認真看待修憲程序：修憲之規範性、制度設計與司法審查

Abstract

摘要 2005年修憲成功納入公民複決，卻也設置極高的修憲門檻，如何評價與回應引發爭論。其反映的不單僅修憲程序本身的問題，更是人們「如何看待與理解修憲程序」的思考。修憲程序指出主權者、定義憲法，亦是憲政遊戲規則的決定者，理想上應為共同憲政生活提供指引，過往卻時常被用以成為競逐利益與意識形態的工具，傷害憲法的公信與穩固。 作為民主法的研究，本文旨在對修憲程序自法理基礎、制度設計、修改之程序理性及司法審查進行框架性的體系討論。自釋字第721號解釋出發，梳理並指出「向來修憲界限理論」存在規範性、經驗性與實效性之問題，建議改以「保障當代全民自然權利」並輔以比例原則作為修憲權行使之界限。此外，傳統訴諸制憲者權威的論述，可能疏離了憲法與真正支持其存續的人民，忽視了良善修憲對於改良憲法、深化民主與鞏固憲政的潛能。台灣亦是透過修憲方才形成今日的憲政民主秩序。 修憲的規範性需放置於民主與憲法的內在衝突中理解，過於虛幻的人民圖像往往僅能達成最表面的調和，難以面對當前更加靈活多變的民主侵蝕。自契約論被提出以降，「自主原則」及「同意原則」仍是當前最主要的憲法正當性論述。本文提出「環繞同意原則的辯證框架」以幫助我們更好的對接不同民主理論，並體現在對於「同意主體、同意內涵、同意方法與同意限制」等的不同思考上，其既可以也應當具體化在修憲程序的制度設計上，實現自主與取得同意即為良善修憲的規範性目標，妥善的修憲權利保障乃憲法持續論證治理正當性的必要前提。 修憲權利雖是國民主權的展現，然當修憲權利集結為修憲權力時亦有遭濫用之可能，而需要細緻的制度設計以為引導也實現。修憲程序區分憲章、憲律與法律，而我們對其之考察不該忽略「廣義修憲程序」（小p程序）的重要性，也即小c憲法如何進行憲法變遷的程式。透過比較憲法的視野，本文耙梳了修憲程序在制度設計上具有「分配改變角色、引入多元同意、促進人民參與或審議、保護少數、控管時間、議題與議程、鎖定現狀或替代革命」等制度功能。現行極為單薄的修憲程序難以兼顧亦實現多元功能，建議導入「多階段、多程序、多角色結合多門檻」的思考，根據不同機關的最適功能進行角色分配。在科技與公民意識持續發展的今日，也應朝向取得更廣泛也品質同意的目標前進，使修憲程序努力成為「共識的共識」。 可惜的是，過往修憲程序不但未能定分止爭，甚且成為政治競逐的對象，被工具化以鎖定特定的利益與立場，連帶使憲法遭受質疑，憲法法院也疏忽介入與保護。當代人民有權決定如何開展理想的憲政生活，亦有基於修憲權利的程序自主權與制度性保障要求，修憲程序建置上應實現自主並契合同意原則，並應預設平等決策之地位，需有合理性方得限制。修憲門檻的建置應以「通過門檻己身為最小必要的前提要件」，其自我修改存在自利與矛盾可能，當要求最高的程序及實體正當性。唯有將修憲程序還予人民，使公民社會得直接且明確的就「何謂良善修憲程序」進行討論，並凝聚如何共識的「前共識」，憲法方不會成為一時勝者的戰利品，而是所有人都願意信服並以之共同生活的民主前提。

關鍵詞：釋字第721號解釋、民主法、修憲界限、憲法變遷、防衛性民主、活的憲法、修憲程序、修憲正當程序、修憲門檻、程序理性

Abstract The 2005 constitutional amendment in Taiwan successfully introduced the popular referendum as the constitutional ratification process, yet it provoked ongoing debates due to excessively high thresholds. The amendment process itself reflects how people understand the constitution and democracy, as it constitutes the “rules for changing the rules,” defines sovereignty, and determines the constitution. Ideally, the amendment process should guide constitutional life and help resolve disputes. This requires the amendment process to achieve acceptance by the citizens, thus becoming a “consensus on consensus.” This study aims to develop a comprehensive framework for analyzing the constitutional amendment process in terms of normativity, procedural design, and judicial review. Beginning with Interpretation No. 721, this study critically examines the traditional UCA (Unconstitutional Constitutional Amendments) theory, highlighting its normative, empirical, and practical limitations. It argues that reliance on the authority of past constituent powers is unnecessary; instead, amendment powers should be constrained by the protection of the natural rights of contemporary citizens. The power of the majority is not unlimited. Moreover, the UCA risks detaching the constitution from the people whose continued support sustains its legitimacy. A well-designed amendment process can also protect the constitution from abuse. Since the advent of social contract theory, the principles of autonomy and consent have remained central to constitutional legitimacy. This paper proposes a dialectical framework on the principle of consent, discussing the subject, content, method, and limits of different consent theories. The amendment process should be designed based on the ideal theory of consent (depending on different democracy theories), not merely to realize autonomy. If legitimacy rests on consent, then a constitution must adequately protect the right to amend to maintain ongoing legitimacy. This study introduces the concept of the “small-p process” to analyze constitutional change more precisely, referring to the process of amending the small-c constitution. Amendment process serves multiple goals or functions, including allocating amending power, incorporating diverse consent, facilitating public participation or deliberation, protecting minorities, regulating timing and agenda, entrenching the status quo, and providing alternatives to revolution. Since these functions may be best fulfilled by different institutions, the paper proposes a multi-stage process involving multiple actors and thresholds, which also serves the purpose of distributing and limiting power. A bad amendment process may be instrumentalized to support Abusive constitutionalism. The process should enhance procedural rationality and strive to become a “consensus on consensus,” which is essential for genuine democratic consolidation. The design of the amendment process is not a privilege of constituent power but a means to realize the autonomy of contemporary citizens. Based on the requirement of equality, any amendment process should presume a “simple majority” as the default decision rule, with any supermajority thresholds requiring proper justification. To prevent the paradox or abuse of self-amendment, any changes to the amendment process must be ratified under both the existing and the proposed new process. This article argues that only a sufficient number of people can set amendment thresholds based on mutual agreement. A temporary majority cannot demand that others must obtain more votes than themselves to overturn their decisions, which is an unjust violation of minority autonomy. That the thresholds can only be established through self-restraint also implies they should be subject to an implicit sunset clause. It is essential to allow the people to engage directly in discussions about the constitutional amendment process which shapes the ideals of the constitution and democracy. By achieving a consensus on consensus, the amendment process can thus become the democratic foundation for all.

Keywords: Interpretation No. 721, Law of democracy, Unconstitutional constitutional amendment, Constitutional change, Militant democracy, Living constitution, Constitutional amendment process, Procedural rationality, Constitutional amendment threshold