models into judicial proof has imposed a contemporary imperative on the legal community to innovate the principles of evidence law. Innovative tasks include how to recognize new forms of evidence, define the positioning of proof models, develop methods for presenting and challenging evidence, and formulate a comprehensive theory of fact-finding supported by data‑driven models.


 

V. Pursuing Advanced “Smart Case Handling”


 

The essence of the “grand evidence‑based thinking” lies in five key aspects, one of which is moving toward “intelligent” evidence review.


 

The renowned Canadian hockey player Wayne Gretzky once said, “Skate to where the puck is going, not where it has been.” This has since become a celebrated maxim for reflecting on how big‑data‑driven intelligence is reshaping society. So what, then, is the “puck’s destination” in the examination of electronic evidence? It is nothing less than sophisticated, AI‑powered case management—underpinned by an endogenous “evidence architecture” and a human‑like “proof‑validation model.”


 

This shift in the review approach not only frees investigators’ hands and feet, but also liberates their minds. What emerges is not manual case handling, but sophisticated intelligent case processing—no longer entirely reliant on human judgment, but infused with the hallmarks of artificial intelligence. Admittedly, fully replacing judges, prosecutors, police officers, lawyers, or forensic experts with machines remains a distant prospect (though it is conceivable that, in internet courts handling simple cases, highly specialized judicial AI might make an appearance). Nor are robotic judges, prosecutors, police officers, lawyers, or forensic experts likely to materialize overnight, given the complex challenges—including value‑based considerations—that underpin judicial decision‑making. Nevertheless, moving toward advanced “intelligent case processing” is a goal that should guide the examination of electronic evidence. To achieve this, we must both look to technological innovation to continuously deliver intelligent review tools and encourage practitioners to proactively supplement these solutions with their own efforts.


 

One form of empowerment is the use of specialized techniques and strategies. In current practice, Electronic evidence review may employ techniques such as time‑range filtering, timestamp verification, parallel line‑by‑line comparison, keyword search, dictionary‑based retrieval, and high‑frequency extraction. Their common feature is that they leverage technological means to dramatically enhance the efficiency of the initial review of electronic evidence and to reduce the time required for manual examination. In doing so, while still requiring investigators to scrutinize the substantive content of electronic evidence, they have embarked on a path toward intelligent automation. Although this approach does not entirely replace the human effort of investigators, it has nonetheless yielded significant results in accelerating the review process. Moreover, in areas such as large-scale financial analysis and deep‑level extraction of communication data, it has demonstrated indispensable and substantial improvements.


 

For example, in Case No. [55] of fraud involving Xie et al., discussed in Chapter Four of this book, much of the incriminating evidence consisted of electronic evidence improperly extracted from seized computers and mobile phones. By matching the timing of the evidence‑collection process with the timestamps of the data retrieved, the defense team was able to precisely identify certain electronic evidence that had been generated after the incident—particularly that which only surfaced during the first and second instance proceedings. Moreover, by relying on the timestamps embedded in the Word documents at issue, they could automatically filter out anomalous files; and through keyword or dictionary‑based searches, they were able to clearly determine whether various departments within the implicated company engaged in fraudulent conduct involving fabrication or concealment of facts. These efforts enabled the defense to achieve significant breakthroughs during the supplementary hearing phase of the second trial. Regrettably, however, the defense team devoted too little time to conducting an intelligent review of the electronic evidence, failing to conclusively uncover potential improper involvement by private investigation firms in the case—and the resulting problems of evidence tampering. Under the current judicial environment, the defense team’s work in scrutinizing the evidence still warrants further self‑reflection.


 

The second benefit is to summarize the major new practices. Our country is carrying out The principal achievements of intelligent review of electronic evidence are reflected in the realm of formal examination, where methods such as checksum verification, timestamp validation, blockchain-based verification, electronic signature authentication, and digital watermark analysis are employed to ascertain whether electronic evidence possesses formal authenticity. By comparison, In China, the intelligent review of electronic evidence still suffers from significant shortcomings in substantive examination, primarily manifested in the lack of awareness and competence among case-handling personnel to conduct basic text searches and sophisticated text analyses. Accordingly, shortcomings in substantive examination constitute the primary focus for improvement, with the key to progress lying in conducting in-depth structural analysis and enriching model development. Moreover, China’s emerging practices in intelligent evidence review can be systematically categorized according to different actors, such as case-handling personnel and platform operators.


 

As in the aforementioned case involving Wang XX and others for the crime of organizing and leading pyramid-scheme activities, one of the key pieces of evidence was the forensic samples subjected to electronic data examination. To expedite verification, investigators may employ several approaches: first, compute the checksums of each sample and automatically match them against relevant documents; second, extract timestamp information from the checksums and automatically correlate it with corresponding records; third, retrieve operating system log entries for system startup and shutdown events and automatically cross‑reference them with related forensic collection activities. These methods can be continuously refined to achieve greater levels of automation and intelligence.

Concluding Remarks


 

In summary, in the new era of big data‑driven intelligence, investigators conducting electronic evidence reviews should undertake a comprehensive rethinking across such dimensions as conceptual framework, volume, structure, analytical models, and automation. In this regard, I propose… An academic proposition on the effective examination of electronic evidence: “Directional Guidance” encompasses substantive review, focused review, microscopic review, and intelligent review; “Principle Innovation” covers reconstructive review, mapping review, comparative review, systemic review, associative review, and chain‑based review; and “Institutional Foundations” involves inspection‑based review, experimental review, expert‑opinion review, auxiliary review, and exclusionary review. How this professional landscape stands remains subject to the critique and guidance of the legal community.


 

The future is already here, and evidence-based thinking will surely give rise to an entirely new approach to case handling!


 


 


 

[1] He Jia-hong, ed., Research on Electronic Evidence Law, “Preface,” p. 4, Law Press, 2002.

[2] Fan Chongyi and Li Siyuan, “On the Arrival of the Era of Electronic Evidence,” Journal of Soochow University (Philosophy and Social Sciences Edition), No. 2, 2016, p. 106.

[3] The inaugural year of big data is generally regarded as 2013, marking the global transition into the big data era. Since then, big data technologies have been recognized as the cornerstone of other emerging information technologies, underscoring the critical importance of big data across diverse sectors.

[4]Evidence is the basis of justice. (Bentham, Rationale of Judicial Evidence, Part III, Chapter 1)

[5] Benjamin N. Cardozo, The Nature of the Judicial Process, trans. Su Li, Commercial Press, 1998, p. 80.

[6] An. Yang. Vysinsky, “The Theory of Litigation Evidence in Soviet Law,” Legal Publishing House, 1950, p. 78.

[7]Actions speak louder than words.

[8] See Hiroya Matsuo, “Japanese Criminal Procedure Law,” Volume II, China Renmin University Press, 2005, p. 4; see also Takeshi Tsuchimoto, “Essentials of Japanese Criminal Procedure,” translated by Dong Naodian and Song Yinghui, Wunan Book Publishing Co., Taiwan, 1994, pp. 310–311. Quoted in Guangzhong Chen and Xi Zheng, “On the Principle of Adjudication Based on Evidence in Criminal Proceedings—With a Discussion of Several Issues in the Revision of the Criminal Procedure Law,” Law Science, No. 9, 2011, p. 7.

[9] “Article 317 Facts are established on the basis of evidence.” See Code of Criminal Procedure of Japan, Law number: Act No. 131 of 1948, Law number: Amendment of Act No. 74 of 2011, http://www.japaneselawtranslation.go.jp/law/detail/?id=3364&vm=02&re=02.

[10] “Article 307 (No Evidence No Trial Principle) (1) Fact finding shall be based on evidence.” See

CRIMINAL PROCEDURE ACT of South Korea, https://elaw.klri.re.kr/eng_service/lawView.do?hseq=52939&lang=ENG.

[11] Edited by the Wunan Legal Research Center of Taiwan: “Simplified Sixth-Grade Legal Code,” Wunan Book Publishing Co., Ltd., 2011 edition, pp. 7–26.

[12] The principal legal provisions in this regard include: (1) Article 61 of the Rules on Criminal Procedure of the People’s Procuratorates, which states: “In ascertaining the facts of a case, the People’s Procuratorate shall base its findings on evidence.” (2) Article 69 of the Interpretations of the Supreme People’s Court on the Application of the Criminal Procedure Law of the People’s Republic of China, which provides: “The ascertainment of the facts of a case must be based on evidence.” (3) Article 2 of the Provisions on Several Issues Concerning the Examination and Adjudication of Evidence in Death Penalty Cases, which stipulates: “The ascertainment of the facts of a case must be based on evidence.” (4) Article 235 of the Opinions of the Supreme People’s Court on Establishing and Improving the Mechanism for Preventing Wrongful Criminal Convictions, which mandates: “Adhere to the principle of adjudication based on evidence. The ascertainment of the facts of a case must be grounded in evidence. Evidence shall be examined and evaluated in accordance with statutory procedures. In finding the defendant guilty, the standard of proof requiring evidence that is both reliable and sufficient shall apply.” (5) Article 1 of the Supreme People’s Court’s Issuance of the “Three Regulations” (for Trial Implementation), which stipulates: “The court shall uphold the principle of adjudication based on evidence. The ascertainment of the facts of a case must be based on evidence.”

[13] Zhang Jun, ed., Understanding and Application of Criminal Evidence Rules, Law Press, 2010, p. 7, “Introduction.”

[14] See Cheng Rongbin, “Building a System of Confessions Must Follow the Path of Legalization,” China Criminal Law Journal, No. 3, 2001, p. 13; and Cui Min, “A Review and Reflection on Three Large-Scale Cases of Coerced Confessions,” Criminal Justice Forum, No. 1, 2010, p. 79.

[15] Xu Shenjian: “A Study on the Issue of Separate Trial Proceedings in Joint Crimes,” Journal of the National Prosecutors College, No. 1, 2022, p. 123.

[16] Leading Case 8-1, Supporting Case 15-4.

[17] Qin Zongwen and Ye Wei, “A Study on the Issue of Supplementing Confessions in Cases Involving Plea Bargaining and Acceptance of Punishment,” Journal of Jiangsu Administrative Institute, No. 2, 2019, p. 123.

[18] See Xie Dengke, “On the Standard of Proof in Summary Criminal Proceedings,” Contemporary Law Science, No. 3, 2015, pp. 142–143.

[19] See Qualcomm, “A Study on the Standard of Proof in Summary Criminal Proceedings,” Law Forum, No. 2, 2017, p. 111.

[20] See Sun Yuan, “On the Standard of Proof in Cases Involving Plea Bargaining and Acceptance of Punishment,” Law Application, No. 11, 2016, p. 17.

[21] Leading Case 15-2, Supporting Case 4-10.

[22] Huang Weizhi, “The Appropriate Positioning of ‘Statement of Circumstances’ in Criminal Cases,” Law Science, No. 7, 2007, p. 153.

[23] Article 175 of the Criminal Procedure Law provides: “When reviewing a case, the People’s Procuratorate may request the public security organs to furnish the evidentiary materials necessary for court trial; if it deems that there may be circumstances involving the unlawful collection of evidence as stipulated in Article 56 of this Law, it may require the public security organs to provide an explanation regarding the legality of the evidence‑gathering process.”

Article 56 provides: “...If the collection of physical evidence or documentary evidence does not comply with statutory procedures and may seriously undermine judicial impartiality, such evidence shall be rectified or reasonably explained; if it cannot be rectified or reasonably explained, such evidence shall be excluded.”

[24] Article 59, Paragraph 2 of the Criminal Procedure Law provides: “If the existing evidence fails to establish the legality of the collection of evidence, the People’s Procuratorate may request the People’s Court to summon the relevant investigators or other persons to appear in court and clarify the circumstances; the People’s Court may likewise summon such investigators or other persons to appear in court and clarify the circumstances. The relevant investigators or other persons may also request to appear in court to provide clarification. Upon being notified by the People’s Court, the relevant persons shall appear in court.”

[25] Calling a tail a leg doesn’t make it one.

[26] This is the author’s position as expressed in a speech delivered on October 16, 2021, at the inaugural ceremony of the Research Institute of Zhou Tai Law Firm in Beijing and the “Seminar on Cybercrime and Electronic Data.”

[27] Leading cases 1-1 and 15-1; supporting cases 6-1, 8-1, and 11-1.

[28] Shao Junwu, “On the Adjudicative Proof of Specialized Issues,” Modern Law Science, No. 6, 2000, p. 98.

[29] Li Xuejun, “Approaches to Resolving Specialized Issues in Litigation—With a Discussion on the Improvement of China’s Expert Appraisal System and Statutory Forms of Evidence,” Political and Legal Forum, No. 6, 2020, p. 40.

[30] Leading cases 4-1 and 5-1.

[31] As stipulated in Article 139, Paragraph 1 of the Interpretation of the Supreme People’s Court on the Application of the Criminal Procedure Law of the People’s Republic of China: “The authenticity of evidence shall be examined in light of all the evidence in the case as a whole.”

[32] He Bangwu, “A Normative Interpretation and Practical Path of ‘Comprehensive Determination,’” Hebei Law Science, No. 8, 2019, pp. 96–97.

[33] Leading Case 7-2.

[34] Article 21 of the Measures of the People’s Courts on Filing and Archiving Litigation Documents.

[35] Supplementary Case 13-5.

[36] Li Shuai, “A Study on the Disclosure of Court Supplementary Files from the Perspective of Judicial Reform,” Yunnan Social Sciences, No. 5, 2018, p. 123.

[37] The court’s specific reasoning in its judgment is as follows: “Pursuant to Article 340 of the Rules on Criminal Procedure of the People’s Procuratorates, after reviewing cases transferred by public security organs, and prior to the people’s court rendering a final judgment, if the people’s procuratorate deems it necessary to obtain additional evidence, it may request the public security organs to provide such evidence. Moreover, Article 422 stipulates that, during the trial, when additional evidence required for the court proceedings or further investigative measures are needed, the people’s procuratorate shall independently collect evidence and conduct investigations, and, where necessary, may request the investigative authorities to render assistance. Accordingly, certain evidence in this case was obtained with the procuratorial organ’s request for assistance from the investigative authorities and may be admitted as evidence. This court does not accept the defense counsel’s argument.”

[38] Han Xu, “A Reflection on the Rationality of the Supplementary Investigation System During the Trial Phase—Against the Backdrop of Trial-Centricism,” Law and Business Research, No. 1, 2022, p. 71.

[39] Some scholars have pointed out that, in contrast to electronic traces—which are static, isolated, and fragmented—the movement trajectories generated invisibly by individuals in real life through the use of GPS, mobile phones, citizen cards, the internet, and other technologies—known as “electronic footprints”—are dynamic and relatively comprehensive. See Wang Yujia, “A Study on the Application of Electronic Footprints in Investigative Work in the Context of the ‘Big Data’ Era,” China Science & Technology Information, No. 13, 2016, p. 35.

[40] Supplementary Case 10-7.

[41] Supplementary Case 10-8.

[42] Leading Case 10-2, Supporting Case 11-7.

[43] He Jia-hong et al., “Challenges Posed by Big Data–Driven Investigation to Evidence Law,” People’s Procuratorate, No. 1, 2018, p. 54.

[44] A brick is not a wall. See McCormick on Evidence, P436.

[45] Xiong Jiansheng, “Structural Thinking: A New Paradigm of Contemporary Thought,” Jianghan Forum, no. 11, 2007, p. 65.

[46] Yuan Yi and Yang Jiayu, “A Study on the Theory of Evidentiary Combination from the Perspective of Proof-Value Assessment,” Journal of Beijing Police College, No. 3, 2022, p. 17.

[47] Leading cases 3-1 and 6-1; supporting cases 1-9 and 13-7.

[48] Leading case 9-1, Supporting case 8-3.

[49] Supplementary Case 9-1.

[50] Leading Case 10-1.

[51]In the view of a logician, the whole life is just a huge chain. If you see one ring of it, all the things of the whole chain could be supposed.

[52] L. Belkin, “Essays on Criminal Investigation,” translated by Li Ruiqin, Mass Publishing House, 1983, p. 48.

[53] Leading Case 11-2.

[54] Supplementary Case 6-3.

[55] Leading Case 4-2.


 

(Reading is beneficial; here’s a reference website for purchasing books: With both knowledge and skill, we charge forward wherever ice hockey leads. )


 

Source: Liu Pinxin, “A Legal Perspective from the Mesh”


 

Layout: Wang Xin


 

Beijing Headquarters Address: 17th Floor, China Resources Building, No. 8 Jianguomen North Street, Dongcheng District, Beijing

 


 

Wuhan Branch Address: Room 1001, 10th Floor, Huangpu International Center, Jiang’an District, Wuhan, Hubei Province

 


 

Related News

CONTACT US

Contact us


Beijing Headquarters

Address: 17th Floor, China Resources Building, No. 8 Jianguomen North Avenue, Dongcheng District, Beijing

Phone: 010-64011566

Email: contact@xinglailaw.com


Wuhan Branch Office

Address: Room 1001, Huangpu International Center, Zhaojiatiao, Jiang'an District, Wuhan City

Phone: 027-82288828

Email: admin@xinglailaw-wuhan.com


Hefei Branch Office

Address:Room 1703-1710, Building E, Weilan Business Port, No. 188 Qianshan Road, Shushan District, Hefei City

Phone: 15850561916

Email: contact@xinglailaw-hefei.com

.

Follow us

.

Digital Star Come

Case Consultation

Experienced lawyers offer free, no-obligation consultations to provide tailored solutions.


%{tishi_zhanwei}%

Copyright 2025 Beijing Xinglai Law Firm

Tags: Website Development:300.cn Beijing Certificate

Business license