制度应该多厚,又如何允许自己被凿
How Thick Should an Institution Be?
大制度不一定厚,小关系不一定薄
人们直觉上认为,国家需要复杂规则,朋友之间则无需制度。但一个大型兴趣社群可能只需几条守则,因为成员随时能无损退出;两个人的婚姻却可能需要精细处理财产、照护、隐私、表达与共同孩子,因为关系密度高,退出成本也高。
SAE 因此不按规模决定制度厚度。更重要的两个变量是:交集有多密,以及退出有多贵。关系越多、彼此影响越深,未处理余项越容易累积;离开越困难,越不能把“你不喜欢就走”当作纠错机制。
公司有员工、用户、股东、伙伴与管理者等多种交集,某些维度甚至比宪法更细。友谊规模极小,却可能因为长期托付而需要清楚的不可侵犯区域。制度厚度来自条件,不来自名称。
最小化不是越少越好
制度的最小化原则是:不需要建构的,不要建构,因为每一项建构都会产生新的余项。规则需要解释、执行和裁决,也可能被权力利用。用一条新制度处理旧问题时,我们同时创造了未来需要修正的新对象。
但最小化不等于一味删减。若退出代价高、权力不对称、关系密集,过薄的制度会把裁决交给强者。真正的最小是“足以保护主体且没有无理由增量”,不是条文数量最少。
好的制度能够说明自己为什么厚到这里:这项程序应对何种权力差;那项权利保护哪一种不可退出的关系;这项义务为何落在拥有更多权力的人身上。
制度也会退化
制度至少有三种典型退化。闭合:质疑和修正的通道被关闭,规则把自己宣布成不可追问。掏空:文本仍然漂亮,实际保护和制衡已经失效。过度建构:制度越过共同交集,开始规定个人应当相信什么、热爱什么、怎样生活。
三者看似不同,都把工具变成了目的。闭合拒绝时间,掏空借外壳掩盖权力,过度建构则进入本应由主体生成目的的涌现层。
因此制度必须包含自修凿机制:修正程序、独立审查、表达保障、外部反馈、定期复核以及真实退出。最关键的条件是,凿的通道不能完全由被凿对象控制。若管理者决定哪些批评可以上达,批评便只是管理的一部分;若行政权决定司法何时独立,独立就只剩名称。
声明余项,而非宣称完美
成熟制度不会说“这一构已经完成”。它会公开已知余项:哪些冲突尚无好答案,哪些条款可能制造新成本,哪些条件变化后必须重审。核心原则可以设置极高修改门槛,却不应伪装成没有余项的永恒构。
到这里,理解 SAE 所需的基础路径已经完整:系统应保护条件而退出目的形成;主体可能在制度、关系和内部遭到殖民;三层需要共同涵育;自由表现为可分叉、可形成自己的不得不;多个主体的余项又使制度成为必要。
后面两篇不再增加新的伦理义务。它们只是继续向下追问:主体性本身从哪里来,“非”是否有更底层的说明。即使不读,前十三篇仍然独立成立;如果愿意理论深潜,再继续往下走。
Large Institutions Are Not Always Thick
We assume that states require complex rules while friendship requires none. Yet a large voluntary community may need only a few procedures because members can leave without meaningful loss. A marriage between two people may require careful arrangements concerning property, care, privacy, expression, and children because relational density and exit costs are high.
SAE therefore does not derive institutional thickness from scale. Two variables matter more: how dense the intersections are and how expensive departure is. Greater density produces more recurring remainders. Higher exit cost makes “leave if you dislike it” an inadequate correction mechanism.
A corporation connects employees, users, shareholders, partners, and managers and may be institutionally thicker than a constitution in certain dimensions. A friendship is small, yet long entrustment can require clear inviolable zones. Conditions, not names, determine thickness.
Minimization Is Not Mere Reduction
SAE's minimization principle says: do not construct what need not be constructed, because every construct generates remainder. Rules require interpretation, enforcement, and judgment; they can also be captured. When a new rule handles an old problem, it creates a new object that future subjects may need to chisel.
Minimization is not the automatic removal of rules. Where exit is costly, power asymmetric, and relationships dense, a thin framework silently assigns judgment to the strong. The relevant minimum is sufficient protection without unjustified increments, not the smallest number of clauses.
A legitimate institution can explain why it is this thick. This procedure answers a particular power difference. That right protects someone within an intersection she cannot readily exit. This obligation belongs to the actor with greater power.
How Institutions Degenerate
Institutions exhibit at least three forms of degeneration. Closure shuts the channels of criticism and revision, presenting the rules as beyond question. Hollowing preserves an attractive text after actual protection and checks have ceased to function. Overconstruction carries institutional authority beyond the common intersection and begins to prescribe what persons should believe, love, and become.
All three turn a tool into an end. Closure refuses time, hollowing uses form to conceal power, and overconstruction enters the Emergence Layer in which purposes should be generated by subjects.
Institutions must therefore contain mechanisms for self-chiseling: amendment, independent review, protected expression, external feedback, periodic audit, and meaningful exit. The essential condition is independence. A channel of criticism controlled entirely by the object criticized is part of management, not a genuine opening. A judiciary independent only when the executive permits it is independent in name.
Declaring Remainders Instead of Perfection
A mature institution does not announce that the construct is complete. It declares known remainders: conflicts for which no adequate answer exists, costs generated by current provisions, and changes of condition that should trigger review. Core principles may receive demanding amendment thresholds without being misrepresented as a remainderless eternity.
At this point, the reader has the complete foundation required to use SAE. Systems secure conditions and withdraw from purpose formation. Colonization can occur institutionally, relationally, and internally. The three layers require cultivation together. Freedom appears as forkability and the growth of one's own cannot-nots. Remainders among multiple subjects make institutions necessary.
The final two essays add no new ethical obligation. They descend to a different question: where subjecthood itself comes from and whether Negativity admits a deeper account. The first thirteen essays stand without them. Readers who want an ontological deep dive may continue.