Free China Movement

Evidence · Human Rights · Democratic Future

Take Action

English 简体中文 繁體中文 Français Deutsch Español

🔎

CCE-062 ｜ What Does the Practice of the Republic of China Constitution in Taiwan Reveal?

Free China Movement

CCE-062 ｜ What Does the Practice of the Republic of China Constitution in Taiwan Reveal?

v1.0 ｜ 2026-10-19

CCE-100 Civic Education Project · Index entry CCE-062

What does the practice of the Republic of China Constitution in Taiwan reveal?

Abstract: The Constitution of the Republic of China, which took effect in 1947, went through three sharply different phases in Taiwan: more than four decades hollowed out by the Temporary Provisions, a decade-plus of democratic transition driven by Additional Articles, and the present, in which constitutional organs under a democratic system repeatedly contend with one another. This article argues that this experience most clearly illustrates one point: a written constitution does not automatically constrain power by itself; its effectiveness depends on whether independent institutions are willing to invoke it, whether political forces are willing to pay the price for it, and whether all sides accept the same set of rules of the game.

I. An Ongoing Debate

On December 19, 2025, the Constitutional Court announced its Judgment 114-Hsien-Pan-1, declaring [1][2] the amended provisions of the Constitutional Court Procedure Act — passed in third reading by the Legislative Yuan on December 20, 2024 and promulgated on January 23, 2025 — unconstitutional and void [13]. The amendments provided that no fewer than 10 justices must participate in deliberations and that a declaration of unconstitutionality requires at least 9 concurring votes [3]; yet only 8 justices were in office at the time, and the Legislative Yuan had not consented to the president’s justice nominees, leaving the Constitutional Court unable to render judgments for an extended period. The judgment was rendered by 5 justices, who found the amendment process had “obvious and serious procedural defects” and violated the principle of separation of powers.

The other side’s position likewise speaks in the name of the constitution. According to a Central News Agency report of October 8, 2025, Justices Tsai Tsung-chen, Yang Hui-chin, and Chu Fu-mei issued a joint statement that day [5], holding that “justices must comply with the currently effective provisions of the Constitutional Court Procedure Act” and that the Constitutional Court “may not expand its own powers by unconstitutional means.” The same constitution was used by both sides to argue for diametrically opposite conclusions. This dispute is the entry point to understanding constitutional practice.

II. Concepts: Constitutional Text, Constitutionalism, and “Constitutional Change”

Having a constitution is different from practicing constitutionalism. The former means there exists a written fundamental law; the latter means power is in fact constrained by that law in actual operation, including regular elections, separation of powers, and independent constitutional review. Constitutional scholarship also distinguishes “constitutional amendment” (revising provisions through statutory procedures) from “constitution-making” (adopting a new constitution), as well as “constitutional change” — altering actual operation through interpretation or convention without changing the text. The ROC Constitution’s journey in Taiwan has seen all three.

III. Phase One: A Constitution Without Constitutionalism (1948–1991)

The constitution was promulgated on January 1, 1947 and took effect on December 25 of the same year, but the Temporary Provisions Effective During the Period of Communist Rebellion, promulgated on May 10, 1948, granted the president emergency powers in a manner “superior to and applicable over the constitution,” and exempted presidential re-election from constitutional limits (the re-election exemption was added when the Temporary Provisions were revised in 1960) [7]; martial law was imposed in Taiwan from May 1949 [15]. First-term central representatives were never re-elected, forming the “ten-thousand-year parliament.”

This phase shows that constitutional provisions can be retained as symbols of legitimacy while losing their binding force in substance. Notably, those in power never discarded the constitution, but suspended part of its effectiveness in the name of the “temporary.” This choice later provided reformers with a basis: demanding a “return to the constitution” was politically easier to win support for than demanding the overthrow of the existing system.

IV. Phase Two: Completing the Democratic Transition Through Constitutional Amendment (1990–2005)

The transition did not take the path of adopting a new constitution. Interpretation No. 261 of June 21, 1990 [8] required first-term central representatives who had never been re-elected to cease exercising their powers by the end of 1991; on May 1, 1991 the Temporary Provisions were abolished and the first set of Additional Articles was promulgated at the same time. After several rounds of constitutional amendment, the president was changed to direct popular election (first implemented in 1996), and the National Assembly’s powers were gradually reduced.

The most illustrative case of this phase is the 1999 constitutional amendment passed by secret ballot by the third National Assembly to extend its own term. On March 24, 2000, the grand justices issued Interpretation No. 499 [9], holding that the amendment process violated the principle of openness and transparency, contained “obvious and serious defects,” and declaring the relevant Additional Articles void. The Constitutional Court’s later English-language selection of leading cases noted that this is one of the rare judgments worldwide in which a constitutional court or supreme court declared an entire constitutional amendment unconstitutional [10]. It shows: when the amendment authority seeks to benefit itself, the judiciary can invoke the constitution’s basic principles to stop it.

The seventh constitutional amendment in 2005 abolished the National Assembly; since then, amendment proposals must be initiated by the Legislative Yuan and passed by citizen referendum. The amendment threshold is so high that no amendment has succeeded in the twenty years since.

V. Phase Three: Guarding the Constitution Under a Democratic System

After democratization, the focus of constitutional practice shifted from “ending the emergency regime” to “protecting fundamental rights” and “resolving conflicts among organs.” Interpretation No. 748 of May 24, 2017 [11] held that the Civil Code’s marriage chapter, by not allowing two persons of the same sex to form a permanent union, violated the freedom of marriage under Article 22 and the right to equality under Article 7 of the Constitution, and required the relevant authorities to complete legislation within two years; the Legislative Yuan passed the Act for Implementation of J.Y. Interpretation No. 748 in 2019. This case shows that constitutional review can force political branches to respond when legislation has long stalled, while leaving “in what form” to the legislature to decide.

The 2024–2025 standstill of the Constitutional Court shows the other side: constitutional review itself depends on the cooperation of other organs. Justice nominations require Legislative Yuan consent, and adjudication procedures are set by the Legislative Yuan through ordinary legislation. When the legislative majority stands opposed to the executive, the operating conditions of the Constitutional Court can be changed by ordinary legislation.

VI. The Counterargument: The Constitution’s Own Legitimacy Is Problematic

A weighty criticism holds that the ROC Constitution was enacted in Nanjing in 1946, with the constituent representatives overwhelmingly from mainland China and almost no participation by the people of Taiwan; its territorial and institutional design (such as the five-power system) also does not match the actual scope of governance in Taiwan today. Proponents of this view argue that patching with Additional Articles cannot remedy the fundamental legitimacy gap, and that Taiwan needs a new constitution enacted by its current electorate.

This criticism identifies a real problem: the Additional Articles themselves are premised on “meeting the needs prior to national unification,” and there is indeed a gap between the constitutional text and the actual political community. But in terms of practical effect, the amendment path avoided the domestic division and external crisis that constitution-making might have triggered, and sustained multiple peaceful transfers of power over thirty years. What critics and supporters debate is less whether the constitution has force, and more whether constitutional legitimacy should come from historical continuity or from renewed authorization by the contemporary people. There is no consensus on this question to this day.

In the current conflict among organs, another counter-view holds that the Legislative Yuan’s statutory regulation of deliberation quorums is legitimate democratic-majority supervision of the judiciary, and that a higher threshold can enhance the legitimacy of unconstitutionality judgments; a judgment rendered by 5 justices excluding the 3 who did not participate in deliberations is, conversely, hard for society to accept. This view likewise invokes separation of powers and cannot simply be dismissed as undermining constitutionalism.

VII. Methodological Limitations

First, whether a constitution “works” is hard to quantify; this article relies mainly on grand-justice interpretations and amendment history as institutional events, with less attention to the everyday implementation of constitutional rights in grassroots administration. Second, the Constitutional Court dispute since 2024 is still developing; related judgments (such as the separate opinions attached to Judgment 115-Hsien-Pan-5 [6]) show continuing internal disagreement among the justices, and conclusions are time-sensitive. Third, this article’s quotation of the three justices’ statement is based on the original Central News Agency report [5], with analysis from an academic blog [16] used only for auxiliary cross-checking.

VIII. Conclusion

The practice of the ROC Constitution in Taiwan shows that a constitution’s power lies not on paper but in three conditions: the existence of an independent judiciary willing to invoke the constitution and bear the political cost (Interpretations No. 261, No. 499, and No. 748); the existence of political forces that accept constitutional rulings (the retirement of senior representatives in 1991; the Legislative Yuan’s 2019 legislation pursuant to Interpretation No. 748); and a basic consensus among all sides on “who has the authority to say what the constitution is.” The first two conditions were gradually established in the 1990s, enabling a long-frozen constitution to become the framework for democratic transition. The third condition was severely tested in 2024–2025.

For observers, Taiwan’s experience supports neither the cynical view that “a constitution is mere decoration” nor the optimistic imagination that “writing it into the constitution guarantees protection.” A constitution is a set of public rules requiring continuous maintenance; its effectiveness ultimately depends on whether participants are willing, even when they lose an argument, to still abide by the rules themselves.

Sources

Review record: Editorial independent review, 2026-10-19. Verdict: PASS.

← Back to the CCE-100 Civic Education Project index

Other language editions

This article is also available in Simplified Chinese, Traditional Chinese, French, German and Spanish.