Seminar Record Session 3｜State Compensation Liability for Judges

I. Comparative Table of Articles

II. Summary of Opinions from the Rapporteur and Discussant Guests

"As judicial power is a part of public authority, if there is unlawful infringement upon people's rights, may the people, and to what extent, seek remedy against the judicial power exercised as a public authority?"

1. What is Judicial Immunity

   Its limitations on state compensation liability encompass the following three major dimensions:

   1. Subject: Civil servants with judicial trial or prosecution duties.
   2. Objective constitutive requirements: Having committed an offense relating to their duties in participating in the trial or prosecution case, upon a final judgment of conviction. This creates extreme limitations; since the implementation of the State Compensation Act over the decades, across all cases searchable on the Judicial Yuan website and legal databases, the success rate of people claiming state compensation has been 0%. Does this comply with the protection of "effective judicial remedy"? Moreover, who watches the watchers?
   3. Subjective constitutive requirements: Intentional liability (for example, the crimes of twisting the law in judgment or abusing prosecution powers are both forms of intentional liability), which differs from the state compensation system of liability based on negligence, creating a systematic narrowing.

2. Retention or Abolition of Judicial Immunity

   Observing the social response to this issue, there are extreme approvals, extreme disapprovals, and partial approvals/disapprovals. Therefore, what stance the draft should adopt is extremely divergent, making it difficult to find common ground, which shows that the issue is worthy of study. In discussion, it can be divided as follows:

   1. Retention or abolition of judicial immunity?

      • (1) Reason 1 for retention: Upholding judicial independence (see the legislative reasoning of the draft, Judicial Yuan Interpretation No. 228)

        • A. Supportive opinions
          First, judicial judgments are not reviewed by other judges (referencing German law BGHZ 50, 14 (19 f.)); second, avoiding future recourse claims.

        • B. Opposing opinions
          First, procedural law inherently contains systems where one judge reviews another judge's judgment, such as appeals, interlocutory appeals, and retrials; second, if judicial independence is used as a justification, it must extend to all duties of a judge, even including non-contentious judges, rather than being limited to trials; third, the principle of proportionality: people petition for state compensation with a current success rate of 0%; taking the protection of judicial independence as the objective, is this measure the least restrictive? For example, cutting off the state's recourse against judges. Fourth, the principle of equality: should members of the Examination Yuan or the Fair Trade Commission not be protected?

        • C. Summary
          Using judicial independence as the justification for the discourse on judicial immunity is questionable.

      • (2) Reason 2 for retention: Protection of the binding force of final judgments

        Res judicata might be shaken, and if shaken, other judges could review it. Therefore, reference can be made to Article 5, Paragraph 1 of the 1981 West German State Liability Act:

        "Where a breach of official duty arises from a binding and final judgment issued by the judicial power which possesses binding force in litigation proceedings, or from a court disposition through which the basis for the judgment was obtained, liability under this Act shall arise only if the breach of official duty constitutes a criminal act and the judgment has been formally set aside. The first sentence shall not apply to third parties not bound by said judgment who suffer damage as a result of such breach of official duty. Other liabilities incurred under this Act concerning the actions of the judicial power shall remain unaffected."

   2. Analysis of the 2021 Executive Yuan Draft

      The legislative reasoning still relies on the protection of judicial independence, which is inappropriate. Another legislative reason is to protect the binding force of final judgments; thus, the constitutive requirements are relaxed to include, first, having committed a criminal offense with a final judgment of conviction, and second, having received a final disciplinary sanction of exemption from duty or dismissal.

      However, whether such relaxed requirements are reasonable is questionable. For example, disciplinary actions might pertain to personal character issues, but may not necessarily involve problems in judicial trials. Furthermore, picking out only one retrial ground as a requirement for filing a state compensation claim raises the question of whether other retrial grounds should also be included.

   3. Summary: Whether judicial immunity should exist is not an absolute proposition.

      • (1) Comparative law perspective

        While German law and Anglo-American law have judicial immunity clauses, they are not universally applicable. The Court of Justice of the European Union (Köbler v. Austria, 2003, EuGH) does not recognize judicial immunity, stating that state compensation liability cannot be restricted by judicial immunity. This is similar to what Professor Lin Hsing-ju mentioned in the morning session regarding Japanese law, where the determination of illegality is stricter, so while the possibility of establishment is smaller, there is no statutory threshold like that in our country.

      • (2) Analysis of the private-sector draft

        The private-sector draft deletes it entirely; I am not opposed to this, but the problem to be faced is how to handle the court's res judicata. Personally, I lean toward maintaining it, otherwise the binding force of judgments cannot be dealt with.

3. Subjects of Judicial Immunity

   Does it include prosecutors? Professor Fan believes that the subjects of judicial immunity do not include prosecutors. Prosecutors belong to the executive power (exclusion theory); they do not make judgments, and their decisions lack binding force. Executive officials are not solely administrative agencies bound by hierarchical command and supervision; reference is also made to Judicial Yuan Interpretation No. 392.

4. Scope of Judicial Immunity

   1. Judgments

      • (1) Judgments terminating the court instance jurisdiction.
      • (2) Rulings replacing judgments
        The objects of adjudication are contentious "litigation cases," including rulings such as declarations of incompetence, litigation cost rulings, and provisional measures.
      • (3) Actions taken to obtain the basis for a judgment
        Aside from the aforementioned violations of substantive law by judges, if a judge violates procedural law during adjudication, such as violating the duty of clarification or infringing upon a witness's personality rights, it also falls within the scope of judicial immunity.
      • (4) Adjudication in non-contentious events?
        The legislative reasoning of our country's state compensation draft mentions that because non-contentious events often possess contentious characteristics, they should be included within the scope of judicial immunity. The professor believes that non-contentious adjudications without contention naturally do not belong to judicial immunity, such as payment orders under the old law.

   2. Excessive duration of litigation

      Unrelated to the judgment itself, this infringes upon the effective legal remedy requirement of fundamental litigation rights and does not constitute judicial immunity; Article 2, Paragraph 2 of the State Compensation Act should apply.

5. Relationship with Criminal Compensation

   Professor Lin Ming-chiang mentioned in the morning session that Article 13 of the State Compensation Act should be directly deleted, reverting to the Criminal Compensation Act. Criminal compensation only compensates for personal liberty, while other rights damages are not compensated, and it operates on a fixed-amount basis, compensating only for the losses suffered. Therefore, while the Criminal Compensation Act certainly undertakes a certain function, the State Compensation Act, beyond personal liberty damages, also encompasses lost profits, showing at least these two differences. Thus, it is argued that Article 13 of the State Compensation Act still retains its necessity for existence.

1. Reference to Anglo-American "Judicial Immunity"

   1. Under Anglo-American law, judges enjoy absolute judicial immunity (in civil matters).

      The historical origin can be traced back to the 16th-century case Floyd v. Barker, where the court held that it lacked jurisdiction to entertain criminal prosecutions against other judges, and that if a judge were to be punished, it should be determined by the King rather than through criminal proceedings. In other words, the court's judicial power stemmed from divine authority, influenced by religious thought.

      Citing cases such as Stump v. Sparkman, Lopez v. Vanderwater, and Dykes v. Hosemann, Judge Chang pointed out that even in circumstances where people's rights were severely infringed and judges clearly engaged in improper conduct, under Anglo-American law they still enjoy judicial immunity and are not subject to tort damages prosecution. Therefore, although according to the research of rapporteur Professor Fan, our country's established state compensation liability for judges is only 1.4% (and a padded 1.4% at that, with the actual being 0%), from the aforementioned case types, it can be seen that the established state compensation liability for American judges is 0%.

   2. Although some commentators (such as Attorney Chiang Jung-hsiang's op-ed "State Non-Liability ≠ Judge Non-Liability") have pointed out using UN documents and other sources that when judges exercise public authority and unlawfully infringe upon people's rights, the state should compensate the people, and Article 13 of the current State Compensation Act should be deleted.

      However, Judge Chang noted that the content of that declaration stems from Anglo-American legal ideology—namely, that errors caused by judges in the exercise of their official duties should be borne by the state without recourse against the judges. Understanding our country's state compensation liability for judges today should not be limited to observing Article 13 of the State Compensation Act alone, but should also incorporate the Criminal Compensation Act. Under the Criminal Compensation Act, when judges exercise public authority unlawfully and cause infringements on people's rights such as personal restraint, state compensation liability is established; thus, the "state compensation" required by the aforementioned international documents should be considered satisfied.

2. Responses and Analyses to the Rapporteur's Paper and the Private-Sector Draft Articles

   1. Responses to the rapporteur's paper

      • (1) The 1981 West German State Liability Act lauded by rapporteur Professor Fan requires conditions such as "breach of official duty or criminal act" and "the judgment being formally set aside," which is actually stricter than current law. What then is the state compensation ratio for German judges? Therefore, Judge Chang harbors doubts regarding the view that "requiring a final judgment of conviction to claim state compensation is an undue restriction on people's right of petition."

      • (2) The rapporteur mentioned that retrial grounds should also be individually examined as conditions for excluding judge liability. However, a retrial indicates a procedural stage of litigation rather than a final determination, and the result of a retrial may still be the same as the original judgment. What is the logical justification for using retrial grounds as a requirement for establishing state compensation liability (i.e., excluding judicial immunity)?

      • (3) He agrees with rapporteur Professor Fan's proposal that litigation delay could constitute a claim under Article 2 of the State Compensation Act. However, paradoxically, while litigation delay constitutes state compensation, in criminal delay—such as Taiwan's currently unique "刑事妥速審判法" (Law for Ensuring Fast and Speedy Trials in Criminal Proceedings)—relief is provided through substantial sentence reductions for defendants. Given that both involve delayed performance, what is the difference in the evaluative basis between civil and criminal proceedings?

   2. Analysis of the private-sector draft articles

      If we revert to general state compensation requirements, when a prosecutor applies for detention and the judge refuses detention, ordering bail instead, would this constitute "failure to exercise official duties" and thus establish state compensation?

      What is the "public authority exercising agency" referred to in Article 3 of the private-sector draft? The legislative reasoning of the private-sector draft mentions that internal recourse may be pursued following state compensation; however, from the perspective of judicial power, the judge is the court, and the court is the judge.

      Furthermore, in investigative actions, which is the public authority exercising agency? Anglo-American law positions prosecutors as executive power, but also treats the prosecutor as the defendant rather than the District Prosecutors Office.

1. Brief Summary of the Basic Stances of the Rapporteur and Discussants

   1. Regarding rapporteur Professor Fan

      Professor Lü noted that in his paper, Professor Fan Wen-ch'ing limits the subject of judicial immunity to judges exercising judicial trial tasks, and restricts its scope to judgments rendered by judges. Other judicial personnel (or administrative personnel with judicial coloration) fall outside the scope of application. As for other non-judgment actions by judges, they revert to Article 2, Paragraph 2 of the State Compensation Act—meaning Article 13 of the State Compensation Act is restrictively interpreted within a certain scope.

   2. Regarding discussant Professor Lü

      Professor Lü stated that he examined Article 13 of the State Compensation Act from the perspective of "trying to explore its underlying legislative intent, necessity for existence, and legitimacy of existence," leading to the conclusion that this article does not need to be deleted and has the necessity to exist.

2. Several Dissenting Opinions of Professor Lü Compared to the Rapporteur's Stance

   1. The legislative intent of Article 13 of the State Compensation Act is likely not about maintaining the binding force of final judgments. This is because such a premise assumes the existence of a judgment, yet acts of infringement upon people's freedom and rights committed by judges or judicial personnel are not necessarily judgments (investigating evidence, summoning witnesses, and other exercises of public authority may all constitute infringements upon people's rights).

   2. Judicial independence as the legislative intent of Article 13 of the State Compensation Act has a major advantage: it allows prosecutors to be included. Reason one: Judges Act and individual judge evaluation procedures basically apply to prosecutors as well. Reason two: Under the Grand Justices' understanding, prosecutors are broadly judicial personnel and less inclined toward executive power. Organizational subordination cannot negate their independence in the exercise of official duties. Actions like suspension of prosecution and non-prosecution dispositions have erga omnes effects, almost equivalent to the res judicata of court judgments. Reason three: Prosecutors, as subjects of public authority, still differ from other agency committees that independently exercise official duties; thus, prosecutors can be protected and regulated alongside judges. To elaborate, the independence of those agencies is institutional decision-making independence, which differs from an individual prosecutor making decisions on non-prosecution, summoning witnesses, and so forth, unlike decisions made by collegiate bodies which are free from interference. Therefore, prosecutors tend to be closer to judges and should be co-subjects of Article 13 of the State Compensation Act.

   3. Professor Lü pointed out that the rapporteur mentioned that the success rate of Article 13 of the State Compensation Act in practice is 0%, interpreting it pessimistically. However, Professor Lü believes statistical interpretation has multiple possibilities, as a 0% success rate does not logically lead to the conclusion that "judges and prosecutors unlawfully exercised public authority leading to state compensation liability, but were excluded due to the requirements of Article 13 of the State Compensation Act." Therefore, a 0% success rate does not necessarily lead to the conclusion that the provisions are overly stringent.

   4. Professor Lü pointed out that the rapporteur cited early German views where judge liability aimed to protect judicial independence, because in some situations, infringements on people are caused even before a final judgment is constituted (such as twisting the law in judgment). State compensation claims might still exist for such infringements of people by judicial power. How to interpret the commission of an offense relating to official duties during trial and prosecution actions actually has no direct relationship with whether the judgment itself is illegal (for example, a corrupt judge whose legal view happens to be correct). The legislature's will has not been amended in the past several revisions, and the retention of its requirements must be meaningful. If a judge or prosecutor commits an offense relating to their official duties, their independence (status, rather than right) should not be protected.

      Finally, there is the issue of the right of recourse. Professor Lü personally believes that even if the state's right of recourse against judges is severed, pressure still exists. Therefore, he believes the current requirements of Article 13 are related to judicial independence. If an offense relating to official duties is committed and there are gross attributable grounds, then the protection of judicial independence does not apply.

1. Analysis of the Fact That the People's Success Rate Is 0% Regarding Judges' State Compensation Liability

   Over the past decades, the exercise of official duties by judges must have caused situations of judicial injustice, yet why has not a single case been discussed as establishing state compensation liability, let alone the exercise of the right of recourse? Professor Chang believes it is not that state compensation cases never materialized, but rather that the data is hidden within the Criminal Compensation Act (or Wrongful Imprisonment Compensation Act). Statistically, what can be confirmed is that there certainly are cases of judicial injustice causing infringement on people's rights; the existence of judicial injustice is also the background for proposing "constitutional review of judgments"—that is, judgments conflicting with constitutional requirements and mandates.

   Therefore, when examining the exclusionary grounds of Article 13 of the State Compensation Act, we should stand from the perspective of protecting people's fundamental rights under Article 24 of the Constitution, treating any restrictive provisions as restrictions on fundamental rights and reviewing whether legitimate grounds exist for such restrictions.

2. Analysis of Requirements in Various Drafts

   This article develops from the two major perspectives of "judicial independence" and "certainty of judgments." Beyond this, we can also observe across the various drafts—the current draft, Article 839, Paragraph 2 of the German Civil Code, and Article 5, Paragraph 1 of the 1981 State Liability Act—that the restrictive provisions on when judicial injustice should establish state compensation liability all include the requirement of "breach of official duty constituting a criminal act."

   In practice under criminal law, establishing cases of "breach of official duty constituting a criminal act" is extremely difficult. For example, in the lobbying case involving a former Supreme Court judge, the prosecutor prosecuted the lobbying judge for twisting the law in judgment, but was ultimately acquitted. Therefore, establishing the requirement that a breach of official duty constitutes a criminal act is extremely difficult. Moreover, whether this requirement is the correct judicial injustice behavior, as the rapporteur stated, is also questionable; thus, the legislature fabricated a fiction here, narrowing and fictionalizing "judicial injustice" into "breach of official duty constituting a criminal act."}

3. Constitutionality of Judicial/Prosecutorial Immunity Clauses

   Article 5, Paragraph 1 of the 1981 State Liability Act, aside from breaches of official duty constituting criminal acts, also includes judgments being formally set aside. This extends to justifying judicial privilege from the perspectives of judicial independence or certainty of judgments. Viewed from the 1981 provision, it is a concept of safeguarding the certainty of judgments to justify it. However, Professor Chang harbors doubts as to "whether the certainty of judgments must necessarily be realized through excluding people's state compensation claims." Although the constitutionality of judicial/prosecutorial immunity is affirmed by Interpretation No. 228 and the Constitutional Court Procedure Act also recognizes constitutional interpretation, with changes in time and space, challenges can be raised; thus Interpretation No. 228 will not constitute the main reason for establishing judge immunity. Therefore, whether using judgment certainty to exclude people's claims is the least restrictive means remains open for discussion.

   German scholars have mentioned that tracking down actions that may violate official duties through judicial administrative supervision does not necessarily exclude people's state compensation claims. It is hoped that the article provides further argumentation on whether excluding people's state compensation claims is necessarily the least restrictive means.

4. Connection and Discrepancy Between State Compensation and Retrial Grounds

   Following the certainty of judgments leads to retrial grounds. The various legislative examples in the drafts do not widely connect with retrial grounds, but if judgment certainty is thoroughly implemented, could grounds for excluding judicial immunity be expanded? For example, does a Supreme Court illegally forming a trial panel constitute twisting the law in judgment (referencing the series "Port of Lies")? Should all current retrial grounds be thoroughly reviewed to see which should be included as constitutive requirements for excluding judicial trial immunity? Under current official draft requirements, gaps exist with retrial grounds; how to design requirements for excluding judicial immunity without sacrificing the protection of people's rights is a legislative challenge.

5. Scope and Definition of Judicial Immunity

   He supports rapporteur Professor Fan's expansion of the concept to "rulings replacing court judgments" and "actions taken to obtain the basis for a judgment." However, excluding provisional legal protection measures on the grounds that they do not constitute final determinations on legal disputes seems questionable. For example, the German Federal Supreme Court has held that provisional orders issued in the form of rulings pursuant to Article 123 of the Administrative Court Act also belong thereto.

   Considering that provisional legal protection proceedings—namely "suspension of execution" and "preservation procedures" prescribed in our country's Petition Act and Administrative Litigation Act—both belong to independent administrative litigation procedures in administrative litigation scholarship, they seem capable of being included within the scope of judicial immunity. This may serve as a reference for further consideration.

6. Application Relationship with the Criminal Compensation Act

   Cases of criminal compensation include lawful and unlawful exercises of public authority by the state. Therefore, it can be seen that under current legal systems, it is "not" the case that because public authority is unlawful and possesses the nature of state compensation that Article 13 of the State Compensation Act applies to exclude people's criminal compensation claims.

   In this situation, it seems to corroborate that there is "no need" to shake the substantive finality of the original judgment in order to compensate victimized citizens. Therefore, from a legislative degege perspective, the direction may head toward narrowing the right of recourse against judges while abolishing the judicial immunity system.

7. State Compensation Requirements for Broadly Defined Judicial Activities

   For broadly defined judicial activities excluded from judicial immunity, such as prosecutors being held liable for state compensation, while such broadly defined judicial activities do not exclude state compensation liability, whether the requirements should be narrowed or revised to incorporate these special circumstances into the framework of culpability requires attention. For comparative law, the German Federal Supreme Court (BGHZ 155, 306, 309f) once held, based on the constitutional principle of judicial independence, that in such individual cases (specifically concerning provisional orders for provisional placement in psychiatric institutions), judge liability is limited to intent and gross negligence, and the substantive correctness of their judgments is not subject to unlimited review, being limited instead to reviewing whether the judgment is supportable while fully respecting the interests of well-functioning judicial administration.