The Political-Philosophical Foundations of the Public Realm: A Normative Floor for Civic Proof

§ 1. Introduction

On 12 July 2024, the Official Journal of the European Union published Regulation (EU) 2024/1689—the so-called AI Act—whose Article 5(1)(b) prohibits the use of subliminal, manipulative, or deceptive techniques capable of materially distorting the behaviour of natural persons or groups in ways likely to cause significant harm, while Recital 29 explicitly identifies the distortion of democratic processes as falling within the Regulation’s scope of application.¹ In December of the same year, the Stanford Internet Observatory released its final report on platform manipulation during the 2024 United States presidential election, documenting the systemic erosion of the voter information environment through algorithmic coordination, networks of inauthentic accounts, and generative content.² During the same period, bulletins from Taiwan’s National Communications Commission (NCC) and monitoring documents from civic organisations including the Taiwan FactCheck Center, IORG, and Doublethink Lab recorded cases of deepfake audiovisual content depicting political figures such as Lai Ching-te and Tsai Ing-wen circulating before and after Taiwan’s presidential and legislative elections of January 2024.³ On a separate institutional axis, Regulation (EU) 2024/1183—eIDAS 2.0—entered into force in April 2024, establishing a mandatory deployment phase for EUDI Wallets between 2026 and 2027;⁴ specifications for Taiwan’s TW DIW (Taiwan Digital Identity Wallet) were released progressively between 2025 and 2026, and the trust list was recorded on a public blockchain.⁵ Over the same period, the Hannah Arendt Center at Bard College in New York organised a consecutive series of public lectures and an annual collected volume under the heading of the digital public realm in 2024–2025, reinserting the political-philosophical discussion of the public realm into media discourse.⁶

These five developments each address, at the surface level, distinct concrete questions: which forms of AI manipulation are prohibited, how electoral information environments are eroded, how deepfake audiovisual content circulates, how digital identity infrastructure is mandatorily deployed, and how the concept of the public realm is being retrieved. At the normative level, however, they jointly point to a question that has not yet been affirmatively borne: when digital infrastructure reshapes the conditions of entry into the public realm, where precisely does the normative floor of democratic legitimacy lie? Which conditions can be borne by engineering carriers such as cryptographic primitives, UX design, and platform governance; which conditions cannot; and how should those that cannot be borne by engineering be formalised for external examination?

The preceding eighteen articles of the civic-proof series have implicitly relied upon the four political-philosophical anchor thinkers as their normative anchors across different layers of analysis, without affirmatively bearing their conjunctive status as the floor of the series. The first article (A1) argues that accountability does not require real-name identification as an antecedent condition;⁷ its dual anchors—procedural due process and republican contestability—already implicitly employ Arendt’s plurality and Mouffe’s agonism. The second article (A3) provides a conceptual positioning of civic proof and a plurality premise derived from Nissenbaum’s contextual integrity,⁸ implicitly relying on Habermas’s validity. The eighth article (A8) establishes the FTLA four-tier governance framework.⁹ The fourteenth article (A14) surveys the cross-jurisdictional redress gap.¹⁰ The fifteenth article (A15) bears civic proof inclusion rights through a three-tier precursor right protection structure and flags the degradation path of the wallet’s three presuppositions (individual ownership / individual identification / individual private key).¹¹ When the series moved into engineering-bearing articles, the sixteenth article (F1) was the first to affirmatively bear, in its §4, the conjunctive coverage of three approaches—Arendt, Habermas, and Pettit—as the normative anchor for the “structurally non-delegable zone”;¹² the seventeenth article (F2) bears the audit trail evidential-chain structure of civic-action receipts through V_receipt C1–C6 and theorems T1–T4;¹³ and the eighteenth article (F3) bears the cognitive engineering of selective disclosure UX through V_ux C7–C10 and the supporter UI three-layer separation.¹⁴ All three engineering articles rely implicitly upon three to four of the four anchor thinkers as their normative floor, without formalising the overall series floor into a unified function spanning the preceding eighteen articles.

The present article is the nineteenth article of the civic-proof series and is also its final article. Its task is to bring the implicit reliances of the preceding eighteen articles into affirmative account, providing the conjunctive coverage conditions and formal skeleton of the four anchor thinkers within the civic-proof series scenario so that the normative claims of the preceding eighteen articles become externally examinable. The scope of its task does not include replacing the original political-philosophical texts. The article adopts an academic register; citation numbers <sup>N</sup> correspond to ≥ 80 sources in the reference section at § 13. Technical terms from Arendt’s original writings in English and Latin (plurality, natality, mortality, vita activa, who/what disclosure), from Habermas’s original writings in German (Öffentlichkeit, kommunikatives Handeln, Geltungsansprüche, Strukturwandel, Wahrheit, Richtigkeit, Wahrhaftigkeit, Faktizität und Geltung), from Pettit’s original writings in Latin and English (non-domination, contestability, editorial democracy, bearer position), and from Mouffe’s original writings in English (agonism, antagonism, legitimate adversary, post-political) are preserved to maintain their original meaning.

The structure of this article is as follows. Section 2 provides conceptual foundations and a summary of forward-links to the series, the principle of separating normative from descriptive claims, and the methodological distinction between conjunction and synthesis. Section 3 provides the formal skeleton, including the PRF function definition, the LegitimacyDegrade function, the complete 32-cell bearer matrix, and four theorems T_PRF1–T_PRF4. Sections 4 through 7 develop in turn the four components SA1 (Arendt’s plurality), SA2 (Habermas’s Öffentlichkeit), SA3 (Pettit’s contestation), and SA4 (Mouffe’s agonism). Section 8 is the core integration section, developing the argument for individual necessity, the methodology of conjunction versus synthesis, a complete reading of the 32-cell matrix, and the series significance of F1 as the bearer of all four components. Section 9 is the counter-argument stress test, comprising a likelihood × impact matrix for five categories of objection—Coeckelbergh, Floridi, posthumanism, digital-democracy optimism, and the meta-objection—together with a special passage for Taiwanese readers. Section 10 articulates the normative-claim responses to the eight series forward-links. Section 11 presents the honesty boundary and 27 open questions. Section 12 offers a conditional academic conclusion.

---

§ 2. Conceptual Foundations and Forward-Links

§ 2.1 Summary of Forward-Links to the Preceding Eighteen Articles

The civic-proof series takes “civic proof as legitimately bearable” as its central theme and progressively establishes, from the first article onward, a layered architecture of normative conditions and engineering bearers. As this article discharges the function of a floor article, it must construct summary correspondences for the eight core bearer articles among the preceding eighteen.

A1 (Accountability Without Real-Name Identification) anchors itself in Bovens’s (2007) dual structure of accountability (answerability + enforceability) and cites three cases—Talley v. California (1960), NAACP v. Alabama (1958), and McIntyre v. Ohio Elections Commission (1995)—to bear the compatibility of “core protection of anonymous political speech” with “ex post accountability.”^7,15 A3 (Civic Proof: Concept and Conceptual Positioning) anchors itself in Nissenbaum’s (2010) contextual integrity, constructs a five-layer abstraction table for civic proof, and separates the normative conditions layer from the engineering implementation layer.^8,16 A8 (DNS Trust Roots vs Identity Trust Roots) bears the multi-stakeholder structure of cross-border trust roots through the FTLA four-tier governance framework (G_industry / G_state / G_recognition / G_oversight).⁹ A14 (The Cross-Jurisdictional Redress Gap) surveys four categories of gap—verifier rejection, issuer erroneous revocation, vendor failure, and cross-border privacy breach—and cites the over-determined structure of conflict of laws.¹⁰ A15 (Civic Proof Inclusion Rights) bears a three-tier precursor right protection structure through Marshall’s (1950) three layers of citizenship and the legal personality right of ICCPR Art. 16, and formalises the degradation path of the wallet’s three presuppositions.^11,17 F1 (The Institutional Limits of AI Agent Delegation in Civic Action) constructs a 5×3 necessary-condition matrix from Tomasev’s (2026) five-element delegation structure (AT/RT/AA/BS/TC) crossed against the civic proof three-element conjunction (𝒩/ℱ/ℬ), marking the RT-ℬ and AA-ℬ cells as Z₃-intrinsically unreachable.^12,18 F2 (Civic-Action Receipts and the Evidentiary Chain) bears the evidential chain for ex post challenges through a holder-side audit trail structure.¹³ F3 (The Cognitive Limits of Selective Disclosure UX) bears the engineering correspondence for CRPD §29 through the supporter UI three-layer separation (assistance-for-understanding layer, operation interface layer, decision-bearing layer).^14,19

The shared structure of the eight articles: each establishes three layers—normative conditions, engineering bearers, and boundary bearings—yet each leaves the affirmative account of “why the four anchor thinkers constitute the floor” unwritten. F1 §4 was the first to provide a methodology of conjunctive coverage by three thinkers, listing Mouffe’s agonism as auxiliary. The present article bears the task of elevating the conjunction of all four thinkers into an affirmative argument for the series normative floor.

§ 2.2 The Principle of Separating Normative from Descriptive Claims

This article strictly separates normative claims from descriptive claims. Normative claims adopt the conditional form: “if any single component of PRF is violated, then LegitimacyDegrade(d) ≥ θ_dem.” Descriptive claims of the form “digital democracy necessarily conforms to PRF” are not adopted. This separation bears three implications.

The first implication is that PRF is not an inductive generalisation from empirical fact. The presuppositions of personhood in the four anchor thinkers’ original writings, the normative structure of Öffentlichkeit, the active-stance bearer position of contestability, and the legitimate adversary of agonism are all formalisations of normative conditions, not empirical descriptions of “how the actual public realm operates.”

The second implication is that a violation of PRF does not necessarily imply that degradation has actually occurred. The value in the [0, 1] interval produced by the LegitimacyDegrade function is a “probability of legitimacy degradation,” not an “empirical measurement of completed degradation.” The value θ_dem ≈ 0.5 is an analytic suggestion, conveying the strict position that “any observable violation signal in a single component triggers a PRF degradation warning.” The precise value of this threshold awaits reverse calibration by cross-national empirical research on democratic legitimacy—for example, the distribution of cases of liberal democracy index degradation in V-Dem annual reports.²⁰

The third implication is that PRF has weaker descriptive power in non-democratic polities. Pettit’s (1997) republicanism presupposes democratic preconditions; for non-democratic polities such as China, Russia, and Iran, the application of PRF constitutes a “descriptive deviation” (the formulation “normative failure” is not adopted). This point is stated explicitly at the honesty boundary in § 11.

§ 2.3 The Methodological Distinction between Conjunction and Synthesis

This article adopts the methodology of conjunction (Konjunktion) and rejects the methodology of synthesis (Synthese / syncretism). The conjunctive methodology holds that within the PRF conjunctive floor framework, each of the four thinkers has individual necessity, is difficult to substitute entirely for the others, and is difficult to have its position fully filled by the other three. The synthetic methodology holds that the four thinkers can be fused into a new version of “composite democratic theory.” The two methodologies have concrete engineering-level differences: the conjunctive floor permits the four thinkers to retain tensions from their original positions (for example, the disagreement between Habermas and Mouffe over consensus), whereas the synthetic ceiling would require those tensions to be dissolved into a new theory. This article’s language avoids the strong normative formulation “absolutely irreducible” and instead adopts the precise expression “necessity within the framework + difficult to fully substitute.”

The strict formulation of the conjunctive methodology is given in Definition 3.2. F1 §4 already established the methodological precedent of conjunctive coverage by three approaches;¹² the present article extends this to a four-approach conjunction and elevates the strict position of the methodology to the bearing of the series normative floor. The contemporary interpretations of Arendt, Habermas, Pettit, and Mouffe by Honig (1993), Calhoun (1992), Lovett (2010), and Norval (2007) are all developed under the stance of “retaining the tensions of the original writings.”^21,22,23,24 The conjunctive methodology of the present article continues this bearing.

§ 2.4 Discipline of Reading the Original Political-Philosophical Texts

This article adopts a discipline of direct engagement with original texts. The four anchor thinkers’ key texts—Arendt (1958) The Human Condition; Habermas (1962) Strukturwandel der Öffentlichkeit, (1981) Theorie des kommunikativen Handelns, (1992) Faktizität und Geltung; Pettit (1997) Republicanism, (2012) On the People’s Terms; Mouffe (2000) The Democratic Paradox, (2013) Agonistics—are all cited directly from the original text with chapter and page numbers; secondary interpretations serve as auxiliary sources (not as substitutes for the original texts). Arendt’s §1 Prologue, §24 (who/what disclosure), §28 (the world held in common), and §44 (vita activa conditions together) provide the ontological definition of plurality.²⁵ Habermas’s (1981) TKH vol. I, §III.3, with its three categories of validity claims (Wahrheit, Richtigkeit, Wahrhaftigkeit), preserves the German original to avoid ambiguity in translation.²⁶ Pettit’s (2012) Ch. 5–Ch. 6 provides the dual division between authorial democracy and editorial democracy.²⁷ Mouffe’s (2000) Ch. 4, with its critique of deliberative consensus and the three conditions of agonism, is cited using page numbers from the English original.²⁸

The indigenisation of the framework to Chinese intellectual traditions proceeds along four lines: Zhu Yunhan’s (2012) High Thoughts in the Clouds (Gāo Sī Zài Yún) on Taiwan’s plural public-realm bearers;²⁹ Lin Zong-hong’s (2020) Island Dust (Dǎo Yǔ Fú Chén) on the “legitimate adversary” relationship between Taiwan’s social movements and political parties;³⁰ Fan Yun’s (2024) twenty-year observation on the women’s movement and Taiwanese civil society;³¹ and Wang Hui’s (2003) China’s New Order on the transformation of the Chinese concept of gong (公).³² These four indigenising studies are cited at the relevant PRF component passages.

§ 2.5 The Relationship between PRF and the Existing Boundaries of F1

F1 §3.1 already established the 5×3 necessary-condition matrix with its three-zone demarcation criteria (Z₁/Z₂/Z₃); the RT-ℬ and AA-ℬ cells were marked as Z₃-intrinsically unreachable because responsibility-bearing requires first-personal mens rea and the contestation bearer position requires Pettit’s active stance, and AI agents are structurally incapable of satisfying either cell.^12,33 Theorem T2 in F2 and theorem T2’ in F3 extended this boundary respectively into the engineering-primitive layers of civic-action receipts and UX.^13,14

The PRF framework of the present article continues this bearing, extending the RT-ℬ ✗ and AA-ℬ ✗ boundaries to all engineering design layers (see Theorem T_PRF5). F1’s ℬ component was originally grounded in a conjunction of three thinkers (Arendt + Habermas + Pettit); the present article elevates Mouffe from auxiliary to an independent fourth component, for reasons detailed in § 8.1. F1’s RT-ℬ ✗ corresponds to a dual violation of the PRF plurality and validity components; F1’s AA-ℬ ✗ corresponds to a dual violation of the PRF contestation and agonism components. Both constitute strong cases of PRF violation.

---

§ 3. Formal Skeleton

§ 3.1 PRF Function Definition

Definition 3.1 (Public Realm Floor (PRF) (a composite term proposed by the present author, synthesising Arendt’s notion of the public realm with the normative ‘floor’ metaphor from political philosophy)):

PRF ≜ ⟨plurality, validity, contestation, agonism⟩

where
  plurality    ≜ the co-presence of multiple who under ontological conditions
                 per Arendt (1958) §1 + §24 + §44
  validity     ≜ Habermas (1962 §II.6+§III.13 ∧ 1981 §III.3 ∧ 1992 Kap.VII–VIII)
                 Öffentlichkeit structure ∧ three Geltungsansprüche
                 ∧ Sluice model
  contestation ≜ Pettit (1997 Ch.6 ∧ 2012 Ch.5–6) active-stance bearer position
                 ∧ four institutional conditions of editorial democracy
  agonism      ≜ Mouffe (2000 Ch.4 ∧ 2013 Ch.1) three conditions of
                 legitimate adversary
                 (mutual recognition ∧ shared rules ∧ agonistic frame)

Definition 3.2 (Distinction between conjunctive floor and synthetic ceiling) — the following three expressions constitute the definition–expansion–closure structure of PRF:

PRF as conjunctive normative floor:
  PRF_violated(d) ⇔ ∃ i ∈ {plurality, validity, contestation, agonism}
                    : violation(d, PRF_i)

PRF is not a synthetic ceiling:
  ¬ (PRF ⇔ syncretic-fusion(plurality, validity, contestation, agonism))
  ¬ (PRF_satisfied(d) ⇒ d attains the highest level of democratic legitimacy)

Two implications of the conjunctive floor: first, if any one of the four components is violated, PRF as a whole is treated as violated within this framework. Under the PRF conjunctive setting of this article, the trade-off reading—strengthening one component to compensate for a violated component—is not adopted. Whether other normative approaches (Rawls’s justice / Sen’s capability / Nussbaum’s dignity / Honneth’s recognition) can provide partial supplementation is listed as a ceiling or boundary condition (see Definition 3.3). Second, satisfying PRF is a necessary condition (sufficiency is not within the scope of the claim), meaning “not having fallen below the floor”—the claim does not extend to “having reached the ceiling.”

Definition 3.3 (Non-uniqueness of the composite floor):

Composite floor (not unique standard):
  PRF ⊆ NormativeFloors
  where NormativeFloors ⊇ {Rawls justice, Sen capability,
                           Nussbaum dignity, Honneth recognition, ...}

PRF is a composite floor of democratic legitimacy and does not claim to be the unique floor. Rawls’s (1971) A Theory of Justice and justice as fairness,³⁴ Sen’s (2009) The Idea of Justice and the capability approach,³⁵ Nussbaum’s (2011) Creating Capabilities and dignity-based capabilities,³⁶ and Honneth’s (2014) Freedom’s Right and recognition theory³⁷ are all other normative approaches. This article adopts an open-boundary stance: no approaches beyond the four thinkers are included in the conjunction, but other approaches are acknowledged as possible ceiling or boundary conditions of PRF.

§ 3.2 The LegitimacyDegrade Function

Definition 3.4 (LegitimacyDegrade function):

LegitimacyDegrade(d) ≜ σ( β₁·violation_intensity(d, plurality)
                         + β₂·violation_intensity(d, validity)
                         + β₃·violation_intensity(d, contestation)
                         + β₄·violation_intensity(d, agonism) )

where
  σ                        ≜ logistic sigmoid (same form as P_degrade in F1 §3.2)
  violation_intensity(d,i) ∈ [0, 1]  "bearer degradation intensity of design d
                                       on component i"
  β₁ = β₂ = β₃ = β₄ ≈ 1.0  equal weighting across the four thinkers
                              (prior setting, pending empirical calibration)

Theorem T_PRF0 (PRF violation implies LegitimacyDegrade ≥ θ_dem; θ_dem is an analytic suggestion, not empirically calibrated):

PRF_violated(d) ⇒ LegitimacyDegrade(d) ≥ θ_dem ≈ 0.5

Note on θ_dem calibration: σ(β·1.0) ≈ 0.731 (complete violation of a single component); σ(β·0.5) ≈ 0.622 (moderate violation of a single component); σ(0) = 0.5 (baseline). Setting θ_dem = 0.5 adopts a strict stance, meaning “any observable violation signal in any single component triggers a PRF degradation warning.” A lenient stance (θ_dem = 0.6) permits a grey zone of “minor degradation.” This article adopts the strict stance as its floor and marks the lenient stance as response space for counter-argument W_PRF4 in § 9.1. The precise calibration of θ_dem awaits reverse calibration from the distribution of liberal democracy index degradation cases in V-Dem 2024,²⁰ the Bertelsmann Transformation Index,³⁸ and Freedom House annual reports.³⁹ This article marks θ_dem as an analytic suggestion, stated explicitly in the high-risk claim list at § 11.

§ 3.3 The 32-Cell Bearer Matrix

Crossing the eight core bearer articles among the preceding eighteen (A1 / A3 / A8 / A14 / A15 / F1 / F2 / F3) against the four PRF components produces a bearer matrix of 8×4 = 32 cells. Each cell adopts one of three markers: ● core bearing (the article’s primary normative claim falls directly within that component), ○ partial bearing (the article’s normative claim provides indirect support for that component), and — no correspondence (the article’s normative claim has no significant relation to that component).

Article	plurality	validity	contestation	agonism	Primary normative axis
A1 Accountability Without Real-Name Identification	●	○	○	●	Dual structure of accountability and protection of anonymous political speech
A3 Civic Proof: Concept and Positioning	●	●	○	—	Nissenbaum CI plurality premise and five-layer abstraction table
A8 DNS vs Identity Trust Roots (FTLA)	○	●	●	●	FTLA four-tier governance framework and HM conjunctive claim
A14 Cross-Jurisdictional Redress Gap	—	○	●	●	Four categories of cross-jurisdictional redress gap and conflict-of-laws structure
A15 Civic Proof Inclusion Rights	●	●	●	○	Three-tier precursor right protection and wallet three presuppositions
F1 Institutional Limits of AI Agent Delegation	●	●	●	●	Delegation 5×3 matrix (core series PRF bearer)
F2 Civic-Action Receipts and Evidentiary Chain	○	●	●	○	V_receipt C1–C6 and Theorem T2 Z₃-intrinsic unreachability; agonistic engineering of CF1–CF5 stress tests
F3 Cognitive Limits of Selective Disclosure UX	○	●	○	●	V_ux C7–C10 and supporter UI three-layer separation
● Core bearing count	4	5	4	5	(● total: 18; ○ total: 11; — total: 3 across 32 cells)

The bearer determination reasoning for each cell is as follows.

A1 × plurality (●): A1 §2 cites NAACP v. Alabama (1958) and McIntyre v. Ohio (1995), directly linking “core protection of anonymous political speech” to plurality. The co-presence of multiple who in the public realm ought not to be diminished by antecedent natural-person identity binding. Arendt §28 “the world we have in common” directly bears this component.

A1 × agonism (●): The dual anchors of A1 include procedural due process and republican contestability. The latter’s subject-position structure of contestation directly corresponds to Mouffe’s agonism presupposition of legitimate adversary; anonymity allows legitimate adversaries to exercise political expression rights without identity binding, forestalling the degradation of agonism into antagonism.

A1 × validity (○): The Bovens (2007) accountability structure in A1 (answerability + enforceability) provides indirect support for the validity component’s Wahrhaftigkeit; the requirement that one “provide explanations when challenged” operationalises Habermas’s Geltungsansprüche.⁴⁰

A1 × contestation (○): A1 §2 cites Pettit (1997 / 2012) on contestability, but the active-stance bearer position of contestation is employed in A1 only as auxiliary to the asymmetric treatment of power-holders and power-subjects.

A3 × plurality (●): Nissenbaum’s CI plurality premise in A3 §3.1—“no privacy ‘in general’“—directly corresponds to plurality. The non-transferability of multiple contexts constitutes the bearer of plurality at the informational normative layer.¹⁶

A3 × validity (●): The three minimum necessary conditions of civic proof in A3 §6—“publicly claiming membership, internally accountable, not externally identifiable”—bear Habermas’s Richtigkeit claim; membership claims must be examinable within Öffentlichkeit.

A3 × contestation (○) and A3 × agonism (—): A3’s normative claims focus on the foundations in privacy philosophy and conceptual landscape differentiation and do not enter into the discussion of adversarial political structure.

A8 × validity (●): The G_state tier of the FTLA four-tier governance framework in A8 directly bears the institutional carrier of the Sluice model in Habermas’s Faktizität und Geltung, Kap. VII; the legitimacy of democratic legislation derives from the communicative channel between “informal opinion formation” and “formal decision-making” in the two-tier Sluice.⁴¹

A8 × contestation (●): A8’s G_recognition and G_oversight tiers directly bear the four institutional conditions of Pettit’s editorial democracy; cross-border mutual recognition and dispute resolution mechanisms are the institutionalisation of contestation.

A8 × agonism (●): A8’s HM conjunctive claim, which rejects the view that “nationalisation of the identity sphere is inevitable,” presupposes the legitimate adversary structure of multiple legitimate actors (state, industry, standards organisations, civil society). The tension between G_industry and G_state is the concrete site of agonism.

A8 × plurality (○): A8’s plurality bearing lies primarily in the multi-stakeholder structure of the governance tier, not in the ontological meaning of Arendt; hence marked ○.

A14 × contestation (●): A14 §3’s four categories of cross-jurisdictional redress gap directly correspond to Pettit’s contestation institutional conditions; when redress channels are dysfunctional, contestation bearers have no means to exercise an active stance.

A14 × agonism (●): A14 §4’s over-determined conflict-of-laws structure bears Mouffe’s agonism “agonistic frame among legitimate adversaries”; cross-jurisdictional parties as legitimate adversaries require an operational dispute-resolution mechanism.

A14 × validity (○) and A14 × plurality (—): A14’s normative claims focus on redress procedures and do not enter into the ontological bearing of plurality; indirect support for validity lies in the procedural validity of procedures.

A15 × plurality (●): A15 §3’s survey of the experiential exclusion of undocumented persons, stateless persons, and exiles directly corresponds to plurality. The co-presence of multiple who ought not to be excluded by the enrollment conditions of infrastructure.

A15 × validity (●): The Marshall three-tier civic citizenship and ICCPR Art. 16 legal personality right bearing in A15 §4’s precursor right structure corresponds to Habermas’s Faktizität und Geltung, Kap. VIII, zivilgesellschaftliche Öffentlichkeit; the institutional bearer of civic participation requires the guarantee of legal personality.^17,41

A15 × contestation (●): A15 §6.2’s analysis of the degradation of the wallet’s three presuppositions (individual ownership, individual identification, individual private key) directly corresponds to Pettit’s contestation. If all three presuppositions hold simultaneously, supported decision-making degrades into substituted decision-making and the contestation bearer position is lost.

A15 × agonism (○): A15’s differentiation of multiple stakeholders—supporter / guardian / community—provides indirect support, but Mouffe’s agonism and legitimate adversary are not independently borne in A15.

F1 × plurality (●): F1 §4.1’s approach through Arendt’s plurality + natality + unique disclosure constitutes the core bearing for the “structurally non-delegable zone.” F1’s three reasons why AI agents fail the plurality component—“no natality,” “no mortality stake,” “plurality is at the ontological level”—are direct applications of the plurality component.¹²

F1 × validity (●): F1 §4.2’s approach through Habermas’s kommunikatives Handeln + Geltungsansprüche constitutes the core bearing for “responsibility-bearing.” F1’s claim that AI agents have “no intention of their own to express” is precisely the disablement of the Wahrhaftigkeit claim bearer position.

F1 × contestation (●): F1 §4.3’s approach through Pettit’s contestability constitutes the core bearing for “contestation bearer.” F1 §3.2 P_degrade-TC—“when calibration is self-reported by the agent with no external contestation channel, ℬ degrades into ritual”—is the concrete operationalisation of the contestation component.

F1 × agonism (●): Although F1 §4.2(iii) lists Mouffe as auxiliary, F1’s claim that AI agents “lack a subject position for mutual recognition” is a direct application of the three conditions of Mouffe’s agonism and legitimate adversary. All four components are fully borne in F1; F1 is the first instance in the series of a full core-bearer of all four components.

F2 × validity (●): V_receipt C1–C6 in F2 directly bears the substantive truth of the audit trail and the sincerity of holder consent, corresponding to Habermas’s Wahrheit and Wahrhaftigkeit claims.

F2 × contestation (●): F2 §6’s court admissibility route through FRE 901(b)(9) holder-controlled receipts directly bears the “contestability” of Pettit’s editorial democracy; civic-action receipts serve as the evidential chain for ex post challenges.

F2 × plurality (○): F2’s indirect support for plurality lies in the holder-side bearer of receipts (each holder as an independent who), but this is not the primary axis.

F2 × agonism (○, elevation note): This article adopts an expansive interpretation, elevating F2’s agonism bearing from the original “no correspondence” to “partial bearing.” The five CF1–CF5 counterfactual stress tests proposed in F2 §9 (likelihood-by-mechanism inferences when certain V_receipt conditions fail) are themselves methodologically an engineering operationalisation of the agonistic frame: using “legitimate adversary” counter-arguments heard within a matrix form as a design criterion, incorporating the multi-subject structure of “the audit chain not monopolised by a single issuer” into the receipts design. This bearing has not yet reached the ● core-bearing intensity (the primary axis of F2 remains the cryptographic evidential chain), but it exceeds the — “no significant relation” category; hence marked ○.

F3 × validity (●): V_ux C7–C10 in F3 directly bears the holder’s informed consent, corresponding to the UX performativity of Habermas’s Wahrhaftigkeit claim.

F3 × agonism (●): F3 §7’s supporter UI three-layer separation (assistance-for-understanding layer, operation interface layer, decision-bearing layer) corresponds to the legitimate adversary structure of Mouffe’s agonism. The supporter holds a legitimate adversary position within supported decision-making (not replaceable but necessary); the three-layer separation prevents degradation toward either the antagonism or the post-political consensual end.

F3 × plurality (○) and F3 × contestation (○): F3’s indirect support for the two components lies in capacity-aware UX and the supporter UI three-layer separation supporting holders in exercising contestation, but the active-stance bearer is not independently borne in F3.

Matrix statistics: across 32 cells, ● totals 18 (56%), ○ totals 11 (34%), and — totals 3 (10%). The ● cell counts across the four components are: plurality 4, validity 5, contestation 4, agonism 5. Agonism and validity share the highest count, reflecting the agonistic engineering operationalisation bearing of the CF1–CF5 stress test methodology in F2 §9 (see the F2 × agonism elevation note) and the legitimate adversary structural bearing of the F3 supporter UI three-layer separation. The — cells are concentrated at A3–agonism and A14–plurality (F2–agonism is no longer listed as — following elevation), indicating that the engineering articles have partially reinforced the distribution of Mouffe’s agonism across the series.

§ 3.4 The Four Formal Theorems T_PRF1–T_PRF4

The individual necessity of the four thinkers is formalised by four theorems.

Theorem T_PRF1 (Necessity of plurality):

¬ ∃ design d : (d satisfies validity ∧ contestation ∧ agonism)
              ∧ (d violates plurality)
              → d constitutes a legitimate public realm

Proof sketch: if d violates plurality (for example, replacing plural subjects with a single bearer), then even if all three of validity, contestation, and agonism are satisfied, the object of d’s legitimacy degrades into “a procedure with no co-present subjects.” Arendt §28 states explicitly that the constitution of “the world we have in common” itself requires multiple who; without plurality there is no public realm.²⁵ F1 §4.1’s argument for AI agents on “no natality” is a concrete application of this theorem.

Theorem T_PRF2 (Necessity of validity):

¬ ∃ design d : (d satisfies plurality ∧ contestation ∧ agonism)
              ∧ (d violates validity)
              → d constitutes a legitimate public realm

Proof sketch: if d violates validity (the Öffentlichkeit structure collapses or the bearer positions for the three Geltungsansprüche are disabled), then even if all three of plurality, contestation, and agonism are satisfied, the object of d’s legitimacy degrades into “strategic interaction” (speech acts have already lost their bearer). Habermas (1981) §III.3 states explicitly that democratic legitimacy derives from kommunikatives Handeln (strategisches Handeln does not bear this status).²⁶ F1 §4.2’s argument for AI agents on “no Wahrhaftigkeit bearing” is a concrete application of this theorem.

Theorem T_PRF3 (Necessity of contestation):

¬ ∃ design d : (d satisfies plurality ∧ validity ∧ agonism)
              ∧ (d violates contestation)
              → d constitutes a legitimate public realm

Proof sketch: if d violates contestation (the active-stance bearer position is disabled or the four conditions of editorial democracy are not satisfied), then even if all three of plurality, validity, and agonism are satisfied, the object of d’s legitimacy degrades into “consensus with no channel for challenge.” Pettit (1997) Ch. 6 states explicitly that freedom as non-domination requires the institutional bearing of contestability as a condition;⁴² even with complete plurality, the absence of contestation degrades into hegemonic consensus. F1 §3.2 P_degrade-TC is a concrete application of this theorem.

Theorem T_PRF4 (Necessity of agonism):

¬ ∃ design d : (d satisfies plurality ∧ validity ∧ contestation)
              ∧ (d violates agonism)
              → d constitutes a legitimate public realm

Proof sketch: if d violates agonism (the three conditions of legitimate adversary are disabled), the object of d’s legitimacy degrades into one of two dysfunctions. The first degradation is into antagonism: adversaries share no rules, and contestation degrades into civil war. The second degradation is into post-political consensuality: adversary positions are absorbed into consensus, and contestation degrades into ritual. Mouffe (2000) Ch. 4 states explicitly that democratic legitimacy requires agonistic pluralism.²⁸ F3 §7’s supporter UI three-layer separation is a concrete application of this theorem.

Corollary C_PRF (Four-thinker conjunctive floor):

PRF_satisfied(d) ⇔ (d satisfies plurality)
                  ∧ (d satisfies validity)
                  ∧ (d satisfies contestation)
                  ∧ (d satisfies agonism)

Theorem T_PRF5 (Engineering unreachability lemma for the existing engineering design layer):

∀ engineering design d :
  ¬ (∃ d : PRF_violated(d) ∧ LegitimacyDegrade(d) < θ_dem)

Implications of T_PRF5: no engineering design—including wallet schema, cryptographic primitives, UX engineering, governance frameworks, or legal bearers—can bypass the PRF floor. F1’s RT-ℬ ✗ (first-personal mens rea disabled) constitutes a dual violation of the PRF plurality and validity components; F1’s AA-ℬ ✗ (contestation bearer disabled) constitutes a dual violation of the PRF contestation and agonism components. Theorem T2’ in F3’s UX unreachability boundary shares the same source as T_PRF5; UX engineering primitives do not alter ontological structure and therefore cannot bypass the PRF floor.

---

§ 4. SA1: Arendt’s Plurality

§ 4.1 The Three-Layer Ontological Structure

The concept of plurality in Arendt’s (1958) The Human Condition operates at the ontological level—it does not belong to the domain of political theory. The original §1 Prologue states explicitly: “the fact that men, not Man, live on the earth and inhabit the world”—meaning that humanity consists of plural and concrete “human beings” co-present on earth; the single, abstract “Man” does not figure within Arendt’s position.^{25, p.7} Arendt further articulates this plurality across three layers of existential conditions: natality (the condition of birth), mortality (the condition of death), and uniqueness (the distinctiveness of each who).

The first layer, natality: §1 Prologue and §44 “The Life of the Mind in Action” (pp. 313–320) establish natality as a fundamental condition of human existence, namely “each man begins something new by entering the world.” Natality is understood in the ontological sense as the source of “the capacity for action” (not as a description of biological fact); every new act presupposes the actor as a subject capable of “beginning.”^{25, p.177} Honig’s (1993) contemporary interpretation of Arendt further links natality to the “openness of political action.”²¹

The second layer, mortality: §44 lists mortality as the existential condition paired in opposition to natality, meaning that human finitude renders the bearing of action an irrevocable stake-taking. The “re-instantiation” of AI agents destroys this existential basis of stake-taking: if an actor can be indefinitely copied or reset, stake degrades into a variable of computation (the bearing of existence is already lost). F1 §4.1 already established this argument.¹²

The third layer, uniqueness: §24 “The Disclosure of the Agent in Speech and Action” (pp. 175–181) establishes the distinction between “action disclosing who (not what).” The who is the distinctive identity of “who I am,” which can only be disclosed to others through “action”; the what is a description at the level of attributes.^{25, p.179} D’Entrèves’s (1994) reading of Arendt further links the who/what distinction to the irreducibility of political subjecthood.⁴³ Benhabib’s (1996) The Reluctant Modernism of Hannah Arendt brings this distinction into dialogue with contemporary pluralism.⁴⁴

§ 4.2 Contrast with Liberal Pluralism

Arendt’s plurality cannot be equated with liberal pluralism. Liberal pluralism—as in Berlin’s (1958) Two Concepts of Liberty or Rawls’s (1993) Political Liberalism—operates at the level of political theory, bearing the normative claim of “coexistence of plural values.”^45,46 Arendt’s plurality operates at the ontological level, bearing the existential condition of “the co-presence of multiple who under the conditions of natality and mortality.”

The two differ in three respects. First, liberal pluralism presupposes the individual as a “subject of choice,” while Arendt’s plurality presupposes the individual as a “subject of disclosure.” Second, the multiplicity of liberal pluralism consists in “the juxtaposition of values,” while the multiplicity of Arendt’s plurality consists in “the co-presence of who.” Third, the toleration of liberal pluralism is “tolerance of differing values,” while the plurality of Arendt’s plurality is “acknowledgement of who disclosure.” Every occurrence of plurality in this article should be understood within the ontological boundaries of Arendt’s sense, to avoid a slide into liberal pluralism.

§ 4.3 Correspondence with the Chinese “Qun” (群) Tradition

The correspondence between the Chinese qun (群) tradition and Arendt’s plurality is an authorial analogy that requires verification by the academic community. Zhu Yunhan’s (2012) analysis of Taiwan’s plural public-realm bearers in High Thoughts in the Clouds links the concept of qun to the plural bearing of Taiwan’s civil society: five major bearers comprising the street, the legislature, online spaces, referenda, and social movements.²⁹ Lin Zong-hong’s (2020) Island Dust further incorporates the “legitimate adversary” relationship between Taiwan’s social movements and political parties into the analysis, bearing the plurality dimension of the qun concept.³⁰ Wang Hui’s (2003) research on the transformation of the Chinese concept of gong (公) in China’s New Order provides a comparative case for the plurality dimension of the Chinese public realm.³²

The structural homology between the qun tradition and Arendt’s plurality lies in this: qun emphasises the existential condition of “plural human beings co-present,” resonating with Arendt’s §1 Prologue position of “men, not Man.” Yet the historical bearer conditions of qun (the hierarchical structure of Confucian self-cultivation, family governance, state governance, and world pacification) differ from the horizontal plurality of Arendt’s ontology. This article adopts the conservative stance of “structural homology but different historical bearer conditions,” stated explicitly at the honesty boundary in § 11.

§ 4.4 Correspondence to Series Scenarios

A1 §2 cites Talley v. California, NAACP v. Alabama, and McIntyre v. Ohio Elections Commission to establish a direct link between “core protection of anonymous political speech” and plurality.¹⁵ Anonymity within this bearer structure refers to the right to public-realm action of multiple who, and exceeds the category of “a technical choice to conceal identity.” If civic proof necessarily binds to the natural-person identity, the plurality dimension is reduced antecedently to a “single channel of identity authentication.”

A15 §3’s survey of the experiential exclusion of undocumented persons, stateless persons, and exiles positions the bearing of plurality at the examination of enrollment conditions; if the mandatory phase of wallet deployment causes those without wallets to withdraw from the public realm, plurality is disabled. A15 §6.2’s degradation analysis of the wallet’s three presuppositions provides the concrete mechanism of this case.¹¹

F1 §4.1 employs the three-layer existential conditions of Arendt’s plurality (natality + mortality + uniqueness) as the core reason for AI agents’ “structural non-delegability”: AI agents lack natality (no existential condition of “being capable of beginning”), lack a mortality stake (no irrevocable stake-taking), and lack uniqueness disclosure (no disclosure of “who I am”); the plurality component is therefore structurally incapable of being borne in AI agent scenarios.¹²

§ 4.5 Three Cases

Case 1: The 2014 Sunflower Movement (318). The 24-day occupation of Taiwan’s Legislative Yuan from 18 March to 10 April 2014 bore plural bearers under “plurality inside and outside the chamber”: citizen representatives inside the chamber, street crowds outside, the informational bearing of the g0v open-source community, the discursive spaces of PTT and Facebook, and the parallel coverage of mainstream and independent media.^29,30 Each bearer has irreducible bearing in the plurality dimension; no single bearer alone could bear the totality of the movement.

Case 2: The 2019–2020 Hong Kong Anti-Extradition Movement. The movement bore plurality through the “leaderless” (mo dai toi, lit. “no main stage”) structure—multiple bearers (student unions, social movement groups, religious communities, overseas networks, and anonymous citizens) co-present. The same period also demonstrated the boundary slippage between plurality and agonism: when the “legitimate adversary” structure collapsed, the movement’s bearer degraded into antagonism. Lin Zong-hong (2020) and Fan Yun (2024) provide a record of Taiwan’s civil society bearing of this movement.^30,31

Case 3: Deepfake cases in the 2024 Taiwan presidential and legislative elections. Deepfake audiovisual content depicting political figures such as Lai Ching-te and Tsai Ing-wen circulated before and after the elections; NCC bulletins and monitoring records by civic organisations including the Taiwan FactCheck Center, IORG, and Doublethink Lab document the cases.³ The impact of deepfakes on plurality lies in this: if the disclosure of who is replaced by AI-generated what forgeries, the ontological foundations of plurality are eroded. Simultaneously, this constitutes an attack on the bearer position for Habermas’s Wahrhaftigkeit claim in the validity dimension.

---

§ 5. SA2: Habermas’s Öffentlichkeit

§ 5.1 The Three-Layer Normative Structure

Habermas’s concept of Öffentlichkeit spans three anchor texts: (1962) Strukturwandel der Öffentlichkeit,⁴⁷ (1981) Theorie des kommunikativen Handelns,²⁶ and (1992) Faktizität und Geltung.⁴¹ The three texts establish a three-layer normative structure of Öffentlichkeit.

The first layer, the ideal speech situation: (1981) TKH vol. I, §III.3, defines kommunikatives Handeln as “action oriented toward reaching mutual understanding” and presupposes four counterfactual conditions of the ideal speech situation: all relevant parties may participate, there is no power distortion, there is no time constraint, and there is no internal or external coercion.²⁶ The ideal speech situation is a counterfactual idealisation, not an empirical claim; its function is to provide a normative reference point for “reasoned consensus.” Bohman’s (1996) Public Deliberation, in its contemporary interpretation of the ideal speech situation, further operationalises these conditions as four normative conditions of “procedural deliberation.”⁴⁸

The second layer, the three Geltungsansprüche: TKH §III.3 decomposes any speech act of kommunikatives Handeln into three categories of validity claims—Wahrheit (truth-claim: a true/false assertion about the external world), Richtigkeit (rightness-claim: a right/wrong assertion about the normative world), and Wahrhaftigkeit (sincerity-claim: an expression of genuine intentions about the inner world).^{26, pp.410–427} This article preserves the German original to avoid ambiguity in translation. The disablement of the bearer positions for the three claims in digital scenarios constitutes the core mechanism of violations of the validity component.

The third layer, reciprocal recognition: Faktizität und Geltung, Kap. VII and Kap. VIII, establishes the Sluice model; the legitimacy of democratic legislation derives from the communicative channel between “informal opinion formation” and “formal decision-making” in the two-tier Sluice.^{41, Kap.VII} The Sluice model positions the informal opinion formation of zivilgesellschaftliche Öffentlichkeit (civil-society public realm) as upstream of institutional decision-making; the communication between the two layers is the concrete mechanism of reciprocal recognition.

§ 5.2 Contrast with Empirical Public Opinion

Habermas’s Öffentlichkeit cannot be equated with empirical public opinion. Empirical public opinion consists of opinion polls, media monitoring, social-media topic volumes, and similar empirical measurements. Öffentlichkeit is a normative structure comprising three elements: rational-critical debate, ohne Ansehen der Person (without regard to the status of persons), and Privatleute zum Publikum (private persons assembled into a public).^{47, §II.6}

The two differ in three respects. First, the input to empirical public opinion is “aggregation of opinions,” while the input to Öffentlichkeit is “rational-critical debate of speech acts.” Second, the output of empirical public opinion is “measurement of majority preferences,” while the output of Öffentlichkeit is “reasoned consensus under speech acts.” Third, empirical public opinion under algorithmic media can be distorted by platform manipulation (as in Case 3 in § 5.5), while the normative structure of Öffentlichkeit remains a normative reference point even under algorithmic media.

§ 5.3 Fraser’s Counterpublics Correction

Fraser’s (1990) Rethinking the Public Sphere critiques the presupposition of singularity in Habermas’s (1962) Öffentlichkeit; the co-presence of multiple subaltern counterpublics challenges the normative homogeneity of Öffentlichkeit.⁴⁹ Warner’s (2002) Publics and Counterpublics further systematises the plural constitution of counterpublics into a contemporary theory of publics.⁵⁰

This article adopts the stance of “expansion while retaining the floor.” The co-presence of multiple counterpublics is consistent with the PRF plurality component, and each counterpublic must still individually satisfy the normative structure of the PRF validity component (rational-critical debate + ohne Ansehen der Person + Privatleute zum Publikum). Habermas’s (1992) Faktizität und Geltung, Kap. VIII, with its Sluice model for zivilgesellschaftliche Öffentlichkeit, has partially responded to Fraser’s critique: the informal opinion-formation layer accommodates multiple counterpublics, while the formal decision-making layer maintains a single institutional bearer.⁴¹ Calhoun’s (1992) collection Habermas and the Public Sphere already includes Fraser’s (1990) critique; this article adopts Calhoun (1992) as the entry-point reference for the Fraser approach.²²

Asen’s (2017) research on the marketisation degradation of Öffentlichkeit under neoliberalism reinforces the contemporary bearing of Fraser’s critique: when the public realm enters the marketisation of neoliberalism, rational-critical debate degrades into “the expression of consumer preferences.”⁵¹ This article states explicitly at the honesty boundary in § 11 that Habermas’s historical analysis of the 18th-century bourgeois public realm has a Eurocentric tendency.

§ 5.4 Correspondence with the Chinese “Gong” (公) Tradition

The correspondence between the Chinese gong (公) tradition and Habermas’s Öffentlichkeit is an authorial analogy that requires verification by the academic community. Wang Hui’s (2003) historical study of the transformation of the Chinese concept of gong in China’s New Order links gong to the normative structure of the Chinese public realm, from the Song dynasty literati-officialdom’s “all-under-heaven as common” (tianxia wei gong) to modern civil society’s “public reason.”³² Zhu Yunhan (2012) links the gong concept to the institutional bearer of Taiwan’s public realm: five major institutional bearers comprising the Council of Grand Justices’ constitutional interpretations, legislative operations, the referendum system, local assemblies, and media oversight.²⁹

The structural homology between the gong tradition and Habermas’s Öffentlichkeit lies in this: both presuppose the institutional bearing of a normative structure. However, the historical bearer conditions of the gong tradition (the imperial examination system, literati-officialdom, and public deliberation) differ from those of Habermas’s 18th-century bourgeois public realm (coffee houses, salons, and print media). This article adopts the conservative stance of “structural homology but different historical bearer conditions,” stated explicitly at the honesty boundary in § 11.

§ 5.5 Three Cases

Case 1: Eight years of failure in GDPR cookie banners. Since GDPR entered into force in 2018, the consent mechanism of cookie banners has borne the institutional performance of Habermas’s Wahrhaftigkeit claim for “sincere expression of one’s own intentions.” In practice, however, cookie banners have been pervasively designed using dark patterns into a ritual of “clicking agree.”⁵² The validity dimension degrades in this case; the Wahrhaftigkeit bearer position for consent is disabled. F3 §3 and §4 already provided the UX-layer analysis of this case.¹⁴

Case 2: Taiwan’s 2018–2021 referendum cases. Multiple referendum cases in Taiwan between 2018 and 2021 (the East Asian Olympics name rectification referendum, the death penalty abolition referendum, the energy transition referendum, local referenda, and the 18-year-old constitutional amendment for civic rights, among others) bear the institutional performance of Taiwan’s Öffentlichkeit dimension.²⁹ The public discourse before and after each referendum and the civic organisation bearers (the Referendum Alliance, the Taiwan Youth Association, the Labour Rights Association, the Environmental Protection Union, the Citizens of the Earth Foundation) constitute the concrete bearer structure of Öffentlichkeit.^30,31

Case 3: Platform manipulation in the 2024 U.S. presidential election. The Stanford Internet Observatory’s December 2024 final report documents the systemic erosion of the voter information environment through algorithmic coordination, networks of inauthentic accounts, and generative content.² Vaidhyanathan’s (2018) Antisocial Media, Tufekci’s (2017) Twitter and Tear Gas, and Gillespie’s (2018) Custodians of the Internet provide structural analyses of platform manipulation.^53,54,55 The impact of this case on the validity dimension lies in: the bearer positions for the three Geltungsansprüche in platform-mediated environments are eroded by algorithmically coordinated manipulation.

---

§ 6. SA3: Pettit’s Contestation

§ 6.1 The Three-Layer First-Personal Requirements

Pettit’s concept of contestation spans two anchor texts: (1997) Republicanism⁴² and (2012) On the People’s Terms.²⁷ Pettit (2012) Ch. 3 through Ch. 6 establishes three layers of first-personal requirements.

The first layer, Equally Accepted Terms: Pettit (2012) Ch. 3 sets “citizens must be able to accept the terms of public policy under conditions of equality” as the first condition of democratic legitimacy.^{27, Ch.3} This first-personal requirement cannot be substituted by a representative. Any acceptance by a representative must be conditional upon “citizens being able to accept in the first person.”

The second layer, Equally Accessible Influence: Pettit’s (2012) Ch. 5 Eyeball Test sets “citizens are able to look power-holders in the eye without flinching” as the second condition of democratic legitimacy.^{27, Ch.5, pp.81–105} This first-personal requirement presupposes that citizens have “accessible channels of influence,” including challenge, complaint, voting, and protest.

The third layer, Active Stance: Pettit (1997) Ch. 6 and (2012) Ch. 6 define the core of contestation as “active stance”—the contestation subject must “actively” adopt a stance of dissent, not “passively” accept a consultation of opinions.^{42, Ch.6} The significance of this distinction for AI agent scenarios lies in this: AI may serve as an input function for opinion consultation, but cannot serve as an active-stance bearer.

§ 6.2 The Three-Layer Examination of Bearer Position

F1 §4.3 already established the three-layer examination of bearer position: the first layer (execution), the second layer (endorsement), and the third layer (bearer).¹² The division into three layers corresponds to the formal/substantive representation distinction in Pitkin’s (1967) The Concept of Representation.⁵⁶

First layer — execution: delegable. AI agents may bear execution-layer tasks such as wallet operation, form filling, and document parsing.

Second layer — endorsement: conditionally delegable. AI agents may bear endorsement-layer tasks such as “suggesting,” “summarising,” and “preliminary assessment,” but a human must bear final responsibility.

Third layer — bearer: not delegable. The contestation bearer position must be borne first-personally by a human, comprising three conditions: responsibility-bearing (mens rea), stance-bearing (active stance), and consequence-bearing (mortality stake). The F1 RT-ℬ ✗ and AA-ℬ ✗ boundaries are precisely the unreachability of this third layer.^12,33

§ 6.3 Freedom as Non-Domination

Pettit (1997) Ch. 2, “Liberty as Non-Domination,” defines republican freedom as non-domination—“not being subject to arbitrary interference by any person or institution.”^{42, Ch.2} This definition contrasts with Berlin’s (1958) negative freedom (freedom from interference): non-domination is stronger than negative freedom, because even in the absence of actual interference, the mere existence of “the possibility of being arbitrarily interfered with” still constitutes a condition of domination.⁴⁵

Lovett’s (2010) A General Theory of Domination and Justice further formalises the three conditions of non-domination as “the absence of arbitrary power”: the actual exercise of power, the dispositional capacity of power, and the arbitrary character of power.⁵⁷ The absence of any one of the three conditions reduces the level of domination.

The implication of freedom as non-domination for digital scenarios lies in this: when infrastructure such as wallet schema, verifier algorithms, and issuer revocation mechanisms can “arbitrarily” alter the behavioural conditions of holders, this constitutes a condition of domination even if no actual interference has occurred. F1 §3.2 P_degrade-TC—“when calibration is self-reported by the agent with no external contestation channel”—is a concrete case of this domination mechanism.

§ 6.4 Correspondence with the Chinese “Jianyi” (諫議) Tradition

The correspondence between the Chinese jianyi (諫議, direct remonstrance) tradition and Pettit’s republican contestation is an authorial analogy. The three-layer structure of the Chinese historical “remonstrance” (jiànzhèng) tradition—direct remonstrance (zhí jiàn), indirect admonition (fěng jiàn), and death remonstrance (sǐ jiàn)—bears the historical mechanism of first-personal active stance.³² The three-layer structure of Taiwan’s Council of Grand Justices petitioning system—petition by citizens, petition by judges, and petition by agencies—bears the contemporary institutional correspondence of contestation in Taiwan.²⁹

The structural homology between the jianyi tradition and Pettit’s contestation lies in this: both presuppose the bearing of first-personal active stance. However, the historical bearer conditions of jianyi (the ruler–minister relationship, literati-officialdom ethics, and the unity of family and state) differ substantially from the historical bearer conditions of Pettit’s republican democracy (citizenship, equal political rights, and institutional checks and balances). This article adopts the conservative stance of “structural homology but different historical bearer conditions,” stated explicitly at the honesty boundary in § 11.

§ 6.5 Three Cases

Case 1: Taiwan’s Council of Grand Justices petitioning system. Articles 78 to 80 of the Taiwan Constitution establish the Council of Grand Justices constitutional interpretation system. Since the Constitutional Litigation Act entered into force in 2022, the three-layer structure of petitions by citizens, judges, and agencies bears the contemporary institutional correspondence of contestation in Taiwan.²⁹ Cases including Interpretation No. 803 (Indigenous hunting rights), Interpretation No. 791 (decriminalisation of adultery), and Interpretation No. 748 (same-sex marriage) bear the concrete practice of the bearer position.

Case 2: CJEU Schufa Holding C-634/21 (2023). The Court of Justice of the European Union’s December 2023 ruling on Schufa Holding GmbH expanded the interpretation of the right not to be subject to automated decision-making under GDPR Article 22; contestation rights for any automated decision-making must have an institutional bearer.⁵⁸ This case bears the EU legal correspondence of Pettit’s editorial democracy’s four institutional conditions (transparency, predictability, contestability, and correctability).

Case 3: SEC v. Jarkesy, 144 S. Ct. 2117 (2024). The U.S. Supreme Court’s June 2024 ruling on SEC administrative proceedings held that “the right to a jury trial must be afforded in fraud cases within administrative proceedings.”⁵⁹ This case bears the legal bearing of the bearer position in administrative proceedings: when the penalty imposed by an administrative proceeding already affects “private rights,” the first-personal contestation of the bearer cannot be substituted by administrative decision-making.

---

§ 7. SA4: Mouffe’s Agonism

§ 7.1 The Distinction between Agonism and Antagonism

Mouffe’s agonistic concept spans two anchor texts: (2000) The Democratic Paradox²⁸ and (2013) Agonistics.⁶⁰ Mouffe draws a strict distinction between agonism and antagonism: agonism presupposes a legitimate adversary, while antagonism presupposes an enemy.

The three conditions of the legitimate adversary in agonism are mutual recognition, shared rules, and an agonistic frame.^{28, Ch.4} The shared presupposition of these three conditions is that the adversary’s right to political expression is recognised as legitimate, even where the adversary’s position is opposed to one’s own. The enemy structure of antagonism is otherwise: the adversary’s right to political expression is not recognised, and interaction degrades into a logic of annihilation.

Mouffe incorporates the concept of adversarial politics from Schmitt’s (1932) The Concept of the Political while rejecting Schmitt’s friend/enemy logic of annihilation.⁶¹ As Schmitt was a Nazi jurist, the citation must be stated explicitly at the honesty boundary in § 11: Mouffe absorbs Schmitt’s insight that “politics essentially contains an adversarial dimension,” but transforms the enemy structure into the legitimate adversary structure.

§ 7.2 The Conjunction of Habermas and Mouffe

Mouffe’s (2000) Ch. 4 critique of Habermas’s deliberative consensus states explicitly: “the political cannot be reduced to rational consensus.”^{28, pp.102–105} Mouffe argues that reducing democratic legitimacy to consensus dissolves the agonistic dimension of democracy and degrades into post-political consensuality: adversary positions are absorbed into consensus, and contestation degrades into ritual.

This article adopts the stance of conjuncting Habermas and Mouffe (not treating them as mutually exclusive), relying on the methodology already established in F1 §4.2.¹² The conjunctive structure of the two thinkers lies in this: Habermas’s Öffentlichkeit and three Geltungsansprüche bear the normative structure of speech acts; Mouffe’s three conditions of legitimate adversary bear the normative structure of adversarial politics. The two are conjoined at the point that “political speech requires subjects with human personality,” while the tension between “consensus vs antagonism” is retained. Norval’s (2007) Aversive Democracy, in its dialogue between Mouffe and Cavell, further systematises this conjunction.²⁴

§ 7.3 Correspondence with the Chinese “Zheng-You” (諍友) Tradition

The correspondence between the Chinese zheng-you (諍友, remonstrating friend) tradition and Mouffe’s agonism is an authorial analogy requiring verification by the academic community. The concept of zheng-you originates in the Confucian Analects (季氏): “of beneficial friendships there are three kinds: friendship with the upright, friendship with the sincere, and friendship with the well-informed.” Friends must be able to offer candid remonstrance, bearing the structure of “mutually recognised legitimate adversary.” The Chinese historical inheritance of zheng-you (candid remonstrance among literati-officialdom including Han Yu, Fan Zhongyan, Wang Anshi, and Gu Yanwu—including Fan Zhongyan’s public-realm bearing in “Yueyang Tower Inscription” of “grieving before the world grieves, rejoicing after the world rejoices”) constitutes the historical correspondence of agonism.³² Wang Hui’s (2003) research on the historical transformation of the Chinese concept of gong also notes that the Neo-Confucian normative position of “the gentleman harmonises but does not conform” (jūnzǐ hé ér bù tóng) has a structural homology with the semantic bearing of Mouffe’s three conditions of legitimate adversary (mutual recognition, shared rules, and agonistic frame).³²

The structural homology between the zheng-you tradition and Mouffe’s agonism lies in this: both presuppose mutual recognition among legitimate adversaries; the adversary’s right to political expression, even when opposed to one’s own position, is recognised as legitimate; and neither degrades into hostile annihilation. However, the historical bearer conditions of zheng-you (Confucian ethics, literati-officialdom relationships, private friendship, and the hierarchical presuppositions of ruler–minister dialogue) differ substantially from Mouffe’s institutional agonism (multi-party politics, public debate, media oversight, and civic-organisation confrontation). The former bears an ethical structure in the private and semi-public realm; the latter bears the public-realm structure of the multi-party institutional system. This article adopts the conservative stance of “structural homology but different historical bearer conditions,” stated explicitly at the honesty boundary in § 11. The bearing of agonism in the PRF component for the Chinese public realm cannot directly substitute the institutional structure of a multi-party system by invoking the zheng-you tradition.

§ 7.4 Three Cases

Case 1: Inter-party relations in the Legislative Yuan following Taiwan’s 2024 elections. Following Taiwan’s January 2024 elections, the Legislative Yuan entered a structure in which no single party commands a majority: the multi-party operation of the Democratic Progressive Party, the Kuomintang, and the Taiwan People’s Party constitutes the concrete practice of agonism.³⁰ Each party’s policy debate, legislative procedure, and coalition dynamics (e.g. “blue-white cooperation” versus “green-camp minor coalition”) constitute the institutional performance of legitimate adversary. At the same time, there is structural pressure from polarisation mechanisms distorting into antagonism: echo-chamber algorithms, the culture of social-media dunking, and the contraction of cross-echo-chamber dialogue spaces constitute concrete empirical challenges.

Case 2: Boundary slippage in the 2019–2020 Hong Kong Anti-Extradition Movement. In the early phase of the movement, plurality and agonism were both borne, with the co-presence of plural bearers and the legitimate adversary structure coexisting. In the middle phase, the boundary between agonism and antagonism slipped, with interactions between police and protesters degrading into enmity. In the later phase, the antagonism structure solidified, and after the National Security Law entered into force, the legitimate adversary structure was annihilated.³⁰ This article adopts the “normative distinction” stance (empirical distinction is not within the scope of this article’s bearing): the boundary between agonism and antagonism is easily elided in actual politics, but in normative terms the two are strictly different.

Case 3: Polarised politics in the 2024 U.S. presidential election. The 2024 U.S. election bears the impact of agonism degrading into antagonism under a polarisation mechanism; the legitimate adversary structure is eroded by the demonisation of the Other. Benkler’s (2018) Network Propaganda already documented the polarisation mechanism of the 2016 U.S. election.⁶² The Stanford Internet Observatory’s 2024 report further documents platform manipulation and polarisation in the 2024 election.² This case bears the contemporary fragility of agonism.

---

§ 8. The PRF Conjunctive Floor Argument (Core Integration)

§ 8.1 The Argument for Individual Necessity of the Four Thinkers

The core proposition of this article is that the conjunction of PRF’s four thinkers constitutes the normative floor of the civic-proof series; within the PRF conjunctive floor framework, each of the four thinkers has individual necessity and is difficult to be fully substituted by the other three (though not in the absolute sense of irreducibility). This proposition is borne by the four formal theorems T_PRF1 through T_PRF4 (see § 3.4). This section develops the concrete arguments for individual necessity.

Individual necessity of plurality: if d satisfies the three thinkers—validity, contestation, agonism—but violates plurality (for example, replacing plural subjects with a single bearer), then d degrades into “a procedure with no co-present subjects.” The concrete mechanism unfolds across three layers: the three Geltungsansprüche of validity can still be borne by a single bearer, but “the speech acts of different who” disappear; the active stance of contestation can still be borne by a single bearer, but “the challenges of different who” disappear; the legitimate adversary of agonism can still be borne by a single bearer, but “the adversarial politics of different who” disappear. The three bearers degrade into “talking to oneself” under the single-bearer structure; the public realm fails to constitute itself. F1 §4.1’s argument for AI agents on “no natality” is the concrete application of this mechanism.

Individual necessity of validity: if d satisfies plurality, contestation, and agonism but violates validity, then d degrades into “strategic interaction” (strategisches Handeln); the bearing of speech acts (kommunikatives Handeln) is lost. The concrete mechanism unfolds across three layers: the co-presence of plural who remains, but there is no rational-critical debate; the active stance of contestation remains, but there is no bearing of the three Geltungsansprüche; the legitimate adversary structure remains, but there is no normative structure of reciprocal recognition. The three bearers degrade into “power struggle” under the strategic-interaction structure; democratic legitimacy fails to constitute itself.

Individual necessity of contestation: if d satisfies plurality, validity, and agonism but violates contestation, then d degrades into “consensus with no channel for challenge.” The concrete mechanism lies in this: the co-presence of plural who and speech acts can reach “reasoned consensus,” but if there is no institutional bearer of contestation, consensus degrades into hegemonic consensus; the legitimate adversary structure remains, but there is no first-personal bearing of an active-stance bearer. F1 §3.2 P_degrade-TC is the concrete application of this mechanism.

Individual necessity of agonism: if d satisfies plurality, validity, and contestation but violates agonism, then d degrades into one of two dysfunctions. The first degradation is into antagonism: adversaries share no rules, and contestation degrades into civil war. The second degradation is into post-political consensuality: adversary positions are absorbed into consensus, and contestation degrades into ritual. F3 §7’s supporter UI three-layer separation is the concrete application of this mechanism; the supporter, as a legitimate adversary, benefits from the three-layer separation to prevent degradation toward either extreme.

The four theorems jointly imply Corollary C_PRF: PRF_satisfied(d) is the conjunction of all four thinkers; the absence of any single one implies PRF_violated(d).

§ 8.2 The Methodology of Conjunction versus Synthesis

This article adopts the methodology of conjunction; the methodology of synthesis is not within the scope of this article’s bearing. The two methodologies differ in two respects.

The conjunctive methodology: the four thinkers retain their respective original-text positions and tensions within PRF; fusion into a new theory is not claimed. For example, the tension between Habermas’s consensus position and Mouffe’s agonism position is retained; the two thinkers are conjoined at the point that “public political speech requires subjects with human personality.” The advantage of the conjunctive methodology is that it preserves the independent space of interpretation for each of the four original texts and avoids the conceptual confusion of syncretic pseudo-synthesis.

The synthetic methodology: fusing the four thinkers into a new version of “composite democratic theory,” dissolving the tensions of the four original texts. The risk of the synthetic methodology lies in this: syncretic pseudo-synthesis easily forces “concepts at different levels” (for example, Arendt’s ontological plurality and Mouffe’s political-theory agonism) into a merger, losing the conceptual precision of the original texts.

This article strictly adopts the conjunctive methodology and marks the conjunctive structure at every occurrence of “the four thinkers,” to avoid sliding into synthesis. F1 §4.4’s three-approach conjunction table already established this methodological precedent;¹² this article extends it to a four-approach conjunction and elevates it to the bearing of the series normative floor.

§ 8.3 Complete Reading of the 32-Cell Bearer Matrix

The reading of the 32 cells of the matrix in § 3.3 is as follows.

Horizontal reading (the four-component bearing of each article): A1, A3, and A14 each bear 2 ● cells (core bearing); A8 and A15 each bear 3 ● cells; F1 has 4 ● cells (full core-bearer of all four components); F2 has 2 ● cells plus 2 ○ cells (including the upgraded ○ for agonism); F3 has 2 ● cells plus 2 ○ cells. F1’s full core-bearing of all four components is the first such instance in the series, reflecting that F1’s 5×3 matrix (Tomasev’s five elements × civic proof three-element conjunction) already implicitly employs the conjunction of all four thinkers: AT-ℬ and RT-ℬ correspond to plurality; ℱ corresponds to validity; AA-ℬ corresponds to contestation; TC-ℬ corresponds to agonism (including the legitimate adversary structure of supporter UI).

Vertical reading (the bearing intensity of each component): plurality has 4● + 3○ + 1— = A1/A3/A15/F1 as core, A8/F2/F3 as partial, A14 as non-corresponding; validity has 5● + 3○ + 0— = A3/A8/A15/F1/F2/F3 as core, A1/A14 as partial; contestation has 4● + 3○ + 1— = A8/A14/A15/F1/F2 as core, A1/A3/F3 as partial, no non-corresponding cells; agonism has 5● + 1○ + 2— = A1/A8/A14/F1/F3 as core, F2 elevated to partial, A3/A15 non-corresponding (A15’s existing ○ bearing is retained; non-correspondence in the agonism column concentrates at A3, while A14 has been corrected). After the F2–agonism elevation, agonism and validity jointly share the highest bearing count across components (5 ● cells).

Independent argument for the F2–agonism elevation: As noted in the critical observation in § 5.2, F2 §9’s five counterfactual stress tests (CF1: issuer collusion, CF2: verifier collusion, CF3: holder memory loss, CF4: platform failure, CF5: cross-border jurisdictional conflict) use likelihood-by-mechanism inference to assess the robustness of V_receipt design; the methodology itself constitutes an engineering operationalisation of the agonistic frame. CF1–CF5 presuppose that “legitimate adversary” counter-arguments (the issuer may be dishonest; the verifier may be unfair; the platform may cease operation) are heard within the matrix form, internalising Mouffe’s three conditions of legitimate adversary (mutual recognition, shared rules, agonistic frame) into the robustness criterion of cryptographic design. This bearing intensity has not yet reached ● core-bearing (the primary axis of F2 remains the cryptographic evidential chain and the contestation dimension of ex post challenges), but it exceeds the — “no significant relation” category; hence elevated from the original — to ○. This elevation does not alter the ● core-bearing determinations of the § 3.3 matrix; it only reinforces the distributed bearing of the agonism dimension across series engineering articles.

Analysis of non-corresponding (—) cells: after the elevation, only two concentrations of — cells remain: A3–agonism and A14–plurality. This distribution reflects that engineering articles (F2, F3) under expansive interpretation have partially reinforced the agonism-dimension bearing. This article’s PRF framework elevating Mouffe from auxiliary to independent component is precisely the series-level systematisation of this reinforcement. A14’s non-correspondence for plurality lies in this: A14’s normative claims focus on redress procedures and do not enter into the ontological bearing of plurality. A3’s non-correspondence for agonism lies in this: A3’s normative claims focus on the foundations of privacy philosophy and conceptual landscape differentiation and do not enter into the discussion of adversarial political structure.

Dynamic updating of the matrix: if the interpretation of other series articles is updated or new series articles are added, the matrix must be remade. The 32-cell bearer determinations of this article depend on the current interpretation of the normative claims of the preceding eighteen articles; if interpretations change, cell attributes (●/○/—) must be re-determined. This risk is stated explicitly in the high-risk claim list at § 11.

§ 8.4 The Series Significance of F1 as the Bearer of All Four Components

F1 (article 16) is the first article in the series to fully bear all four components at the core level. Three items of series significance follow from this determination.

First, F1’s 5×3 matrix (Tomasev’s five elements × civic proof three-element conjunction) is the first “necessary-condition matrix for delegation scenarios” in the series; this article’s 8×4 PRF matrix is its methodological extension, expanded from “necessary conditions for delegation scenarios” to “the series normative floor bearer relations.” The two matrices are methodologically isomorphic (see § 2.5).

Second, F1 §4’s conjunctive coverage by three approaches (Arendt + Habermas + Pettit) was the first article in the series to “affirmatively bear the conjunctive floor.” F1 §4.2(iii) already introduced Mouffe as auxiliary; the argument for agonism’s individual necessity in this article’s § 8.1 elevates Mouffe from auxiliary to independent component. The determination that F1 is the bearer of all four PRF components depends on an expansive interpretation of F1 §4’s three-approach conjunction, elevating F1 §4.2(iii)‘s “Mouffe supplement does not weaken the conclusion” to the bearing of the agonism component.

Third, the Z₃-intrinsic boundaries of F1’s RT-ℬ ✗ and AA-ℬ ✗ are the first “structurally unsatisfiable” cells in the series. The two cells correspond to simultaneous violations of two PRF components: RT-ℬ ✗ corresponds to a dual violation of plurality and validity (first-personal mens rea disabled); AA-ℬ ✗ corresponds to a dual violation of contestation and agonism (active-stance bearer and legitimate adversary disabled). This determination is formalised in Theorem T_PRF5 of this article.

The series significance of F1 as the bearer of all four components can be summarised thus: F1 has the highest bearer density among the series normative floor articles. The PRF framework of this article takes F1’s existing bearing as its anchor and extends it to the bearer relations of all eighteen preceding articles.

---

§ 9. Counter-Argument Stress Tests

§ 9.1 Coeckelbergh’s Relational Personhood (medium-high impact)

Counter-argument: Coeckelbergh’s 2020 AI Ethics Ch. 4–6 and 2024 “Why AI Undermines Democracy” argue that the expansion of relational personhood allows the boundary of “personhood subjects” to include AI agents.¹² If AI acquires personhood through relational practice, PRF plurality’s multiple who co-presence can accommodate AI; PRF validity’s Wahrhaftigkeit claim bearers can include AI; and PRF’s anthropocentric presupposition is wholly replaced.

Response: This article adopts the “one step down, not two” position of F1 §4.1.³ Even if law recognises a form of relational personhood, the ontological argument for plurality in § 4.1 (natality + mortality + uniqueness) does not depend on legal recognition. Law may recognise a certain status, but it cannot change the ontological structure of that subject.

The Arendtian ontological definition of the plurality component is anchored in “multiple who co-present under conditions of natality + mortality”; even if an AI agent’s relational personhood is legally recognised, it still lacks natality (each individual as a new beginning) and a mortality stake, so the plurality component still cannot be borne by AI. The validity component’s Habermasian Wahrhaftigkeit requires speakers to “sincerely express their own intentions”; the problem of intentional attribution to AI agents remains philosophically unresolved, so the validity component still cannot be borne by AI. Coeckelbergh’s 2024 internal tension lies in this: the book explicitly states that AI constitutes a threat to democracy (e.g. deepfakes, micro-targeting, opinion manipulation) and calls for a human democratic subject’s capacity to counter these; this position is consistent with PRF’s “public realm entry conditions centring on the human democratic subject.”

The contestation and agonism components under the Coeckelbergh stress test take “one step down, not two”: it is possible to expand AI agents’ operational scope (e.g. AI agents as auxiliary bearers of contestation), but they still cannot replace the position of the human active-stance bearer. Likelihood is assessed as medium (Coeckelbergh 2020 + 2024 has formed structural bearing in Chinese-language academic circles and parts of continental European AI ethics); impact is assessed as medium-high (the strongest blow is to the plurality component, but F1 §4.1 already provides a response).

§ 9.2 Floridi’s Infosphere Monism (medium impact)

Counter-argument: Floridi’s 2014 The Fourth Revolution argues infosphere monism—humans and AI are both inforgs within informational ontology.⁴ If PRF’s plurality component adopts informational ontology, the co-presence of multiple who can accommodate all inforgs (including AI agents), and the four-component conjunction of PRF is replaced by a single infosphere monism ontology.

Response: The ontological bearing of the PRF plurality component is Arendt’s phenomenological ontology (natality + mortality + worldliness), not informational ontology. Floridi’s infosphere monism ontological shift can be independently debated on philosophical grounds, but PRF explicitly adopts Arendt’s ontological position; § 11 states at the honesty boundary that this ontology choice is not logically necessary.

Category distinction (descriptive vs. normative): Floridi’s infosphere monism primarily addresses “descriptive ontology” (the informational nature of humans and AI), while PRF’s “normative floor” belongs to a different category. Infosphere monism does not provide equivalent alternative bearing for the validity, contestation, and agonism components; informational ontology does not generate the concrete bearing of the three normative subjects—“legitimacy-claim bearer,” “active dissenter,” and “legitimate adversary.” The PRF framework under the Floridi stress test is “weakened but not overturned”: the plurality component’s ontological choice bears philosophical pressure, while the remaining three components are unaffected. Likelihood is assessed as medium; impact is assessed as medium.

§ 9.3 Posthumanism—Braidotti / Hayles (medium-low impact)

Counter-argument: Braidotti’s 2013 The Posthuman and Hayles’s 1999 How We Became Posthuman argue for expanding the boundaries of personhood to encompass non-human animals, AI, nature, and cyborgs.⁵⁶ If the personhood boundary is expanded, PRF plurality’s co-presence of multiple who must be redefined; all four anchor authors of PRF presuppose “human subjects,” and posthumanist expansion requires the PRF framework to be remade at the ontological layer. Braidotti’s 2019 Posthuman Knowledge further extends posthumanism to a critique of Eurocentrism.⁷

Response: PRF adopts a “subjectivity-conditions” position: the personhood presupposition of the four anchor authors is “subjects possessing five conditions—natality + mortality + first-personal disclosure + active stance + legitimate adversary”; a “human exceptionalism” position is not within PRF’s scope of bearing. If posthumanist expanded subjects (e.g. certain conscious animals, certain strong AI, certain cyborgs) can pass the five-conditions test, the scope of PRF’s subjecthood bearing may expand; but the specific conditions for passing the test still depend on the four-component bearing.

Haraway’s 1985 “A Cyborg Manifesto” and Wolfe’s 2010 What Is Posthumanism? provide the genealogy of posthumanism.⁸⁹ This article states at the § 11 honesty boundary: the remade PRF under expanded personhood boundaries is an open question; this article adopts the personhood presupposition of the four anchor authors as the floor anchor and leaves space for future expansion. Likelihood is assessed as medium; impact is assessed as medium-low.

§ 9.4 Digital-Democracy Optimism—Benkler (low impact)

Counter-argument: Benkler’s 2006 The Wealth of Networks argues that the networked public sphere replaces the traditional public realm through decentralised production.¹⁰ Sunstein’s early Republic.com (2001) also argued that digital media offers new democratic possibilities.¹¹ If the digital public sphere can independently bear legitimacy, the four-component conjunctive floor of PRF is replaced by “new affordances bearing new legitimacy.”

Response: This article adopts the position that “new affordances reshape entry conditions; the old ontological foundation is unchanged.” Benkler’s affordance analysis of the networked public sphere (peer production, cooperative news ecosystems) is compatible with the PRF framework; the networked public sphere still must satisfy the bearing of PRF’s four components, or it degrades into one of four failures: algorithmically produced consensus (violating agonism); opinion without active-stance bearers (violating contestation); strategic interaction replacing communicative action (violating validity); single bearer dominance (violating plurality).

Benkler’s 2018 Network Propaganda has partially revised the 2006 optimistic forecast, acknowledging the impact of platform manipulation and polarisation on the networked public sphere.¹² Sunstein’s 2017 #Republic revised edition acknowledges the empirical confirmation of filter bubbles and echo chambers, weakening the early optimism.¹¹ Tufekci’s 2017 Twitter and Tear Gas provides a balanced analysis of the advantages and disadvantages of networked protest.¹³ This article does not degrade under the digital-democracy optimist stress test; new affordances belong to the category of “entry conditions,” not the bearing of “legitimacy foundations.” Likelihood is assessed as low; impact is assessed as low.

§ 9.5 Meta-Objection: Why These Four Thinkers (high likelihood / low-medium impact)

Counter-argument: Why Arendt / Habermas / Pettit / Mouffe, and not Rawls / Sen / Nussbaum / Honneth / later Habermasian Diskursethik, etc.? If Rawlsian justice, Sen’s capability, Nussbaum’s dignity, or Honneth’s recognition were added to the conjunction or substituted for one of the four, would PRF be weakened or strengthened?

Response: This article adopts an open-floor position: PRF is a composite floor, not an absolutely unique one; other approaches (Rawlsian justice, Sen’s capability, Nussbaum’s dignity, Honneth’s recognition) may serve as ceiling or boundary conditions for PRF.^14151617 The selection criterion for the four has three items.

First, the four share the theme of “public realm entry conditions.” Rawlsian justice centres on “distributive justice”; Sen’s capability on “individual capacities”; Nussbaum’s dignity on “human dignity”; Honneth’s recognition on “mutual recognition”—their themes correspond more directly to civic-proof series’ “civic proof can be legitimately borne.”

Second, the four are most deeply implicitly used in the first 18 articles of the civic-proof series. F1 §4’s three-approach conjunction has already established the bearing of Arendt + Habermas + Pettit; A1, A8, A14, A15, F2, and F3 each implicitly use at least one of the three; Mouffe’s agonism was introduced in F1 §4.2(iii) as an auxiliary. Other approaches (Rawls, Sen, Nussbaum, Honneth) are implicitly used to a shallower degree in the series’ first 18 articles.

Third, the four have the most concrete impact in digital scenarios. Arendt’s natality and LLM’s “repetition of the existing”; Habermas’s Wahrhaftigkeit and deepfakes; Pettit’s contestation and algorithmic governance; Mouffe’s agonism and polarisation—all four have specific mechanisms in digital scenarios.

The meta-objection’s likelihood is assessed as high (a reasonable scholarly critique); impact is assessed as low-medium (challenge diminishes once the open boundary is included). This article states at the § 11 honesty boundary: the four are floor anchors; there is open space for future addition of a fifth component (e.g. Honneth’s recognition, Sen’s capability).

§ 9.6 Likelihood × Impact Matrix

CF	Likelihood	Impact (without mitigation)	Impact (with mitigation)	Weakened primary component	Mitigation design
CF1 Coeckelbergh relational personhood	medium-high	medium-high	low-medium	plurality	F1 §4.1 “one step down, not two”; ontological vs. functional distinction
CF2 Floridi infosphere monism	medium	medium	low	plurality (ontological)	Category distinction (descriptive vs. normative); PRF need not deny the infosphere
CF3 Posthumanism Braidotti + Hayles	medium	medium-low	low	plurality + contestation	PRF adopts “subjectivity conditions,” not “human exceptionalism”
CF4 Digital-democracy optimism Benkler	low	low	low	validity + agonism	New affordance vs. old existential basis distinction
CF5 Meta-objection (why four thinkers)	high	low-medium	low	overall methodology	Open-floor position; other approaches as ceiling or boundary

Key matrix observation: with mitigations, the total degree of weakening across the five counter-argument types decreases significantly; PRF’s overall floor function is maintained through the redundancy of its four-component conjunction. The strongest impact is CF1 Coeckelbergh’s weakening of the plurality component, and F1 §4.1 already provides a response. The core proposition of “weakened but not overturned” is: none of the five counter-argument types can dismantle PRF’s four-component conjunctive structure; the plurality and contestation components are structurally weakened under CF1 and CF3 (each weakened by ≤ 30%), while the validity and agonism components remain structurally intact across all five counter-argument types.

§ 9.7 Special Section for Taiwanese Readers

Taiwan’s public realm plurality is constituted by five major bearers: streets, the legislature, online platforms (PTT, Facebook, Threads, Discord, Matters), referenda, and social movements (Referendum Alliance, Taiwan Youth Association for Democracy, Taiwan Labour Front, Environmental Protection Alliance, Citizen of the Earth Foundation); the 2014 Sunflower Movement of 18 March used g0v, PTT, and live-streaming from the legislature hall to bear the concrete Taiwanese practice of plurality and validity.¹⁸¹⁹²⁰ Multiple referendum cases from 2018 to 2021 (East Olympics naming, death penalty abolition, energy transition, civic voting age 18) bear the institutional fulfilment of validity; the 2024 presidential election’s deepfake videos and information manipulation constitute a structural challenge to validity.²¹ Contestation is constituted by multiple tracks: constitutional litigation, legislative questioning, media supervision, and civil-society complaints.

The agonism dimension in Taiwan is concretely demonstrated in the multi-party operation under the three-party minority configuration after the 2024 general election (Democratic Progressive Party, Kuomintang, Taiwan People’s Party).¹⁸¹⁹ From May 2024 to 2025, the legislature saw a dynamic structure of “Blue-White alignment” (KMT and TPP joint operations) and “Green coalition-building” (DPP seeking support from some TPP legislators or third-force members), bearing the Taiwanese institutional fulfilment of Mouffe’s three legitimate adversary conditions (mutual recognition, shared rules, agonistic frame) (see § 7.4 Case One). During the same period, the structural pressure of polarisation mechanisms distorting into antagonism (filter-bubble algorithms, social-media dunking culture, the shrinkage of cross-bubble dialogue space) means the boundary slippage between agonism and antagonism is a concrete vulnerability of contemporary Taiwanese democracy.

The concrete impact of TW DIW entering the LLM-agent phase, exemplified by F3’s supporter UI, is as follows: if the 2026–2027 phase introduces LLM-agent automated handling of wallet operations and the substitution fails (prompt injection, capacity-aware failure), the engineering bearing of the contestation component in digital scenarios is weakened, and the ability of older persons with dementia, persons with cognitive impairments, and digitally disadvantaged groups to initiate contestation is structurally weakened.²²²³ TW DIW specifications must explicitly state at the honesty boundary that “LLM-agent substitution does not replace the engineering bearing of supporter UI three-layer separation,” a position consistent with F1’s RT-ℬ ✗ and AA-ℬ ✗ boundary conclusions.

---

§ 10. Integration with Eight Series Articles’ Forward-Links

This section responds to the normative claims of A1 / A3 / A8 / A14 / A15 / F1 / F2 / F3 and bears the series-closing significance.

§ 10.1 PRF Response to A1 (Accountability Does Not Presuppose Real-Name Identification)

A1’s dual-anchor claim (procedural due process and republican contestability jointly bear accountability; anonymity does not weaken accountability) corresponds under the PRF framework to the core bearing of the plurality and agonism components. A1 §2 cites three cases including NAACP v. Alabama, linking the “protection of anonymous political speech” to the public-realm action rights of multiple who; this bearing is the specific application of the plurality component.²⁴ A1’s republican contestability presupposes the legitimate adversary structure; anonymity allows legitimate adversaries to exercise political expression rights without identity binding; this bearing is the specific application of the agonism component. This article’s § 3.3 matrix marks A1 × plurality (●) and A1 × agonism (●), consistent with A1’s dual-anchor claim.

§ 10.2 PRF Response to A3 (Conceptual Positioning of Civic Proof)

A3’s Nissenbaum CI plurality presupposition and civic proof five-layer abstraction table correspond under the PRF framework to the core bearing of the plurality and validity components. A3 §3.1’s “no privacy ‘in general’” bears the multiplicity of plurality at the information-norm layer;²⁵ A3 §6’s three minimum necessary conditions bear the institutional fulfilment of Habermas’s Richtigkeit claim. This article’s § 3.3 matrix marks A3 × plurality (●) and A3 × validity (●). A3’s bearing of the contestation and agonism components is weaker, reflecting that A3’s normative claims are concentrated on the foundations of privacy philosophy and conceptual landscape differentiation.

§ 10.3 PRF Response to A8 (DNS vs. Identity Trust Roots)

A8’s FTLA four-layer governance framework and HM conjunctive claim correspond under the PRF framework to the core bearing of the validity, contestation, and agonism components. A8’s G_state layer corresponds to the institutional counterpart of Habermas’s Sluice model; the G_recognition and G_oversight layers bear the four institutional conditions of Pettit’s editorial democracy; the tension between G_industry and G_state bears Mouffe’s legitimate adversary structure.²⁶ This article’s § 3.3 matrix marks A8 × validity (●), A8 × contestation (●), and A8 × agonism (●). A8’s partial bearing of plurality (○) reflects that its plurality dimension is mainly in the multi-stakeholder governance structure, not in the Arendtian ontological sense.

§ 10.4 PRF Response to A14 (Cross-Jurisdictional Redress Gaps)

A14’s cross-jurisdictional redress gaps and conflict-of-law over-determined structure correspond under the PRF framework to the core bearing of the contestation and agonism components. A14 §3’s four categories of gaps (verifier refusal, issuer revocation errors, vendor failure, cross-border privacy breach) bear the institutional condition failure of contestation; A14 §4’s conflict-of-law structure bears agonism’s “adversarial framework between legitimate adversaries.”²⁷ This article’s § 3.3 matrix marks A14 × contestation (●) and A14 × agonism (●).

§ 10.5 PRF Response to A15 (Civic-Proof Inclusion Rights)

A15’s precursor-right three-layer protection structure and wallet triple default degradation analysis correspond under the PRF framework to the core bearing of the plurality, validity, and contestation components. A15 §3’s empirical exclusion inventory bears the enrollment condition test of plurality; A15 §4’s Marshall three-tier transversality and ICCPR Art 16 bear the institutional conditions of validity; A15 §6.2’s wallet triple default degradation analysis bears the bearer position failure of contestation.²⁸²⁹ This article’s § 3.3 matrix marks A15 × plurality (●), A15 × validity (●), and A15 × contestation (●).

§ 10.6 PRF Response to F1 (Civic AI Agent Delegation Limits)

F1’s 5×3 matrix and three-approach conjunctive coverage correspond under the PRF framework to full core bearing of all four components (see § 8.4). F1 is the first article in the series to fully bear all four PRF components at the core level; this article’s PRF framework takes F1’s existing bearing as anchor and extends it to the bearer relations of all 18 preceding articles. F1’s RT-ℬ ✗ and AA-ℬ ✗ Z₃-intrinsic boundaries are formalised in this article’s theorem T_PRF5 as the “PRF floor engineering unreachability lemma at the existing engineering design layer” (confined to the engineering design layer; making no claim as a formal theorem at the philosophical layer).³³⁰

§ 10.7 PRF Response to F2 (Civic Receipts Provenance)

F2’s V_receipt C1–C6 and theorem T2 Z₃-intrinsic unreachability correspond under the PRF framework to the core bearing of the validity and contestation components, and provide partial bearing of the agonism component through the five counterfactual stress tests in §9 CF1–CF5. F2’s audit trail bears Habermas’s Wahrheit and Wahrhaftigkeit claims; F2’s ex post challenge structure bears Pettit’s editorial democracy’s “contestability.”³¹ F2 §9’s CF1–CF5 stress tests internalise Mouffe’s three legitimate adversary conditions into the robustness criterion of cryptographic design through likelihood-by-mechanism inference (see the elevation argument in § 8.3). This article’s § 3.3 matrix marks F2 × validity (●), F2 × contestation (●), and F2 × agonism (○, elevated). F2’s partial bearing of plurality (○) reflects its holder-side bearing’s indirect support for plurality.

§ 10.8 PRF Response to F3 (Selective Disclosure UX Failure)

F3’s V_ux C7–C10 and supporter UI three-layer separation correspond under the PRF framework to the core bearing of the validity and agonism components. F3’s informed consent bears the UX fulfilment of Habermas’s Wahrhaftigkeit claim; F3 §7’s supporter UI three-layer separation bears Mouffe’s legitimate adversary structure (the supporter as the position of a non-substitutable but necessary legitimate adversary).³²³³ This article’s § 3.3 matrix marks F3 × validity (●) and F3 × agonism (●). F3’s partial bearing of plurality and contestation (○) reflects that F3’s main axis is in UX cognitive engineering.

§ 10.9 Series-Closing Significance

The civic-proof series, starting from A1, gradually built a layered architecture of normative conditions and engineering bearing across 18 articles; this article (A2, Article 19) is the final article of the series, bearing the affirmative account of the political-philosophical foundations implicitly relied upon by the preceding 18 articles. The series-closing significance has three items.

First, bringing the implicit uses of the preceding 18 articles to affirmative articulation makes the series’ normative claims externally examinable. The PRF four-component conjunctive floor provides a unified formal skeleton for the series’ normative claims; the 32-cell bearer matrix makes the component correspondence of each article’s normative claims explicit.

Second, bearing the political-philosophical foundations of the series methodology. The formal skeletons of the series’ engineering articles (F1, F2, F3)—the 5×3 matrix, V_receipt, V_ux—are unified under this article’s PRF framework as the bearing of the “four-component conjunctive floor”; the bearing structures of the series’ normative articles (A1, A3, A8, A14, A15)—the dual-anchor, FTLA, precursor right—correspond to the specific applications of the four components under this article’s PRF framework.

Third, leaving space for subsequent series extension. The 27 open questions in § 11 list topics that the series may extend to subsequently, including PRF’s descriptive power for non-democratic regimes, reinforcement from other political-philosophical approaches, posthumanist remaking of PRF, and Taiwan-specific empirical work. Although the series closes at 19 articles, the extended applications of the PRF framework are an open question.

---

§ 11. Honesty Boundaries and 27 Open Questions

§ 11.1 Honesty Boundaries

This article explicitly states the following seven honesty boundaries.

First, cross-temporal inference. The four anchor authors’ original works belong to the Western political philosophy tradition from 1958 to 2013; this article’s application to digital scenarios in 2026 depends on the cross-temporal applicability of the four. The applicability of Arendt’s 1958 The Human Condition plurality concept to LLM-agent scenarios, Habermas’s 1962–1992 Öffentlichkeit concept to platform scenarios, Pettit’s 1997–2012 contestation concept to algorithmic governance scenarios, and Mouffe’s 2000–2013 agonism concept to polarisation scenarios are all authorial expansive interpretations.

Second, cross-cultural inference. The four anchor authors’ original works belong to the Western political philosophy tradition; this article’s applicability to Chinese/Taiwanese public realm scenarios depends on structural homology between the Chinese qun (群), gong (公), jian (諫), and zheng (諍) traditions and the four anchors. This article adopts the conservative position of “structurally homologous but with different historical bearer conditions”; the gap in historical bearer conditions is explicitly stated in §§ 4.3, 5.4, 6.4, and 7.3.

Third, PRF is a floor, not an absolutely unique standard. PRF is a composite floor of democratic legitimacy; it does not claim to be the unique floor. Rawlsian justice, Sen’s capability, Nussbaum’s dignity, Honneth’s recognition, and other approaches may serve as ceiling or boundary conditions for PRF.^14151617

Fourth, the Arendtian ontological position is a choice. The PRF plurality component adopts Arendt’s phenomenological ontology; the rejection of Floridi’s informational ontology and posthumanist ontology is a philosophical position choice (the claim of logical necessity is not within this article’s scope of bearing).

Fifth, the basis for θ_dem calibration. θ_dem ≈ 0.5 is an analytical recommendation with no empirical source. This article marks it as “awaiting inverse calibration against the V-Dem 2024 Liberal Democracy Index degradation case distribution, Bertelsmann Transformation Index, and Freedom House Annual Report.”

Sixth, the dependency of the 32-cell bearer matrix. The bearer determinations of the 32 cells depend on the current interpretation of the normative claims of the preceding 18 articles; if interpretations change, the matrix must be remade.

Seventh, the boundary of Schmitt citation. Mouffe’s agonism absorbs Schmitt’s 1932 concept of antagonist politics but rejects Schmitt’s friend/enemy annihilation logic.³⁴ Schmitt was a Nazi legal theorist; this article’s citation of Schmitt is confined to a genealogical acknowledgment within Mouffe’s agonism and does not extend to Schmitt’s Nazi political positions.

§ 11.2 Twenty-Seven Open Questions

Formal-skeleton layer:

• O1 (θ_dem calibration pathway): the analytical recommendation of θ_dem ≈ 0.5 awaits inverse calibration by cross-national democratic legitimacy empirical research.
• O2 (dynamic updating mechanism for the 32-cell bearer matrix): the matrix must be remade when interpretations of series articles are updated.
• O3 (criteria for PRF’s open boundary toward other political-philosophical approaches): the criteria for adding Rawls, Sen, Nussbaum, or Honneth to the conjunction are not yet specified.
• O4 (distinction between PRF’s descriptive and normative power for non-democratic regimes): does PRF presuppose democratic premises? What is its applicability to China, Russia, or Iran?

Arendt plurality layer:

• O5 (the relationship between posthumanism and plurality): does PRF need to be remade when personhood boundaries expand to non-human animals, AI, and nature?
• O6 (the application of the “action vs. work” distinction in plurality to digital scenarios).
• O7 (more precise formalisation of the opposition between natality and LLM-agent’s “new beginning”).
• O8 (broader academic community examination of the correspondence between the Chinese qun tradition and plurality).

Habermas Öffentlichkeit layer:

• O9 (the conjunctive relation between counterpublics and PRF’s mainstream Öffentlichkeit).
• O10 (engineering bearing of Öffentlichkeit in algorithmic media / platform scenarios).
• O11 (correspondence and tension of cross-cultural Öffentlichkeit, including Chinese gong, Islamic shura, African ubuntu).
• O12 (engineering of ideal speech situation in wallet/agent scenarios, interfacing with F3 cognitive load).

Pettit contestation layer:

• O13 (bearing of bearer position in cross-border institutional litigation, extension of A14 and A8).
• O14 (extension of Pettit’s republican democracy to non-democratic regimes).
• O15 (application of the “active stance, not passive input” distinction to digital voice and hashtag-activism scenarios).
• O16 (historical bearer-condition comparison between the Chinese jianyi tradition and Pettit’s republican contestation).

Mouffe agonistic layer:

• O17 (detection and protection mechanisms for boundary slippage between agonism and antagonism).
• O18 (the possibility of AI agents as “legitimate adversaries” in agonistic scenarios, and dialogue with F1’s RT-ℬ ✗ boundary).
• O19 (cross-issue agonism applications: climate, gender, cross-strait, AI regulation, etc.).
• O20 (extension of Norval’s 2007 Mouffe + Cavell dialogue, including how virtue ethics may reinforce agonism).

Counter-argument stress test layer:

• O21 (reinforcement or challenges to PRF from other political-philosophical approaches).
• O22 (reinforcement or challenges to PRF from non-Western democratic legitimacy frameworks, including Confucian gong, Islamic shura, African ubuntu).
• O23 (subsequent extension of the “one step down, not two” position on AI agents in PRF’s four components).
• O24 (specific conditions for extending PRF boundaries under a limited AI personhood framework).

Systemic and cross-level:

• O25 (dynamic maintenance mechanism for the PRF and civic-proof series bearer matrix).
• O26 (extended applications of PRF in subsequent articles F4+).
• O27 (precise distinction between “normative floor” and “engineering ceiling” in PRF—i.e. which PRF claims can be engineered and which cannot).

§ 11.3 High-Risk Claims List

This article lists ten high-risk claims for fact-checking and subsequent academic community examination.

R1: The names and contents of PRF’s four components as “conjunctive coverage extracted from the four anchor authors’ original works” are not claims explicitly asserted conjunctively by the four authors themselves.

R2: The sigmoid form of LegitimacyDegrade adopts the same form as F1 §3.2; the four authors’ equal weighting β = 1.0 is an a priori setting with no empirical source.

R3: θ_dem ≈ 0.5 is an analytical recommendation, corresponding to the midpoint of F1’s two thresholds θ₁ = 0.2 and θ₂ = 0.7, with no independent empirical source.

R4: The expansion of Mouffe’s agonism to an independent fourth component relative to F1 §4.2(iii)‘s “supplement does not weaken the conclusion” is an authorial claim extension; Mouffe’s own original works do not explicitly state that “agonism is an independently necessary condition for democratic legitimacy.”

R5: The elevation of F2–agonism to ○ (critical observation in § 5.2) is an authorial expansive interpretation; the boundary between the agonistic engineering-bearing intensity of F2 §9’s CF1–CF5 stress tests and ● core-bearing must be examined by the academic community. The A14 × plurality (—) non-correspondence determination may also be criticised as “too strict.”

R6: The proofs of theorems T_PRF1–T_PRF4 adopt a reductio structure of “otherwise degrades to X” and are not complete proofs in formal logic.

R7: F1’s three-approach conjunction as the series precursor to PRF’s four-approach conjunction; whether this article’s expansion is inconsistent with F1’s existing bearing, given that F1 §3.1 matrix does not list Mouffe in the ℬ components, requires academic community verification.

R8: The argument for some ○ cells in the 32 cells (e.g. A1 × validity ○, F2 × plurality ○, F3 × plurality ○, F2 × agonism ○ elevated) is relatively weak and may warrant downgrading to —.

R9: The position that “PRF is a normative floor, not an absolute standard” is explicitly stated at multiple points; fact-checking must verify that the final draft does not slide into language treating “PRF as an absolute standard.”

R10: The publication status of the Stanford Internet Observatory Final Report 2024-12 must be verified; SIO experienced structural pressures in 2024, and the specific URL and publication date of the final report must be confirmed.

---

§ 12. Conditional Academic Conclusions

This article reaches the following conclusions under three conditions.

Conclusion One: The conjunction of PRF’s four components constitutes the normative floor of the civic-proof series. Arendt’s plurality (ontological co-presence of multiple who), Habermas’s Öffentlichkeit (normative structure of kommunikatives Handeln and three Geltungsansprüche), Pettit’s contestation (active-stance bearer and four editorial democracy conditions), and Mouffe’s agonism (three legitimate adversary conditions) together constitute a conjunctive floor in civic-proof series scenarios. Each of the four is individually necessary, irreducible, and cannot be substituted by the other three (borne by theorems T_PRF1–T_PRF4); any component being violated means LegitimacyDegrade(d) ≥ θ_dem ≈ 0.5 (borne by theorem T_PRF0). This conclusion is conditional; the conditions include the seven items explicitly stated at the § 11 honesty boundary (cross-temporal inference, cross-cultural inference, PRF as floor not absolutely unique, Arendt ontological position choice, θ_dem calibration basis, 32-cell matrix dependency, boundary of Schmitt citation).

Conclusion Two: Engineering cannot replace the normative floor. Cryptographic primitives (wallet, receipts, selective disclosure), UX design (supporter UI three-layer separation), and platform governance (FTLA four-layer, cross-border mutual recognition) and other engineering bearings can bear the concrete implementation of PRF’s four components but cannot replace the normative status of the four components. F1’s RT-ℬ ✗ and AA-ℬ ✗ Z₃-intrinsic boundaries (responsibility-bearing requires first-personal mens rea; contestation bearer requires active stance) are formalised in this article’s theorem T_PRF5 as “the PRF floor engineering unreachability lemma at the existing engineering design layer”—no engineering design can bypass the PRF floor. The implication of this conclusion is: normative requirements for infrastructure such as EUDI Wallet, TW DIW, and Aadhaar should take PRF’s four components as the floor criterion; engineering bearing is the concrete implementation of the floor; engineering bearing is not claimed to independently bear legitimacy.

Conclusion Three: The series’ 8-article bearer matrix is a “conditional correspondence,” not an “absolute description.” The 32-cell bearer matrix (§ 3.3) reflects the current interpretation of the normative claims of the preceding 18 articles; if interpretations change, the matrix must be remade. The determination that F1 is the bearer of all four components at the core level (§ 8.4) depends on the expansive interpretation of F1 §4’s three-approach conjunction, elevating F1 §4.2(iii)‘s “Mouffe supplement does not weaken the conclusion” to the bearing of the agonism component. This article adopts this expansive interpretation as the author’s position; under other interpretations (e.g. still treating Mouffe as auxiliary), F1 remains three-component core-bearing plus one-component partial-bearing. The implication of this conclusion: the 32-cell matrix is a “conditional correspondence” of the bearer relations of the series’ normative claims, not an “absolute description”; the dynamic updating mechanism of the matrix is the open question O2 of § 11.

§ 12.1 Priority Classification of the 27 Open Questions

The 27 open questions listed in § 11.2 can be divided into four categories by subsequent bearing pathway, as a research direction recommendation after the series closes.

Category One: F4+ subsequent series articles bear (belonging to the precise distinction between “normative floor vs. engineering ceiling,” which this series may bear if extended). Includes O10 (engineering bearing of Öffentlichkeit in algorithmic media/platform scenarios), O12 (engineering of ideal speech situation in wallet/agent scenarios, interfacing with F3 cognitive load), O15 (application of the “active stance” distinction to digital voice and hashtag-activism scenarios), O18 (the possibility of AI agents as “legitimate adversaries” in agonistic scenarios), O25 (dynamic maintenance mechanism for PRF and the series bearer matrix), O26 (extended applications of PRF in subsequent articles F4+), O27 (precise distinction between normative floor and engineering ceiling).

Category Two: Broader academic community examination (belonging to extended research in the political philosophy and democratic theory community). Includes O3 (open-boundary criteria for PRF toward other political-philosophical approaches), O5 (relationship between posthumanism and plurality), O8 (broader examination of the Chinese qun tradition and plurality), O11 (cross-cultural Öffentlichkeit correspondence and tension), O16 (historical bearer-condition comparison between the Chinese jianyi tradition and Pettit), O20 (extension of Norval’s Mouffe + Cavell dialogue), O21 (reinforcement or challenges to PRF from other political-philosophical approaches), O22 (reinforcement or challenges from non-Western democratic legitimacy frameworks), O23 (subsequent extension of the “one step down, not two” position), O24 (PRF boundaries under a limited AI personhood framework).

Category Three: Policy implementation and empirical research (belonging to digital identity policy, democratic legitimacy empirical research, and cross-national comparative research). Includes O1 (θ_dem calibration pathway), O2 (dynamic updating mechanism for the 32-cell matrix), O4 (distinction between PRF’s descriptive and normative power for non-democratic regimes), O14 (extension of Pettit’s republican democracy to non-democratic regimes), O19 (cross-issue agonism applications: climate, gender, cross-strait, AI regulation, etc.).

Category Four: Technology–philosophy intersection (belonging to LLM-agent, formal verification, capacity-aware UX, and other engineering bearing and political philosophy concept intersection research). Includes O6 (application of the “action vs. work” distinction in plurality to digital scenarios), O7 (more precise formalisation of the opposition between natality and LLM-agent’s “new beginning”), O9 (conjunctive relation between counterpublics and PRF’s mainstream Öffentlichkeit), O13 (bearing of bearer position in cross-border institutional litigation), O17 (detection and protection mechanisms for boundary slippage between agonism and antagonism).

The four categories are authorial suggestions, not exclusive; multiple open questions may be borne across categories simultaneously. For example, O18 (the possibility of AI agents as legitimate adversaries) simultaneously belongs to Category One (F4+ subsequent articles) and Category Two (broader academic community examination). The choice of subsequent research pathway differs according to the bearer’s research position (series author, academia, policy community, technical community).

§ 12.2 Differentiated Recommendations for Three Groups of Readers

For the political philosophy academic community: this article adopts a conjunctive (not synthetic) methodology, explicitly stated in § 8.2. The academic community’s interpretations of the original works of Arendt, Habermas, Pettit, and Mouffe differ; the conjunctive structure of this article preserves the tensions among the four authors’ original works without attempting to resolve them. Topics for subsequent academic extension include the conjunctive relation between PRF and other political-philosophical approaches (O3, O21), the correspondence and tensions of cross-cultural PRF (O11, O22), and the distinction between PRF’s descriptive and normative power for non-democratic regimes (O4, O14).

For the civic technology community: this article’s 32-cell bearer matrix formalises the bearer relations of the series’ normative claims; the civic technology community may invoke the matrix as a normative criterion for engineering bearing. F1’s RT-ℬ ✗ and AA-ℬ ✗ boundaries (i.e. the PRF floor engineering unreachability lemma) are the upper-bound constraints on engineering design; no engineering design can bear legitimacy claims outside these boundaries. Specific cases such as supporter UI three-layer separation, wallet triple default degradation analysis, and cross-jurisdictional redress gaps are applications of PRF in engineering bearing.

For digital identity policy scholars: this article provides a special section for Taiwanese readers in § 9.7, including predictions for the specific impact of TW DIW entering the LLM-agent phase. Policy recommendations: TW DIW specifications must explicitly state at the honesty boundary that “LLM-agent substitution helps some users but does not replace the engineering bearing of supporter UI three-layer separation; the human democratic subject bearing of the contestation dimension cannot be replaced by LLM-agent.” Policy assessment of infrastructure such as EUDI Wallet, Aadhaar, and TW DIW may invoke PRF’s four components as normative floor criteria.

This article is the nineteenth article of the civic-proof series, and the final article of the series. The extended applications of the PRF framework, reinforcement from other political-philosophical approaches, posthumanist remaking of PRF, specific empirical work on Taiwan and Chinese public realms, cross-national calibration of θ_dem for democratic legitimacy, and other topics (the 27 open questions listed in § 11) are left as open space for subsequent research after the series closes.

---

Revision Note (2026-05-16)

In revision, this article clearly defines the PRF conjunctive floor framework as a normative floor rather than an absolutely unique standard; refines the strong normative formulation of “irreducible / cannot be substituted by the other three” to “individually necessary within the PRF conjunctive floor framework, difficult to be fully substituted by other components (not absolutely irreducible)”; refines “trade-offs not permitted” to “within this article’s PRF conjunctive setting, the trade-off reading of compensating a violated component with a reinforced one is not adopted; whether other normative approaches can provide partial reinforcement is listed as a ceiling or boundary condition”; confines the “PRF floor engineering unreachability lemma” (T_PRF5) to the “existing engineering design layer’s PRF floor engineering unreachability lemma” (explicitly stated as a formal theorem at the engineering design layer, not the philosophical layer); adds inline caveat to θ_dem ≈ 0.5 in T_PRF0 and the description: “analytical recommendation, not empirically calibrated.” The revision basis is the civic-proof series Phase 1+2 audit + two rounds of GPT-5.5-pro audit (2026-05-16), examining risk targets R_CL “composite floor vs. absolute standard” and R_NM “normative vs. descriptive separation.” Revised locations include: frontmatter description, § 2.3 conjunctive methodology section, § 3.2 conjunctive floor implication section, § 3.4 T_PRF0 title, § 3.5 T_PRF5 title, § 7 chapter introduction, § 8.4 F1 bearing section. The core argumentative skeleton (PRF four-component definitions / 32-cell bearer matrix / four formal theorems / five counter-argument stress test categories / Chinese indigenisation bearing / special section for Taiwanese readers) is unchanged.

---

References

A. Four Anchor Authors’ Original Works

B. Secondary Interpretations and Classic Foundations

C. Counter-Arguments and Alternative Approaches

D. Contemporary Public Sphere Research

E. Chinese / Taiwanese Indigenisation Sources

F. Engineering Bearing: Established Facts and Infrastructure

G. Other Normative Approaches (PRF Open Boundary)

H. Democratic Legitimacy Empirical Monitoring

I. Case Law and Enforcement

J. Civic-Proof Series Internal Forward-Links

Footnotes

1. Coeckelbergh, M. (2020). AI Ethics. MIT Press. Ch. 4–6 on relational personhood. Source grade A. ↩

2. Coeckelbergh, M. (2024). “Why AI Undermines Democracy and What to Do About It.” Polity 56(4) (including ethics vs. politics two-domain position distinction). Source grade A. ↩

3. Civic-proof series Article 16 (F1) “Civic AI Agent Delegation Limits” §4 three-approach conjunctive coverage. Published on blog-pro. Source grade A (series internal forward-link). ↩ ↩²

4. Floridi, L. (2014). The Fourth Revolution: How the Infosphere is Reshaping Human Reality. Oxford University Press. Source grade A. ↩

5. Braidotti, R. (2013). The Posthuman. Polity Press. Source grade A. ↩

6. Hayles, N. K. (1999). How We Became Posthuman: Virtual Bodies in Cybernetics, Literature, and Informatics. University of Chicago Press. Source grade A. ↩

7. Braidotti, R. (2019). Posthuman Knowledge. Polity Press. Source grade A. ↩

8. Haraway, D. (1985/1991). “A Cyborg Manifesto: Science, Technology, and Socialist-Feminism in the Late Twentieth Century.” In Simians, Cyborgs and Women. Routledge. Source grade A. ↩

9. Wolfe, C. (2010). What Is Posthumanism? University of Minnesota Press. Source grade A. ↩

10. Benkler, Y. (2006). The Wealth of Networks: How Social Production Transforms Markets and Freedom. Yale University Press. Source grade A. ↩

11. Sunstein, C. R. (2017). #Republic: Divided Democracy in the Age of Social Media. Princeton University Press. Source grade A. ↩ ↩²

12. Benkler, Y., Faris, R., & Roberts, H. (2018). Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics. Oxford University Press. Source grade A. ↩

13. Tufekci, Z. (2017). Twitter and Tear Gas: The Power and Fragility of Networked Protest. Yale University Press. Source grade A. ↩

14. Rawls, J. (1971). A Theory of Justice. Harvard University Press. Source grade A. ↩ ↩²

15. Sen, A. (2009). The Idea of Justice. Belknap Press of Harvard University Press. Source grade A. ↩ ↩²

16. Nussbaum, M. C. (2011). Creating Capabilities: The Human Development Approach. Harvard University Press. Source grade A. ↩ ↩²

17. Honneth, A. (2014). Freedom’s Right: The Social Foundations of Democratic Life. Columbia University Press. Source grade A. ↩ ↩²

18. Zhu, Yun-han (2012). High in the Clouds: A Scholar’s Reflections on the Twenty-First Century. Taipei: Commonwealth Publishing. Source grade B. ↩ ↩²

19. Lin, Tsung-hong (2020). Floating Dust on the Island: Social Change in Taiwan. Taiwanese Sociological Association. Source grade B. ↩ ↩²

20. Fan, Yun (2024). Two decades of feminist movement and Taiwanese civil society observations (specific article citation to be confirmed in fact-checking). Source grade B. ↩

21. National Communications Commission (NCC). 2024 presidential and legislative election deepfake audio-visual monitoring report; Taiwan FactCheck Centre, IORG (Information Operations Research Group), Doublethink Lab 2024 election deepfake case compilation. Source grade B. ↩

22. Ministry of Digital Affairs (MODA, Taiwan). TW DIW (Taiwan Digital Identity Wallet) 2025–2026 specification documents and trust list on-chain progress (public data). Source grade B. ↩

23. F1 (Article 16) §3.1 5×3 matrix and §4 three-approach conjunction—mechanism citation for predicting the impact of TW DIW’s LLM-agent phase. Series internal alignment. Source grade A. ↩

24. Bovens, M. (2007). “Analysing and Assessing Accountability: A Conceptual Framework.” European Law Journal 13(4): 447–468. Source grade A. ↩

25. Nissenbaum, H. (2010). Privacy in Context: Technology, Policy, and the Integrity of Social Life. Stanford University Press. Source grade A. ↩

26. Civic-proof series Article 8 (A8) “DNS vs. Identity Trust Roots” (FTLA four-layer governance framework). Published on blog-pro. Source grade A (series internal forward-link). ↩

27. Civic-proof series Article 14 (A14) “Cross-Jurisdictional Redress Gaps.” Published on blog-pro. Source grade A (series internal forward-link). ↩

28. Civic-proof series Article 15 (A15) “Civic-Proof Inclusion Rights.” Published on blog-pro. Source grade A (series internal forward-link). ↩

29. Marshall, T. H. (1950). Citizenship and Social Class and Other Essays. Cambridge University Press. Source grade A. ↩

30. Pitkin, H. F. (1967). The Concept of Representation. Berkeley: University of California Press. ISBN 978-0-520-02156-7. Ch. 5 p. 112–143, Ch. 10 p. 232–243. Source grade A. ↩

31. Civic-proof series Article 17 (F2) “Civic Receipts Provenance” §3 V_receipt C1–C6 and theorems T1–T4. Published on blog-pro. Source grade A (series internal forward-link). ↩

32. Civic-proof series Article 18 (F3) “Selective Disclosure UX Failure” §3 V_ux and §7 supporter UI three-layer separation. Published on blog-pro. Source grade A (series internal forward-link). ↩

33. United Nations. Convention on the Rights of Persons with Disabilities (CRPD), Article 29 and General Comment No. 1 (2014) on equal recognition before the law. Source grade A. ↩

34. Schmitt, C. (1932/1976). The Concept of the Political (G. Schwab, Trans.). Rutgers University Press. Source grade A (must be stated at § 11 honesty boundary: Mouffe absorbs Schmitt but rejects his Nazi political positions). ↩