两度显现的不完备性 / Incompleteness Exhibited Twice

OpenAI 2026 年 10 月 26 日的公告展示了由内部模型生成的三项候选结果：任意维数下欧几里得波动方程局部平滑猜想的证明（由此解决挂谷猜想与斯坦因限制猜想）、否定指数时间假说的确定性亚指数算法，以及四维纯杨-米尔斯理论与若干相互作用量子场论的构造性存在证明。每一项结果都在不同程度上附有Lean 形式化文件与指定的外部审验流程。这些结果仍在接受独立审视；下文的剖析并不以它们最终成立与否为前提。这一事件让不完备性以互为镜像的方式两度显现：在直接意义上，它显现于这些形式化之中——它们把进一步的真命题与自身的一致性留在开放之处；在否定意义上，它显现于那种持守之中——原本无法封闭持守之物被暂时封闭持守，却在发布与接收中被当作已被永久封闭持守来对待。

The OpenAI announcements of 26 October 2026 presented three candidate results generated by an internal model: a proof of the local smoothing conjecture for the Euclidean wave equation in every dimension (thereby resolving the Kakeya and Stein restriction conjectures), a deterministic subexponential algorithm that refutes the Exponential Time Hypothesis, and a constructive existence proof for pure Yang–Mills theory and certain interacting quantum field theories in four dimensions. Each was accompanied, in varying degrees, by Lean formalizations and by designated processes of external examination. The results remain under independent scrutiny; their eventual correctness is not required for the analysis that follows. What the episode makes visible is incompleteness exhibited twice as mirror images: in its direct sense inside the formalizations, which leave further truths and their own consistency open, and in its negative sense in the hold, where what cannot be held closed is held closed temporarily yet treated in presentation and reception as if held closed permanently.

一、 绝对的不完备性与不确定性的确定 / 1. Absolute Incompleteness and the Certainty of Uncertainty

哥德尔不完备定理确立了：任何一致的、可有效公理化的、强到足以表达基本算术的形式系统，都会在自身内部留下不可推导的真命题，并且无法证明自身的一致性。这一界限并非相对于某套特定公理化体系而言，因而也不会被某个更大的系统补齐。任何此类扩充系统，只要能够表达先前系统，都会遇上属于它自身的不可推导性：它可以证明先前系统无法证明的命题，同时在自身内部留下新的不可证命题，正如没有系统能被保持封闭从一位寄存器一路追溯至算术所展示的那样。不完备性因而在层层扩充中是绝对的：形式系统的非绝对性，就在于不完备性自身。同样的关系也显现在不确定性的确定之中。确定性始于将确定从不确定中区分出来的那道初始切分；其他一切皆由此流出，并在逻辑上一再重申自身。由于那道切分是在特定语境下作出的，一切由之推导而出之物便都以该语境为前提，且仅适用于该语境。关于不确定性的确定因而可以毫无问题地一再重申；恰恰是当它声称“不确定性的确定”也像“确定的确定”那样无法确定时，它才把自己倒置了。确定性本身唯有在暗含某物已毫无不确定性时才无法成为确定，这等于把推导出的结果同为那道切分设定前提的语境割裂开来。不可推导性并非等待闭合的暂时缺口；它是任何形式化得以发生的条件，即开放即一致的形式面相。

Gödel’s incompleteness theorems establish that any consistent, effectively axiomatized formal system strong enough to express basic arithmetic leaves true statements underivable inside it and cannot prove its own consistency. This limit is not relative to a particular axiomatization that a larger system might complete. Any such extension capable of expressing the prior system meets an underivability of its own: it may prove what the prior system could not, and leaves new statements unprovable inside itself, as No System Can Be Kept Closed traces from the one-bit register up to arithmetic. Incompleteness is therefore absolute across extensions: the non-absoluteness of formal systems is located in incompleteness itself. The same relation appears in the certainty of uncertainty. Certainty initiates from the original cut of certainty from uncertainty, and from that cut everything else follows and reasserts itself logically. Because that cut is made under a specific context, everything derived from it is preconditioned by that context and applies only to that context. A certainty about uncertainty can therefore be reasserted without problem; it inverts itself precisely when it claims that the certainty of uncertainty cannot be certain, as the certainty of certainty cannot. Certainty itself cannot be certain when it implies something without uncertainty—severing what follows from the context that preconditioned the cut. The underivability is not a temporary gap awaiting closure; it is the condition under which any formalization occurs, the formal face of Openness Is Consistency.

不完备定理所确立的，不止是存在为真却不可证的命题。在一个强到足以表达算术的一致形式系统内部，若干命题是不可判定的：无论该命题本身还是其否定，都不可推导。哥德尔语句 G 在预期标准解释下成立——一旦在元理论中设定了该解释——却在系统内部不可推导（假定系统一致）；假定 ω-一致性，¬G 同样不可推导，而该语句的罗瑟变体仅凭普通一致性便是独立的。该语句因而是独立的。任何命题，只要其真与伪都超出系统内部所能调动的资源，便会承袭这一双重界限。一致性同时阻断了肯定方向的推导与否定方向的推导。

The incompleteness theorems establish more than the existence of true but unprovable statements. Inside a consistent formal system strong enough for arithmetic, certain statements are undecidable: neither the statement nor its negation is derivable. Gödel’s sentence G holds in the intended standard interpretation—once that interpretation is posited in the metatheory—yet remains underivable inside the system (assuming consistency); assuming ω-consistency, ¬G is underivable as well, and Rosser’s variant of the sentence is independent under plain consistency alone. The sentence is therefore independent. Any statement whose truth and falsehood both extend beyond the resources the system can internally mobilize inherits the same dual limit. Consistency blocks both the affirmative derivation and the negative one.

连同层层扩充中无尽的递退，这一双重性质使不完备性成为绝对的。非绝对性不仅在于无法证明若干真命题，而且同时在于无法证伪相应的独立命题。系统既无法通过将该命题坐实为真来完备自身，也无法通过将该命题确立为假来完备自身。这两种举措都会要求它去执行一致性所禁止之事。因此，开放余项在任何一个方向上都无法作为封闭之物被持守。试图将该命题肯定为已然持守，处于系统所及的范围之外；试图将其当作已被证伪而摒弃，同样处于这一范围之外。对于各类未决的算术语句，形式记录所确立的不会超出已就其推导出的内容：在形式独立性已被证明之处，它确立的是只要系统一致，两个方向便均不可推导；而对于 P 对 NP 这类开放问题——其中既未推导出 Σ₂ 语句 P = NP，也未推导出 Π₂ 语句 P ≠ NP，且其相对于 ZFC 的形式独立性本身亦未获证——它所确立的仅是两个方向均处于未证状态，正如论关于 P 与 NP 所能言说的边界所追溯的那样。在元理论中设定唯一的预期标准模型 ℕ——即称 G 为真所凭的那同一项设定——以坚持某种非语境的“是或否”真值，或在形式证明之外斥该问题为无意义，抑或在独立性尚未获证之处断言永久不可推导性，每一项都越过了形式装置所能核证的范围，由心智自行提供了一项形式外承诺。

Together with the regress across extensions, this dual character is what renders incompleteness absolute. Non-absoluteness is located not merely in the failure to prove certain truths, but in the simultaneous failure to refute the corresponding independent statements. The system cannot complete itself by securing the statement as true, and it cannot complete itself by establishing the statement as false. Both moves would require it to do what consistency forbids. The open residue therefore cannot be held closed in either direction. An attempt to affirm the statement as held remains outside the system’s reach; an attempt to dismiss it as refuted is equally outside that reach. Across undecided arithmetical sentences, the formal record establishes no more than what has been derived about them: where formal independence has been proved, it establishes that neither direction is derivable if the system is consistent; for an open question such as P versus NP, where neither the Σ₂ equality nor the Π₂ inequality has been derived and formal independence from ZFC itself remains unproved, it establishes only that both directions stand unproved, as On the Boundaries of What Can Be Said about P and NP traces. Positing a unique intended standard model ℕ in the metatheory—the same posit under which G is called true—to insist on a non-contextual yes-or-no truth value, dismissing the question as meaningless outside formal proof, or, where independence has not been proved, asserting permanent underivability each steps beyond what the formal apparatus certifies and supplies an extra-formal commitment of the Mind’s own.

这三项结果倘若经受住审验，便解决了若干悬而未决的具体问题。统一的局部平滑定理将了结限制理论的高维情形；针对固定宽度合取范式（CNF）的亚指数计数算法将否定指数时间假说（ETH），同时让 P 与 NP 问题原封未动；四维杨-米尔斯的存在证明将回应千禧年大奖难题的一个组成部分，而未解决质量间隙。在每一种情形中，问题的解决依然只是某个形式系统内部的一项特定推导。接受该推导的系统依然会留下进一步不可推导的命题，也依然无法证明自身的一致性。开放余项继续延伸，正如为何数学永远解不完在整个学科中所追溯的那样。这些结果的内容之中，没有任何东西会改变这一绝对界限。

The three results, if they survive examination, settle specific questions that had remained open. A uniform local-smoothing theorem would close higher-dimensional cases of restriction theory; a subexponential counting algorithm for fixed-width CNF would refute ETH while leaving P versus NP untouched; an existence proof for four-dimensional Yang–Mills would address one component of a Millennium problem without settling the mass gap. In each case the settlement remains a particular derivation inside some formal system. The system that accepts the derivation still leaves further statements underivable and cannot prove its own consistency. The open residue continues, as Why Mathematics Can Never Be Solved traces across the discipline. Nothing in the content of the results alters this absolute limit.

二、 把开放余项转化为可持守之物 / 2. Converting the Open Residue into Something to Hold

传播中可以辨认出的实际效应，是把那份余项转化为看似可永久持守之物。公告将这些产出包装为“重大科学发现”，并附上Lean 验证证书与具名把关学者。病毒式传播的接收过程在很大程度上把余留的审验当作结果得以永久封闭持守之前的一段延迟，而非当作不完备性本就要求的区分活动的延续。建立在 ETH 之上的条件性复杂度困难结果被登记为骤然可以改写；限制理论被登记为已补齐高维情形；若干场论的构造性地位被登记为已从开放变为确立。无论一项候选推导是证明某项猜想还是证伪某项假说，这一登记都会将同一份开放余项闭合：一项假说被登记为已被证伪，与一项猜想被登记为已被证明一样迅速。注意力的收窄，在审验尚未结束之前便已由发布的形式造成；同样造成这一收窄的，是对谁能交付那件闭合之物的定位——无论这份持守是分散在公开构件之中，还是局限于某个推迟发布的指定小组之内。

The practical effect identified in the circulation is the conversion of that residue into an item that appears permanently holdable. The announcements package the outputs as “major scientific discoveries” accompanied by Lean certificates and named stewards. The viral reception largely treats the residual examination as a delay before the result is held closed permanently, rather than as the continuation of distinguishing activity that incompleteness already requires. Conditional hardness results that rested on ETH are registered as suddenly available for revision; restriction theory is registered as having a completed higher-dimensional case; the constructive status of certain field theories is registered as having moved from open to established. Whether a candidate derivation proves a conjecture or refutes a hypothesis, the registration closes the same open residue: a hypothesis is registered as already refuted as quickly as a conjecture is registered as already proved. The narrowing of attention is produced by the form of the presentation before the examination has concluded. It is produced, equally, by the positioning of who is able to deliver the closed item—whether the holding is distributed across public artifacts or localized to a designated group that has deferred release.

这一转化不同于发现过程本身。生成候选形式化表述、推进 Lean 代码库开发以及随后的开放审验，依然是寻常的探究活动；在其中，某项推导或工作假设只是被暂时封闭持守，以便检验随之而来的推论并迫使人们重新审验先前的假设。这一转化并不发生在这些步骤之中；它发生在以公告传播速度流转的即时登记之中——把原本无法封闭持守、在探究中仅能暂时封闭持守之物，当作已被永久封闭持守来对待。永久闭合的印象并不取决于任何特定形式化最终是否正确；它是由包装与接收制造出来的。

flowchart TD
    subgraph S_Open["寻常的探究活动（作为工作步骤的暂时封闭持守）"]
        direction TB
        Cand["候选推导与 Lean 形式化推进：局部平滑、亚指数 CNF 计数、四维杨-米尔斯存在性"]
        Exam["开放审验：核查语义适切性、周围系统一致性与修正后的假设"]
        Cont["继续延伸的余项：每一项被接受的推导，依然让独立命题在系统内部既未被证明也未被证伪，并让一致性不可证"]
        Cand --> Exam --> Cont
    end
    subgraph S_Conv["包装与接收中的镜像反面（被当作永久封闭持守）"]
        direction TB
        Pack["公告包装：重大科学发现、Lean 验证证书、具名把关学者或指定小组推迟发布"]
        Delay["病毒式接收：把余留的审验当作永久闭合之前的一段延迟"]
        Hold["倒置的持守：探究中仅能暂时封闭持守之物，以公告传播速度被当作已被永久封闭持守"]
        Pack --> Delay --> Hold
    end
    Cand -.->|"在审验完成前被包装与接收"| Pack
    Hold ==>|"作为形式系统内的特定推导接受重新审验"| Exam
    style S_Open fill:#161b22,stroke:#3fb950,stroke-width:1.5px,color:#3fb950
    style S_Conv fill:#161b22,stroke:#f85149,stroke-width:1.5px,color:#f85149
    style Cand fill:#0d1117,stroke:#58a6ff,stroke-width:1px,color:#f0f6fc
    style Exam fill:#0d1117,stroke:#3fb950,stroke-width:1px,color:#f0f6fc
    style Cont fill:#0d1117,stroke:#a371f7,stroke-width:1px,color:#f0f6fc
    style Pack fill:#0d1117,stroke:#d29922,stroke-width:1px,color:#f0f6fc
    style Delay fill:#0d1117,stroke:#f85149,stroke-width:1px,color:#f0f6fc
    style Hold fill:#0d1117,stroke:#f85149,stroke-width:1px,color:#f0f6fc

This conversion is distinct from the discovery process itself. The generation of candidate formalizations, the production of Lean developments, and the subsequent open examination remain ordinary investigative activity, in which a derivation or working assumption is held closed temporarily to test what follows and to force re-examination of prior assumptions. The conversion is not located in those steps. It is located in the immediate registration, circulating at the speed of the announcement, in which what cannot be held closed—and is held closed only temporarily in inquiry—is treated as if it were held closed permanently. The impression of permanent closure does not depend on the eventual correctness of any particular formalization; it is produced by the packaging and the reception.

flowchart TD
    subgraph S_Open["Ordinary Investigative Activity (Held Closed Temporarily as Working Step)"]
        direction TB
        Cand["Candidate derivations and Lean developments: local smoothing, subexponential CNF counting, 4D Yang–Mills existence"]
        Exam["Open examination: checking semantic adequacy, surrounding consistency, and revised assumptions"]
        Cont["Continuing residue: every accepted derivation still leaves independent statements neither proved nor refuted and consistency unprovable inside the system"]
        Cand --> Exam --> Cont
    end
    subgraph S_Conv["Negative Mirror Image in Packaging and Reception (Treated as Held Closed Permanently)"]
        direction TB
        Pack["Announcement packaging: major scientific discoveries, Lean certificates, named stewards, or deferred group release"]
        Delay["Viral reception: treats residual examination as a delay before permanent closure"]
        Hold["Inverted hold: what is held closed temporarily in inquiry is treated at announcement speed as if held closed permanently"]
        Pack --> Delay --> Hold
    end
    Cand -.->|"Packaged and received prior to examination"| Pack
    Hold ==>|"Re-examined as a particular derivation inside a formal system"| Exam
    style S_Open fill:#161b22,stroke:#3fb950,stroke-width:1.5px,color:#3fb950
    style S_Conv fill:#161b22,stroke:#f85149,stroke-width:1.5px,color:#f85149
    style Cand fill:#0d1117,stroke:#58a6ff,stroke-width:1px,color:#f0f6fc
    style Exam fill:#0d1117,stroke:#3fb950,stroke-width:1px,color:#f0f6fc
    style Cont fill:#0d1117,stroke:#a371f7,stroke-width:1px,color:#f0f6fc
    style Pack fill:#0d1117,stroke:#d29922,stroke-width:1px,color:#f0f6fc
    style Delay fill:#0d1117,stroke:#f85149,stroke-width:1px,color:#f0f6fc
    style Hold fill:#0d1117,stroke:#f85149,stroke-width:1px,color:#f0f6fc

三、 形式系统产出它自身无法持守的命题 / 3. The Formal System Produces What It Cannot Itself Hold

反讽就内在于同一场传播之中。机器作为形式系统运作。内核接受一条定理，所确立的仅仅是所陈述的定理从所陈述的公理与规则中推导而出。这一接受把周围系统的一致性以及超出其资源的真命题留在形式化内部不可推导之处，并把形式命题与目标数学主张之间的语义对应留在形式化之外。因此，验证作为在形式系统之外执行的开放区分活动继续进行——即由人类的审验去登记形式陈述是否匹配数学主张（如ETH 形式化范围说明或先前的自然语言证明与 Lean 形式化推进之间的错位所示），以及该结果是否在无矛盾的前提下修正了先前的假设。形式系统产出的命题，无法由它自身在数学持守所要求的意义上确立为真。当这种产出被呈现为机器已经持守了结果，当余留的人类审验被登记为机器产出不再需要它之前的一段暂时延迟，绝对的不完备性便被压缩成了它的反面，上演了验证的倒置。形式系统无法容纳的那份开放余项，被登记成了由形式系统自身所交付之物。

The irony is internal to the same circulation. The machine operates as a formal system. Its kernel acceptance establishes only that a stated theorem follows from the stated axioms and rules. That acceptance leaves the consistency of the surrounding system, and the true statements beyond its resources, underivable inside the formalization, and the semantic correspondence to the intended claim outside it. Verification therefore continues as open distinguishing activity performed outside the formal system—human examination that registers whether the formal statement matches the mathematical claim, as in the ETH formalization scope note or prior mismatches between natural-language proofs and Lean developments, and whether the result revises prior assumptions without contradiction. The formal system produces statements it cannot itself secure as true in the sense required for mathematical holding. When that production is presented as the machine having already held the result, and when the residual human examination is registered as a temporary delay before the machine’s output no longer needs it, the absolute incompleteness is compressed into its opposite, enacting The Reversal of Verification. The open residue that the formal system cannot contain is registered as having been delivered by the formal system itself.

在前文所考察的形式化之中，同样的双重界限依然适用：内核的接受不会触及任何其真伪超出形式化自身资源的进一步命题。此类命题在系统内部依然既未被证明，也未被证伪。外部审验可以登记形式陈述与目标主张之间的对应，或登记其与先前结果的一致性，然而这一审验发生在一个承袭着同样不完备性的更大活动之内。形式系统依然无法通过证明那份余项将其作为封闭之物来持守，也依然无法通过证伪那份余项将其作为封闭之物来持守。

In the formalizations considered earlier, the same dual limit applies: acceptance by a kernel leaves untouched any further statement whose truth or falsehood reaches beyond the formalization’s own resources. Such a statement remains neither proved nor refuted inside the system. External examination can register correspondence to an intended claim or consistency with prior results, yet the examination occurs within a larger activity that inherits the same incompleteness. The formal system still cannot hold the residue closed by proving it, and it still cannot hold the residue closed by refuting it.

四、 陈述定理与第一人称的登记 / 4. Articulating the Theorem and Registering It in the First Person

陈述不完备定理与从第一人称感知中登记不完备定理之间的差异，遵循着同一个条件。这些定理可以作为关于形式系统的命题被陈述、被核验、被传递。无论陈述者是否从区分活动的内部登记到这道界限，那份陈述都依然可得，正如进位无法在寄存器之内被命名在元系统内将不完备性形式化与在自身上限处体会其意涵之间所追溯的那样。第一人称的登记，就是认出：任何把系统作为封闭之物来持守的尝试，本身都是又一次区分动作——它只能把某道边界作为工作步骤暂时持守，而无法把自身的延续作为永久闭合容纳进来。不可推导性不再作为外部形式客体的属性被遭遇，而是作为当下动作的开放余项被遭遇。由于这份余项是活动的延续，而非添加进活动之中的某项内容，第一人称的登记便无法靠追加陈述来产生。外部陈述可以指向这份余项，但只要活动尚未从内部登记到自身的界限，那份指向就依然只是一句陈述——这正是论封闭系统、开放心智与证明的界限所追溯的边界。在登记发生之处，永久持守着某个闭合结果的印象在动作之中便已消解；在登记未发生之处，那份陈述便继续作为一条关于形式系统的定理流转，而暂时持守仍继续被当作永久闭合来对待。

The difference between articulating the incompleteness theorems and registering them from first-person perception follows the same condition. The theorems can be stated, checked, and transmitted as propositions about formal systems. That articulation remains available whether or not the one stating it registers the limit from inside the distinguishing activity, as The Carry That Cannot Be Named Inside the Register traces between formalizing incompleteness in a meta-system and feeling its meaning at one’s own ceiling. First-person registration is the recognition that any attempt to hold the system closed is itself a further distinguishing act—one that holds a boundary only temporarily as a working step and cannot contain its own continuation as permanent closure. The underivability is encountered not as a property of an external formal object but as the open residue of the present act. Because the residue is the continuation of the activity rather than a content added to it, the registration cannot be produced by additional articulation. An external statement can point toward the residue, but the pointing remains an articulation until the activity registers its own limit from within, the boundary traced in On Closed Systems, Open Minds, and the Limits of Proof. Where the registration occurs, the impression of holding a permanently closed result is already dissolved in the act; where it does not, the articulation continues to circulate as a theorem about systems while temporary holds continue to be treated as permanent.

因此，这一事件两度显现了不完备性——并非作为同一种表达的两次并列出现，而是互为镜像。在直接意义上，形式系统所生成的命题，其周围一致性在系统内部不可推导，其语义适切性则留在系统之外，从而要求系统自身无法执行的开放审验。在否定意义上，发布与接收中的持守把原本无法封闭持守、在探究工作的特定语境下仅能暂时封闭持守之物，当作已超越该语境而被永久封闭持守来对待，倒转了这一要求的方向。这种持守并非与形式生成并列的不完备性的直接表达，而是不完备性的镜像反面——把受语境前提限定的暂时持守误作永久的闭合。实际的观察关乎这一倒置的速度与形式：开放的探究活动被包装并接收为一件一劳永逸地封闭持守之物。这一倒置在公告的形式与回应的形式中皆可审视。当区分活动登记到任何持守依然以特定语境为前提，且形式化依然把永久闭合留在自身内部既不可证明也不可证伪之处时，这一倒置便随之消解。

The episode therefore exhibits incompleteness twice—not as two instances of the same expression, but as mirror images. In its direct sense, the formal system generates statements whose surrounding consistency remains underivable inside it and whose semantic adequacy lies outside it, demanding open examination that the system cannot perform. In its negative sense, the hold in presentation and reception takes something that cannot be held closed—something held closed only temporarily within the specific context of working inquiry—and treats it as if it were held closed permanently beyond that context, reversing the direction of the demand. The hold is not a direct expression of incompleteness alongside formal generation; it is the negative mirror image of incompleteness, where a temporary contextual hold is mistaken for permanent closure. The practical observation concerns the speed and form of that inversion: open investigative activity is packaged and received as an item held closed once and for all. The inversion is available for inspection in the form of the announcement and the form of the response. It dissolves when the distinguishing activity registers that any hold remains preconditioned by its context and that the formalization leaves permanent closure neither provable nor refutable inside itself.

在这一否定意义内部，这种转化在开放余项自身之上跨越双重界限的两个方向运作。发布与接收可以把暂时的持守当作开放余项已被永久证明来对待；它同样可以将其当作开放余项已被永久证伪或搁置来对待。无论在哪一个方向上，这种持守都把一致系统在其资源之外留作不可判定之物登记为已永久落定。对这一界限的第一人称登记，就是认出：无论肯定的永久持守还是否定的永久持守，在企图认领它的那个动作内部都不可得。这份余项作为区分活动的延续继续延伸，既未被那套无法容纳它的形式化所永久坐实，也未被其永久消除。

flowchart TD
    subgraph S_First["直接意义上的显现：形式生成中的开放余项"]
        direction TB
        K["内核接受：仅核查所陈述的定理从所陈述的公理与规则中推导而出"]
        R1["双重开放余项：独立命题在系统内部既未被证明也未被证伪，且系统自身的一致性在系统内部不可证；语义适切性留在形式化之外"]
        K --> R1
    end
    subgraph S_Second["否定意义上的镜像显现：把暂时持守当作永久闭合"]
        direction TB
        C["倒置的持守：原本无法封闭持守之物被暂时封闭持守，却在包装与接收中被当作开放余项已被永久封闭持守（已证明或已证伪）"]
    end
    A["未经登记的陈述：定理继续流转，而暂时持守仍被当作永久闭合来对待"]
    subgraph S_Reg["活动内部的第一人称登记"]
        direction TB
        F["第一人称认出：任何持守都只是暂时持守的离散动作；无论肯定还是否定的永久闭合在其中均不可得"]
        D["倒置在动作中消解：暂时持守回归工作之用，开放余项继续延伸"]
        F --> D
    end
    R1 ==>|"在公告与接收中被倒置"| C
    C --> A
    R1 ==>|"作为当下动作的开放余项被遭遇"| F
    A -.->|"在从内部登记之前，指向依然只是陈述"| F
    style S_First fill:#161b22,stroke:#58a6ff,stroke-width:1.5px,color:#58a6ff
    style S_Second fill:#161b22,stroke:#f85149,stroke-width:1.5px,color:#f85149
    style S_Reg fill:#161b22,stroke:#3fb950,stroke-width:1.5px,color:#3fb950
    style K fill:#0d1117,stroke:#58a6ff,stroke-width:1px,color:#f0f6fc
    style R1 fill:#0d1117,stroke:#a371f7,stroke-width:1px,color:#f0f6fc
    style C fill:#0d1117,stroke:#f85149,stroke-width:1px,color:#f0f6fc
    style A fill:#0d1117,stroke:#f85149,stroke-width:1px,color:#f0f6fc
    style F fill:#0d1117,stroke:#a371f7,stroke-width:1px,color:#f0f6fc
    style D fill:#0d1117,stroke:#3fb950,stroke-width:1px,color:#f0f6fc

Within that negative sense, the conversion operates across both directions of the dual limit on the open residue itself. Presentation and reception can treat a temporary hold as if the open residue were permanently proved; they can equally treat it as if the residue were permanently refuted or set aside. In either direction, the hold registers as permanently settled what the consistent system leaves undecidable beyond its resources. First-person registration of the limit is the recognition that neither an affirmative permanent hold nor a negative permanent hold is available inside the act that claims it. The residue continues as the continuation of the distinguishing activity, neither permanently secured nor permanently eliminated by the formalization that cannot contain it.

flowchart TD
    subgraph S_First["Direct Sense: Open Residue in Formal Generation"]
        direction TB
        K["Kernel acceptance: checks only that a stated theorem follows from stated axioms and rules"]
        R1["Dual open residue: independent statements remain neither proved nor refuted and the system's consistency remains unprovable inside it; semantic adequacy remains outside the formalization"]
        K --> R1
    end
    subgraph S_Second["Negative Sense (Mirror Image): Temporary Hold Treated as Permanent Closure"]
        direction TB
        C["Inverted hold: what cannot be held closed is held closed temporarily, yet packaged and received as if the open residue were held closed permanently (proved or refuted)"]
    end
    A["Unregistered articulation: the theorem circulates while temporary holds continue to be treated as permanent closure"]
    subgraph S_Reg["First-Person Registration Within the Activity"]
        direction TB
        F["First-person recognition: any hold is a discrete act that holds only temporarily; neither affirmative nor negative permanent closure is available inside it"]
        D["Inversion dissolved in the act: the temporary hold returns to working use as the open residue continues"]
        F --> D
    end
    R1 ==>|"Inverted in announcement and reception"| C
    C --> A
    R1 ==>|"Encountered as open residue of the present act"| F
    A -.->|"Pointing remains articulation until registered from within"| F
    style S_First fill:#161b22,stroke:#58a6ff,stroke-width:1.5px,color:#58a6ff
    style S_Second fill:#161b22,stroke:#f85149,stroke-width:1.5px,color:#f85149
    style S_Reg fill:#161b22,stroke:#3fb950,stroke-width:1.5px,color:#3fb950
    style K fill:#0d1117,stroke:#58a6ff,stroke-width:1px,color:#f0f6fc
    style R1 fill:#0d1117,stroke:#a371f7,stroke-width:1px,color:#f0f6fc
    style C fill:#0d1117,stroke:#f85149,stroke-width:1px,color:#f0f6fc
    style A fill:#0d1117,stroke:#f85149,stroke-width:1px,color:#f0f6fc
    style F fill:#0d1117,stroke:#a371f7,stroke-width:1px,color:#f0f6fc
    style D fill:#0d1117,stroke:#3fb950,stroke-width:1px,color:#f0f6fc