> Full-text translation from Chinese. The original remains authoritative for its research claims; this translation is not independent mathematical verification. # Reverse Goldbach Structural Closure Program 07 ## Minimal Counterexamples, Shift Transport, and Demand–Capacity Closure ### Reverse Goldbach Structural Closure Program 07 ### Minimal Counterexamples, Shift Transport, and Demand–Capacity Closure **Series name:** Reverse Goldbach Structural Closure Program **English name:** Reverse Goldbach Structural Closure Program **Abbreviation:** RG-SCP **Article:** 07 / planned 08–10 **Authors:** Neo.K; Aletheia (GPT-5.6 Sol) **Date:** 2026-09-28 **Version:** v0.1 **Nature:** Minimal-counterexample draft / transport-operator draft / demand–capacity closure interface **Status:** Not a proof of the Goldbach conjecture; this article converts the minimality information that “all smaller even integers succeed” into smooth-shift transport constraints and new demand lower bounds --- ## Abstract The first six articles of the Reverse Goldbach Structural Closure Program established, in sequence: 1. counterexample universe; 2. proper obstruction cover; 3. M6/Wheel coverage cost; 4. fixed-point section hole; 5. approximation/skew heuristic layer; 6. exact demand–capacity inequalities. The central necessary condition in the sixth article is: $$ F_G(N)=1 \Longrightarrow D_N(y)\le I_N(y), $$ where: $$ D_N(y) $$ is the prime-complement demand avoiding all small prime factors not exceeding $y$, whereas: $$ I_N(y) $$ is the available deep proper-factor incidence capacity. This article further assumes the existence of a minimal Goldbach counterexample: $$ \boxed{ N_* = \min\mathcal C_G. } $$ Thus, for every smaller even integer: $$ 4\le M0. $$ This article first expresses this as the “first empty even anti-diagonal” of the prime-pair lattice: $$ \boxed{ \forall M If $h$ has no additional structure, this lift condition mostly restates the original complement-obstruction condition without adding a factor source. The genuine increment occurs with **smooth shifts**. Let: $$ M_k = \prod_{i=1}^{k}p_i $$ be the $k$-th primorial. If: $$ M_k\mid h $$ and: $$ p>p_k $$ is prime, Then: $$ \gcd(p+h,M_k)=1. $$ Thus, if $N_*$ is a failure and $p$ is a representation endpoint in a smaller successful layer, then: $$ p+h $$ is not only composite, but must also avoid the first $k$ prime factors: $$ \boxed{ \operatorname{lpf}(p+h)>p_k. } $$ This provides the central transport principle of this article: $$ \boxed{ \text{minimality} + \text{smooth shift} \Longrightarrow \text{deep obstruction demand}. } $$ Notably, this conclusion does not require: $$ M_k\mid N_*. $$ It requires only: $$ M_k\mid h. $$ Thus, even when $N_*$ itself is not a primorial multiple, smooth-shift transport can still force a collection of $p_k$-rough composite complements. This article then defines a **minimal-counterexample transport graph**. The left vertices are: $$ h\in\mathcal H_k(N_*), $$ where: $$ M_k\mid h $$ and: $$ N_*-h>2p_k. $$ The right vertices are reduced prime endpoints of smaller Goldbach representations: $$ q\le\frac{N_*-h}{2}. $$ If: $$ q $$ and: $$ p=N_*-h-q $$ are both prime, an edge is added. Minimality guarantees that every left vertex has at least one edge. For any edge: $$ h\sim q, $$ we have: $$ N_*-q=p+h, $$ and: $$ \operatorname{lpf}(N_*-q)>p_k. $$ Thus, all right neighbors of the transport graph lie in: $$ \mathbb H_{p_k}(N_*). $$ If its maximum matching size is: $$ \nu_k(N_*), $$ then we obtain an exact demand lower bound: $$ \boxed{ D_{N_*}(p_k) \ge \nu_k(N_*). } $$ Then, using the failure capacity inequality from the sixth article: $$ D_{N_*}(p_k) \le I_{N_*}(p_k) \le U_{N_*}(p_k), $$ we obtain the minimal-counterexample necessary inequality: $$ \boxed{ \nu_k(N_*) \le U_{N_*}(p_k). } $$ If it can be proved in the future that: $$ \nu_k(N_*) > U_{N_*}(p_k), $$ then the minimal counterexample does not exist. This article further uses the maximum right degree: $$ \Delta_k^{R}(N_*) $$ to give a general matching lower bound: $$ \boxed{ \nu_k(N_*) \ge \left\lceil \frac{ |\mathcal H_k(N_*)| }{ \Delta_k^{R}(N_*) } \right\rceil. } $$ whereas: $$ \Delta_k^{R}(N_*) $$ can itself be controlled by primes in arithmetic progressions, because, for a fixed right endpoint $q$, all corresponding large endpoints: $$ p=N_*-q-h $$ lie in the same residue class: $$ p\equiv N_*-q\pmod{M_k}. $$ Thus, this article ultimately obtains a new proof architecture: $$ \boxed{ \text{minimality} \to \text{occupied lower diagonals} \to \text{smooth-shift graph} \to \text{matching demand} \to \text{deep-factor capacity} \to \text{contradiction}. } $$ This is more natural than the naive fixed-point descent attempted in the fourth article: rather than trying to move failure downward, it reverses “all states below succeed” into additional obstruction demand on the minimal failure. --- # 1. Minimal-Counterexample Assumption Define the Goldbach counterexample universe: $$ \mathcal C_G = \left\{ N\in2\mathbb N: N\ge4, \quad R_G(N)=0 \right\}. $$ Assume: $$ \mathcal C_G\neq\varnothing. $$ then the well-ordering of the natural numbers gives the existence of: $$ \boxed{ N_* = \min\mathcal C_G. } $$ Therefore: $$ F_G(N_*)=1. $$ and: $$ \boxed{ \forall M, \quad 4\le M0. } $$ --- # 2. Prime-pair lattice Define: $$ \boxed{ \mathcal G(M) = \left\{ (p,q)\in\mathbb P^2: p+q=M \right\}. } $$ This is the ordered prime-pair representation fiber. we have: $$ R_G(M)>0 \Longleftrightarrow \mathcal G(M)\neq\varnothing. $$ Thus, minimality is equivalent to: $$ \boxed{ \mathcal G(M)\neq\varnothing } $$ for every smaller even integer $M$, but: $$ \boxed{ \mathcal G(N_*)=\varnothing. } $$ --- # 3. First Empty Even Anti-Diagonal On: $$ \mathbb P\times\mathbb P $$ the lattice, the equation: $$ p+q=M $$ is an anti-diagonal. Thus, the geometric meaning of a hypothetical minimal counterexample $N_*$ is: > Every lower even anti-diagonal meets at least one prime lattice point, but the even anti-diagonal at $N_*$ is the first to become completely empty. This article calls this: # First Empty Even Anti-Diagonal Formally: $$ \boxed{ N_* = \min \left\{ M\in2\mathbb N_{\ge4}: \mathcal G(M)=\varnothing \right\}. } $$ --- # 4. Representation count frontier Define the ordered representation count: $$ \boxed{ r_2(M) = |\mathcal G(M)|. } $$ Then minimality gives: $$ r_2(M)>0 $$ for every smaller even integer $M$, and: $$ r_2(N_*)=0. $$ Therefore: $$ \boxed{ r_2(N_*-2)>0, \quad r_2(N_*)=0. } $$ Thus, the minimal counterexample must be accompanied by a terminal drop: $$ r_2(N_*)-r_2(N_*-2) = -r_2(N_*-2) <0. $$ This is exact, but does not yet form a contradiction. --- # 5. Why “The Representation Count Suddenly Drops to Zero” Is Not Enough Even if: $$ r_2(M) $$ appears smooth or positive on average at many scales, empirical smoothness alone cannot exclude: $$ r_2(N_*)=0. $$ Thus: $$ \boxed{ \text{first zero} } $$ is a target condition, not a contradiction in itself. If an analytic, spectral, or combinatorial theorem can prove in the future that: $$ r_2(M) $$ cannot drop from positive values to zero in a certain interval, only then would genuine closure be formed. --- # 6. Even shift Let: $$ h $$ be a positive even integer, and let: $$ M=N_*-h. $$ provided that: $$ 4\le M None of the Goldbach endpoints of $N_*-2$ can be twin-prime starts. This is an exact structural constraint. --- # 13. General Even-Gap Blocking For any even: $$ h, $$ a minimal counterexample requires: > No Goldbach endpoint of $N_*-h$ can meet another prime at distance $h$. That is: $$ \boxed{ \mathcal E_h(N_*) \cap \left( \mathbb P-h \right) = \varnothing. } $$ This connects minimality with prime-pair gaps. This article does not, however, assume any unproved prime-gap conjecture. --- # 14. The Limited Information Increment of a Naive Lift If all we know is that: $$ p+h $$ is composite, then, because: $$ p+h = N_*-q, $$ this is merely: $$ N_*-q $$ the complement obstruction that this is composite. Thus: $$ \boxed{ \text{unstructured lift blocking} } $$ is mostly just reindexing of the old obstruction map. Thus, if minimality is to supply genuinely new power, additional structure must be imposed on: $$ h $$ this shift. --- # 15. Smooth shift Let: $$ M_k = \prod_{i=1}^{k}p_i. $$ If: $$ M_k\mid h, $$ then: $$ h $$ is called a **primorial-smooth shift** at layer $k$. Because: $$ p_i\mid h $$ for every: $$ i\le k. $$ --- # 16. Shift Self-Suppression ## Lemma RG-SCP-07.2 If: $$ M_k\mid h, $$ and: $$ p>p_k $$ is prime, Then: $$ \boxed{ \gcd(p+h,M_k)=1. } $$ --- ## Proof For any: $$ p_i\le p_k, $$ we have: $$ p_i\mid h. $$ If also: $$ p_i\mid(p+h), $$ Then: $$ p_i\mid p. $$ but: $$ p $$ is a prime greater than: $$ p_k $$ this cutoff, Thus: $$ p_i\nmid p. $$ a contradiction. Therefore: $$ p_i\nmid(p+h) $$ for every: $$ i\le k. $$ Hence: $$ \gcd(p+h,M_k)=1. $$ This completes the proof. --- # 17. Smooth-Shift Deep Obstruction Theorem ## Theorem RG-SCP-07.3 If: $$ F_G(N_*)=1, $$ $$ M_k\mid h, $$ and: $$ (p,q)\in\mathcal G(N_*-h) $$ and satisfies: $$ p>p_k, $$ Then: $$ \boxed{ p+h \text{ composite} } $$ and: $$ \boxed{ \operatorname{lpf}(p+h)>p_k. } $$ --- ## Proof By lift-blocking: $$ p+h $$ is composite. By shift self-suppression: $$ \gcd(p+h,M_k)=1. $$ Thus: $$ p+h $$ has no prime divisor not exceeding: $$ p_k $$ this cutoff. Therefore: $$ \operatorname{lpf}(p+h)>p_k. $$ This completes the proof. --- # 18. What Is Genuinely Added Here Primorial suppression in the third article requires: $$ M_k\mid N. $$ This article instead obtains: $$ \boxed{ M_k\mid h } $$ as sufficient to impose, on the transported endpoint: $$ p+h $$ the same deep-factor requirement. Thus: $$ \boxed{ M_k\nmid N_* } $$ does not prevent us from using primorial smoothness. It suffices to choose: $$ h $$ is: $$ M_k $$ as a multiple of this primorial. This is where minimality transport genuinely differs from the original obstruction geometry. --- # 19. Reduced representation endpoint For an even: $$ M, $$ Define: $$ \boxed{ \mathcal R^{-}(M) = \left\{ q\in\mathbb P: q\le\frac M2, \quad M-q\in\mathbb P \right\}. } $$ If: $$ R_G(M)>0, $$ Then: $$ \mathcal R^{-}(M)\neq\varnothing. $$ Every: $$ q\in\mathcal R^{-}(M) $$ corresponds to: $$ p=M-q $$ and: $$ p\ge\frac M2. $$ --- # 20. Why Reduced Orientation Matters If: $$ M>2p_k, $$ Then: $$ \frac M2>p_k. $$ Thus, for: $$ q\in\mathcal R^{-}(M), $$ its partner: $$ p=M-q $$ satisfying: $$ p\ge\frac M2>p_k. $$ Thus, every reduced representation automatically supplies: $$ p>p_k. $$ This is precisely what is needed to apply the smooth-shift theorem. --- # 21. Smooth-shift layer set For a hypothetical minimal counterexample: $$ N_*, $$ Define: $$ \boxed{ \mathcal H_k(N_*) = \left\{ h\in\mathbb N: h\ge M_k, \quad M_k\mid h, \quad N_*-h>2p_k \right\}. } $$ Because: $$ M_k $$ contains $2$, Thus: $$ h $$ is automatically even. For: $$ h\in\mathcal H_k(N_*), $$ we have: $$ M_h = N_*-h $$ is a smaller even integer and: $$ M_h>2p_k. $$ --- # 22. Minimality Guarantees Occupancy of Every Layer From: $$ N_* $$ being minimal, for every: $$ h\in\mathcal H_k(N_*), $$ we have: $$ R_G(M_h)>0. $$ Thus: $$ \boxed{ \mathcal R^{-}(M_h)\neq\varnothing. } $$ Thus, every smooth-shift layer has at least one reduced Goldbach witness. --- # 23. Transport graph Define the bipartite graph: $$ \boxed{ \mathcal T_k(N_*) = (L_k,R_k,E_k) } $$ where the left side is: $$ L_k = \mathcal H_k(N_*). $$ On the right, take all possible reduced prime endpoints: $$ R_k = \left\{ q\in\mathbb P: q<\frac{N_*}{2} \right\}. $$ Define the edge: $$ h\sim q $$ if and only if: $$ q\in \mathcal R^{-}(N_*-h). $$ --- # 24. Every Left Vertex Is Nonisolated By minimality: $$ \boxed{ \deg_L(h)\ge1 } $$ for every: $$ h\in L_k. $$ This is the most direct graph-theoretic manifestation of minimality in the transport graph. --- # 25. The Large Endpoint of an Edge If: $$ h\sim q, $$ Let: $$ p = N_*-h-q. $$ Because: $$ q\le\frac{N_*-h}{2}, $$ Thus: $$ p\ge\frac{N_*-h}{2}. $$ Also: $$ N_*-h>2p_k. $$ Hence: $$ \boxed{ p>p_k. } $$ --- # 26. Edge-to-hard-survivor theorem ## Theorem RG-SCP-07.4 If: $$ h\sim q $$ in: $$ \mathcal T_k(N_*), $$ Then: $$ \boxed{ q\in \mathbb H_{p_k}(N_*). } $$ --- ## Proof By the definition of an edge, $$ p=N_*-h-q $$ is prime, and by the preceding section: $$ p>p_k. $$ Also: $$ M_k\mid h. $$ Thus: $$ \gcd(p+h,M_k)=1. $$ but: $$ p+h = N_*-q. $$ Therefore: $$ \gcd(N_*-q,M_k)=1. $$ Also: $$ q<\frac{N_*}{2}. $$ Thus: $$ q\in \mathbb H_{p_k}(N_*). $$ This completes the proof. --- # 27. More Strongly: Right Neighbors Have Rough Composite Complements Because: $$ F_G(N_*)=1, $$ For: $$ q\in R_k $$ if it is a transport neighbor, Then: $$ N_*-q $$ is composite. Then, by RG-SCP-07.4: $$ \gcd(N_*-q,M_k)=1. $$ Thus: $$ \boxed{ \operatorname{lpf}(N_*-q)>p_k. } $$ Thus, the right neighborhood of the transport graph is a deep-obstruction subset forced by minimality. --- # 28. Transport neighborhood Define: $$ \boxed{ \Gamma_k(N_*) = \Gamma_{\mathcal T_k}(L_k) } $$ as the union of the right neighbors of all smooth-shift layers. Then: $$ \boxed{ \Gamma_k(N_*) \subseteq \mathbb H_{p_k}(N_*). } $$ Thus: $$ \boxed{ D_{N_*}(p_k) \ge |\Gamma_k(N_*)|. } $$ --- # 29. Why Counting Layers Alone Is Not Enough Although: $$ |L_k| $$ may be large, different smooth shifts may share the same: $$ q. $$ Thus, one cannot directly write: $$ D_{N_*}(p_k)\ge|L_k|. $$ This would count the right endpoint repeatedly. It is necessary to control: $$ \boxed{ \text{transport collision}. } $$ --- # 30. Maximum matching Let: $$ \boxed{ \nu_k(N_*) = \nu \left( \mathcal T_k(N_*) \right) } $$ be the maximum matching size of the transport graph. A matching selects: $$ \nu_k $$ distinct left shifts, and: $$ \nu_k $$ distinct right prime endpoints. Thus: $$ \boxed{ |\Gamma_k(N_*)| \ge \nu_k(N_*). } $$ --- # 31. Matching Demand Theorem ## Theorem RG-SCP-07.5 If: $$ N_* $$ is a minimal Goldbach counterexample, Then: $$ \boxed{ D_{N_*}(p_k) \ge \nu_k(N_*). } $$ --- ## Proof By RG-SCP-07.4, all right neighbors of the transport graph lie in: $$ \mathbb H_{p_k}(N_*). $$ A maximum matching uses: $$ \nu_k $$ distinct right vertices. Therefore: $$ \mathbb H_{p_k}(N_*) $$ contains at least: $$ \nu_k $$ elements. Hence: $$ D_{N_*}(p_k) \ge \nu_k. $$ This completes the proof. --- # 32. This Is Minimality-Derived Demand The demand in the sixth article: $$ D_N(p_k) $$ is defined directly by the complement structure of $N$. This article instead derives, from: $$ \text{all smaller evens are successful} $$ the lower bound: $$ \boxed{ D_{N_*}(p_k) \ge \nu_k(N_*). } $$ Thus: $$ \nu_k $$ is: # Minimality-Derived Transport Demand --- # 33. Connection with Capacity in the Sixth Article In the sixth article, if: $$ F_G(N_*)=1, $$ Then: $$ D_{N_*}(p_k) \le I_{N_*}(p_k). $$ If there is also a rigorous upper bound: $$ I_{N_*}(p_k) \le U_{N_*}(p_k), $$ Then: $$ \boxed{ \nu_k(N_*) \le U_{N_*}(p_k). } $$ this is a new necessary inequality that a minimal counterexample must satisfy. --- # 34. Minimality Demand–Capacity Closure ## Theorem Template RG-SCP-07.A If, for a hypothetical minimal counterexample $N_*$, one can prove: $$ \boxed{ \nu_k(N_*) > U_{N_*}(p_k), } $$ Then: $$ \boxed{ N_* \text{ does not exist}. } $$ Thus, the Goldbach counterexample universe is empty. Note: > This is only a closure template; this article has not yet established general bounds that make the two sides cross. --- # 35. Maximum Right Degree Define: $$ \boxed{ \Delta_k^{R}(N_*) = \max_{q\in R_k} \deg(q). } $$ It measures: > The maximum number of smooth-shift layers that can share the same reduced prime endpoint $q$. If: $$ \Delta_k^R $$ is large, the transport layers collide heavily, and matching demand becomes smaller. --- # 36. Bounded-Degree Matching Lemma ## Lemma RG-SCP-07.6 If every left vertex of a finite bipartite graph has degree at least $1$, and its maximum right degree is: $$ \Delta_R, $$ then its maximum matching: $$ \nu $$ satisfying: $$ \boxed{ \nu \ge \left\lceil \frac{|L|}{\Delta_R} \right\rceil. } $$ --- ## Proof Take a maximal matching of size: $$ m. $$ If an unmatched left vertex is adjacent to an unmatched right vertex, that edge can be added, contradicting maximality. Thus, every left vertex is adjacent to some matched right vertex. Every matched right vertex is adjacent to at most: $$ \Delta_R $$ left vertices. Therefore: $$ |L| \le m\Delta_R. $$ Hence: $$ m \ge \frac{|L|}{\Delta_R}. $$ The maximum matching: $$ \nu\ge m, $$ Thus: $$ \nu \ge \left\lceil \frac{|L|}{\Delta_R} \right\rceil. $$ This completes the proof. --- # 37. Transport demand lower bound Applying RG-SCP-07.6: $$ \boxed{ D_{N_*}(p_k) \ge \nu_k(N_*) \ge \left\lceil \frac{ |\mathcal H_k(N_*)| }{ \Delta_k^{R}(N_*) } \right\rceil. } $$ This is the first minimality demand lower bound that does not require directly computing: $$ D_{N_*} $$ this demand. --- # 38. Smooth-shift layer count For: $$ k\ge2, $$ From: $$ N_*-h>2p_k $$ and both sides are even, is equivalent to: $$ h\le N_*-2p_k-2. $$ Thus, if: $$ N_*>2p_k+2+M_k, $$ Then: $$ \boxed{ |\mathcal H_k(N_*)| = \left\lfloor \frac{ N_*-2p_k-2 }{ M_k } \right\rfloor. } $$ If the right side is less than $1$, this layer family is empty. --- # 39. Minimality Produces a Linear Number of Smooth Layers Fix: $$ k, $$ when: $$ N_*\to\infty, $$ we have: $$ |\mathcal H_k(N_*)| \asymp \frac{N_*}{M_k}. $$ This is merely an elementary layer count. The real question is: $$ \Delta_k^R(N_*) $$ whether it is likewise large enough to swallow the matching lower bound completely. --- # 40. Arithmetic-Progression Structure of the Right Degree Fix a right vertex: $$ q. $$ If: $$ h\sim q, $$ Then: $$ p = N_*-q-h $$ is prime. Also: $$ M_k\mid h. $$ Thus: $$ \boxed{ p \equiv N_*-q \pmod{M_k}. } $$ Thus, for the same: $$ q $$ the partner primes corresponding to all its left neighbors: $$ p $$ all lie in a single arithmetic progression. --- # 41. Degree AP bound Because: $$ h\ge M_k, $$ we have: $$ p \le N_*-q-M_k. $$ Thus: $$ \deg(q) $$ does not exceed: $$ \boxed{ \pi \left( N_*-q-M_k; M_k, N_*-q\bmod M_k \right) } $$ when the residue is reduced. If: $$ \gcd(N_*-q,M_k)>1, $$ then for: $$ p>p_k $$ there is no edge at all. --- # 42. Global right-degree envelope Define: $$ \boxed{ B_k(N_*) = \max_{\substack{ a\bmod M_k\\ \gcd(a,M_k)=1 }} \pi \left( N_*-M_k-2; M_k,a \right). } $$ Then: $$ \boxed{ \Delta_k^R(N_*) \le B_k(N_*). } $$ This is a rigorous but raw AP envelope. --- # 43. AP-controlled transport demand Therefore: $$ \boxed{ D_{N_*}(p_k) \ge \left\lceil \frac{ |\mathcal H_k(N_*)| }{ B_k(N_*) } \right\rceil. } $$ This combines, for the first time: - minimality; - primorial smooth shift; - arithmetic progression occupancy; into a demand lower bound. --- # 44. External AP Upper-Bound Interface If one uses any rigorous: $$ B_k(N_*) \le \widetilde B_k(N_*), $$ one obtains: $$ \boxed{ D_{N_*}(p_k) \ge \left\lceil \frac{ |\mathcal H_k(N_*)| }{ \widetilde B_k(N_*) } \right\rceil. } $$ One can use: - Brun–Titchmarsh-type upper bound; - explicit AP prime-count bound; - stronger estimates for specific moduli. This article does not embed any external bound as completed closure. --- # 45. Hall transport criterion More strongly, If the transport graph satisfies: $$ \boxed{ |\Gamma(S)| \ge |S| } $$ for every: $$ S\subseteq L_k, $$ then, by Hall’s marriage theorem, there exists a matching covering the entire left side. Thus: $$ \nu_k(N_*) = |L_k|. $$ Therefore: $$ \boxed{ D_{N_*}(p_k) \ge |\mathcal H_k(N_*)|. } $$ --- # 46. Meaning of the Hall Condition The Hall condition is equivalent to: > Smooth-shift layers cannot crowd excessively onto only a few right prime endpoints. If lower Goldbach representations have sufficient expansion, minimality forces: $$ N_* $$ to have many distinct rough composite complements. This is much stronger than mere existence of a witness at every layer. --- # 47. Transport congestion Define: $$ \boxed{ \kappa_k(N_*) = \frac{ |L_k| }{ |\Gamma_k(N_*)| } } $$ when the right neighborhood is nonempty. It measures, on average: > How many smooth-shift layers share one right endpoint. Hall-like expansion corresponds to: $$ \kappa_k $$ being close to $1$. Heavy collision corresponds to: $$ \kappa_k\gg1. $$ --- # 48. Congestion Is Not Arbitrary Fix: $$ q, $$ if it is shared by many: $$ h $$ of these shifts, Then: $$ N_*-q-h $$ must repeatedly be prime in the same: $$ \bmod M_k $$ residue class. Thus, high transport congestion itself requires prime concentration in an AP. Therefore: $$ \boxed{ \text{small matching demand} } $$ and: $$ \boxed{ \text{large AP congestion} } $$ trade off against one another. --- # 49. The First Minimality Trade-Off From: $$ D_{N_*}(p_k) \ge \frac{|L_k|}{\Delta_k^R}, $$ one can read: $$ \boxed{ \text{To make transport demand small, some right endpoints must bear high congestion}. } $$ High congestion, in turn, requires the corresponding AP capacity. This is the first genuinely nontrivial trade-off supplied by minimality. --- # 50. Dual AP Structure with the Sixth Article’s Capacity Interestingly: ### Transport side controlling: $$ \Delta_k^R $$ requires counting: $$ p \equiv N_*-q \pmod{M_k}. $$ ### Obstruction side controlling: $$ I_{N_*}(p_k) $$ requires counting deep factor classes: $$ q \equiv N_* \pmod r. $$ Thus, a minimal counterexample must simultaneously maintain two sets of AP capacities: $$ \boxed{ \text{lower-success transport AP} } $$ and: $$ \boxed{ \text{upper-failure obstruction AP}. } $$ This may form new cross-source coupling. --- # 51. Two-Capacity Squeeze This can be written abstractly as: $$ \boxed{ \frac{ |\mathcal H_k(N_*)| }{ \text{TransportCongestionCapacity} } \le D_{N_*}(p_k) \le \text{ObstructionCapacity}. } $$ If the lower bound on the left exceeds the upper bound on the right, the minimal counterexample is excluded. This is the central squeeze architecture of this article. --- # 52. Transport demand lower envelope Define any rigorous: $$ \Delta_k^R(N_*) \le C_k^{\mathrm{trans}}(N_*). $$ Then: $$ \boxed{ L_k^{\mathrm{trans}}(N_*) = \left\lceil \frac{ |\mathcal H_k(N_*)| }{ C_k^{\mathrm{trans}}(N_*) } \right\rceil } $$ satisfying: $$ D_{N_*}(p_k) \ge L_k^{\mathrm{trans}}(N_*). $$ --- # 53. Minimal-counterexample closure criterion If there is also the rigorous capacity upper bound from the sixth article: $$ I_{N_*}(p_k) \le U_k^{\mathrm{obs}}(N_*), $$ then a minimal counterexample must satisfy: $$ \boxed{ L_k^{\mathrm{trans}}(N_*) \le U_k^{\mathrm{obs}}(N_*). } $$ If one can prove: $$ \boxed{ L_k^{\mathrm{trans}}(N_*) > U_k^{\mathrm{obs}}(N_*), } $$ then no minimal counterexample exists. --- # 54. This Is More Natural than Naive Descent The fourth article studies: $$ N_* \to N'p_k $$ being prime, it automatically avoids all: $$ p_i. $$ Thus, a smooth shift directly transports: $$ \boxed{ \text{prime endpoint} } $$ into: $$ \boxed{ \text{small-factor-free shifted endpoint}. } $$ This is an exact transport, not a heuristic. --- # 57. Shift-space source novelty The third article uses: $$ M_k\mid N. $$ This article uses: $$ M_k\mid h. $$ The two have different provenance: - the former is target-state divisor structure; - the latter is transport-step structure. Thus, shift space provides a new operational degree of freedom. The ultimate factor witness, however, is still the least-prime-factor source. Thus, source novelty is: $$ \boxed{ \text{new constraint geometry}, } $$ not: $$ \boxed{ \text{new primality oracle}. } $$ --- # 58. Smooth shift ladder One can fix: $$ k $$ and examine: $$ h_j=jM_k. $$ This gives: $$ N_*-M_k, $$ $$ N_*-2M_k, $$ $$ N_*-3M_k, $$ $$ \ldots $$ this sequence of lower even layers. Minimality guarantees a Goldbach representation at every layer, until: $$ N_*-h_j $$ becomes too small. This forms: $$ \boxed{ \text{smooth-shift success ladder}. } $$ --- # 59. Ladder terminal condition Every layer: $$ h_j $$ produces at least one reduced endpoint: $$ q_j. $$ and its partner: $$ p_j $$ satisfying: $$ p_j>p_k. $$ Thus: $$ N_*-q_j = p_j+h_j $$ is: $$ p_k $$ -rough composite. If many: $$ q_j $$ are distinct, they rapidly increase: $$ D_{N_*}(p_k). $$ --- # 60. Collision Is One of the Only Ways Out Thus, if a hypothetical minimal counterexample is to avoid excessive transport demand, one way is to make many: $$ h_j $$ share the same: $$ q. $$ that is, to have heavy collision. But heavy collision crowds the partner primes: $$ p_j = N_*-q-jM_k $$ into the same arithmetic progression. Thus: $$ \boxed{ \text{avoid demand growth} \to \text{pay AP congestion cost}. } $$ --- # 61. A Second Layer of “Counterexamples Pay Their Own Cost of Existence” The principle in the first article is: > Counterexamples must pay the cost of existence. The seventh article now obtains a more precise version: > If a minimal counterexample is to avoid smooth-shift transport producing many distinct rough complements, it must use a small number of right endpoints to bear many lower Goldbach layers; this, in turn, requires the partner primes to be highly concentrated in fixed arithmetic progressions. Thus, the cost of existence has two modes of payment: $$ \boxed{ \text{many distinct rough complements} } $$ or: $$ \boxed{ \text{high transport congestion}. } $$ --- # 62. Transport dichotomy Conceptually, this can be written as: $$ \boxed{ \text{Minimal Counterexample} \Longrightarrow \begin{cases} \text{large matching demand},\\ \text{or high AP congestion}. \end{cases} } $$ This is not an additional theorem, but a structural reading of the bounded-degree matching inequality. --- # 63. Local transport subsets It is not necessary to use all of: $$ \mathcal H_k(N_*). $$ One can choose: $$ H\subseteq\mathcal H_k(N_*). $$ Define the induced transport graph: $$ \mathcal T_k(N_*;H). $$ Then, likewise: $$ D_{N_*}(p_k) \ge \nu \left( \mathcal T_k(N_*;H) \right). $$ This permits choosing: - short shift windows; - special multipliers; - low-congestion layers; - heuristic-selected layers. --- # 64. Interface Between Heuristic and Proof The fifth article can use: - representation counts; - congestion profiles; - skew; - pressure field; to search for a high-value shift subset: $$ H. $$ But the seventh article accepts only: $$ \boxed{ \text{rigorous matching lower bound} } $$ as proof input. Thus: $$ \text{heuristic subset selection} \to \text{exact graph proof}. $$ --- # 65. Further Refinement of the Transport Graph’s Right Degree The raw AP bound ignores that: $$ q $$ must itself form, with its partner: $$ p $$ a smaller Goldbach representation. That is: $$ p $$ being prime is already included, whereas: $$ q $$ is fixed and also prime. Thus, transport degree is: $$ \text{prime count in one AP} $$ rather than an arbitrary integer count. This is already much sparser than general layer collision. --- # 66. Further Restricting the $q$ Family If one chooses: $$ q $$ to lie in a residue family, the partner AP is further constrained by: $$ N_*-q $$ the residue of this complement. Thus, the local transport graph may have a smaller: $$ \Delta_R. $$ This gives a reason for subset optimization. --- # 67. Hall-expansion search The ideal transport theorem is not a bound on the maximum degree, but directly proves that a class of: $$ H $$ satisfying: $$ |\Gamma(S)|\ge|S| $$ for every: $$ S\subseteq H. $$ Then: $$ \nu=|H|. $$ This turns every smooth-shift layer into a distinct rough-complement demand. --- # 68. Hall theorem target shape A candidate theorem could take the form: > For some $k$ and all sufficiently large minimal-counterexample candidates $N$, the smooth-shift transport graph $\mathcal T_k(N)$ satisfies the Hall condition. If this can be proved, Then: $$ D_N(p_k) \ge |\mathcal H_k(N)|. $$ then compare it with the capacity upper bound. This article does not currently have such a theorem. --- # 69. A More General Version of Shift Divisor Suppression It is not necessary to require: $$ M_k\mid h. $$ For any shift: $$ h, $$ Define: $$ \mathcal D(h) = \left\{ r\in\mathbb P: r\mid h \right\}. $$ If: $$ p $$ is prime and: $$ p\notin\mathcal D(h), $$ Then: $$ r\in\mathcal D(h) \Longrightarrow r\nmid(p+h). $$ Therefore: $$ \boxed{ \text{prime divisors of }h } $$ are the suppressed obstruction colors of the shifted endpoint. --- # 70. General shift suppression theorem If: $$ F_G(N_*)=1, $$ $$ (p,q)\in\mathcal G(N_*-h), $$ and: $$ p $$ is unequal to every prime divisor of $h$, Then: $$ p+h $$ is composite, whereas: $$ \operatorname{lpf}(p+h) \notin \mathcal D(h). $$ A primorial shift merely designs: $$ \mathcal D(h) $$ to be: $$ \{p_1,\ldots,p_k\}. $$ --- # 71. Shift design problem Thus, future work need not even use only primorials. One can design: $$ h $$ so that its prime divisor set: $$ \mathcal D(h) $$ specifically suppresses a chosen set of obstruction channels. This forms: $$ \boxed{ \text{transport-step design}. } $$ The primorial is the most natural first version. --- # 72. Smoothness budget The smoother the shift, usually: $$ h $$ the larger it is. whereas: $$ h $$ the larger it is, the lower state: $$ N_*-h $$ the smaller it is. Thus: $$ \boxed{ \text{suppression depth} } $$ and: $$ \boxed{ \text{available lower-layer range} } $$ have a trade-off between them. --- # 73. The Trade-Off in $k$ Increasing: $$ k $$ will: ### Benefit suppress more small obstruction primes: $$ p_1,\ldots,p_k. $$ ### Cost $$ M_k $$ grows rapidly, Thus: $$ |\mathcal H_k(N_*)| $$ decreases rapidly. Therefore: $$ k $$ is a natural optimization parameter for transport demand. --- # 74. Transport objective Define the conceptual quantity: $$ \boxed{ \mathcal T\mathrm{Gain}_k(N) = \frac{ |\mathcal H_k(N)| }{ \Delta_k^R(N) } } $$ It approximates the unrounded version of the matching lower bound. The heuristic channel can search for: $$ k $$ such that: $$ \mathcal T\mathrm{Gain}_k $$ is maximized. The proof channel then proves the corresponding bounds. --- # 75. Duality with Cutoff Optimization in the Sixth Article The sixth article optimizes: $$ y. $$ The seventh article optimizes: $$ k $$ or: $$ h $$ its smoothness. The two correspond to: ### Obstruction side choosing a cutoff: $$ y=p_k. $$ ### Transport side choosing a shift: $$ M_k\mid h. $$ Thus, the same: $$ k. $$ can be chosen simultaneously, producing a very natural two-sided squeeze. --- # 76. Same-$k$ Closure Program Fix: $$ k. $$ Left side: $$ \boxed{ L_k^{\mathrm{trans}}(N_*) } $$ comes from smooth-shift transport matching. Right side: $$ \boxed{ U_k^{\mathrm{obs}}(N_*) } $$ comes from deep-factor obstruction capacity. Target: $$ \boxed{ L_k^{\mathrm{trans}}(N_*) > U_k^{\mathrm{obs}}(N_*). } $$ This is minimality closure at the same primorial scale. --- # 77. Different Scales May Be Needed It may also be optimal to have: $$ k_{\mathrm{trans}} \neq k_{\mathrm{obs}}. $$ Thus, more generally: $$ L_{k_1}^{\mathrm{trans}}(N_*) > U_{k_2}^{\mathrm{obs}}(N_*). $$ provided that the transport-neighbor set really lies in: $$ \mathbb H_{p_{k_2}}(N_*). $$ If: $$ k_2\le k_1, $$ Then: $$ p_{k_1}\text{-rough} \Rightarrow p_{k_2}\text{-rough}. $$ Thus, comparison across scales is possible. --- # 78. Cross-scale monotonicity If: $$ k_1\ge k_2, $$ Then: $$ \mathbb H_{p_{k_1}}(N) \subseteq \mathbb H_{p_{k_2}}(N). $$ Thus, distinct right neighbors generated by transport at a deep scale, can also serve as demand at a shallower scale. This provides cross-scale flexibility. --- # 79. Minimality-derived local subset The right vertices selected by a matching: $$ Q_k^{\mathrm{match}} $$ form: $$ \mathbb H_{p_k}(N_*) $$ an exact subset of this set. One can directly establish for: $$ Q_k^{\mathrm{match}} $$ a local capacity upper bound without controlling the whole: $$ D_{N_*}(p_k). $$ This may be easier than whole-domain capacity. --- # 80. Local capacity closure If a matching gives: $$ |Q_k^{\mathrm{match}}| = \nu_k, $$ and one can prove that the total deep-obstruction capacity of these q satisfies: $$ U_k^{\mathrm{local}} < \nu_k, $$ there is likewise a contradiction. Thus: $$ \boxed{ \text{matching-selected local demand} } $$ may be better suited to closure than global demand. --- # 81. But the Matching Subset Itself Depends on the Graph If the proof uses: $$ Q_k^{\mathrm{match}}, $$ it must provide: - matching certificate; - the smaller Goldbach representation for every edge; - distinct right endpoints; - smooth-shift divisibility; - local capacity bound. This is well suited to certificate-carrying proof architecture. --- # 82. Finite matching certificate For a fixed candidate: $$ N, $$ One can output: ```text k M_k shift list h_i for each h_i: smaller even M_i = N-h_i reduced Goldbach pair (q_i,p_i) matching: all q_i distinct checks: p_i + q_i = M_i p_i,q_i prime p_i > p_k M_k | h_i gcd(N-q_i,M_k)=1 ``` If one further proves: $$ N-q_i $$ are all composite, a finite transport obstruction certificate is formed. --- # 83. Symbolic Family Proof Is More Important But the ultimate goal of RG-SCP is not, for every: $$ N $$ to output an enormous matching. Instead, it is to establish a symbolic theorem: $$ \nu_k(N)\ge L(N,k). $$ and pair it with: $$ U(N,k). $$ If: $$ L(N,k)>U(N,k) $$ holds for the entire family, only then is there genuine closure. --- # 84. Interface Between Minimality and Prime-Gap Theorems When $h=2$, all lower endpoints: $$ p $$ must satisfy: $$ p+2 $$ is composite. For general $h$, $$ p+h $$ is composite. Thus, any theorem guaranteeing: > Among certain lower Goldbach representation endpoints, there must be an endpoint that also forms a prime pair at gap $h$ this condition, would directly exclude a minimal counterexample. There is currently no sufficiently strong universal theorem of this kind. --- # 85. Why Bounded-Gap Results Cannot Be Applied Directly Even knowing that infinitely many prime pairs have gap: $$ \le H, $$ does not mean: > Every specific lower Goldbach endpoint set must contain such a pair. Thus, a global bounded-gap theorem cannot directly imply: $$ S_h(N_*)>0. $$ RG-SCP must maintain consistency of the target population. --- # 86. Population discipline Every external theorem must explicitly answer: - Which primes? - In what interval? - What proportion? - Is it compatible with the representation endpoint condition? - Is it for a fixed $h$? - Is it uniform in $N$? One cannot take: $$ \exists^\infty p $$ as a direct substitute for: $$ \forall N_* \exists p\in\mathcal E_h(N_*). $$ --- # 87. Minimality as additive-history source Prime-Surface Insufficiency emphasizes the need for new sources. The source provided by minimality is not a new primality predicate. It provides: $$ \boxed{ \text{additive history below }N_*. } $$ That is: $$ \forall M0. $$ This is prefix information from the target problem itself. --- # 88. Is Prefix Information Sufficient? Unknown. It is possible that: $$ \boxed{ \text{all prior even sums succeed} } $$ is still insufficient to force: $$ N_* $$ to succeed. But this article at least turns prefix information into: - occupied lower diagonals; - smooth-shift transport graph; - matching demand; - AP congestion. Thus, it is no longer merely the semantic statement that “all smaller cases hold.” --- # 89. Prefix serialization If studying a hypothetical: $$ N_*, $$ One can retain: ```yaml minimal_counterexample_prefix: N_star: all_even_below_success: assumed_by_minimality selected_shift_scales: - k: M_k: shifts: transport_graph: left_count: right_count: max_right_degree: matching_lower_bound: matching_certificate: transport_demand: obstruction_capacity: closure_gap: ``` --- # 90. Formalization targets The elementary core of this article is well suited to formalization. ### L1 minimal counterexample implies smaller-even success. ### L2 lift-blocking: $$ (p,q)\in\mathcal G(N_*-h) \Rightarrow p+h,q+h \text{ composite}. $$ ### L3 shift self-suppression: $$ M_k\mid h \land p>p_k \Rightarrow \gcd(p+h,M_k)=1. $$ ### L4 smooth-shift deep obstruction. ### L5 transport graph left-totality. ### L6 edge-to-hard-survivor theorem. ### L7 matching demand: $$ D_{N_*}(p_k)\ge\nu_k. $$ ### L8 bounded-degree matching lower bound. ### L9 AP right-degree envelope. ### L10 minimality demand–capacity closure theorem. --- # 91. External theorem isolation If one uses: - Hall; - Brun–Titchmarsh; - explicit PNT in AP; - matching theorem; - sieve bounds; the interfaces should be separated. Hall and the basic matching theorem can be directly formalized. Analytic number-theoretic bounds go into a separate external interface. --- # 92. Source provenance graph ```text minimality ↓ all lower even layers successful ↓ smooth shift selection ↓ transport graph left-totality ↓ matching demand ↓ rough complements at N* ↓ deep obstruction demand ↓ capacity upper bound ``` and: ```text smooth shift collision ↓ right degree ↓ prime AP congestion ↓ matching lower bound ``` --- # 93. The Exact Theorem Group of This Article The central exact statements currently established by this article: ## T1 If: $$ (p,q)\in\mathcal G(N_*-h), $$ Then: $$ p+h,q+h $$ are both composite. ## T2 If: $$ M_k\mid h $$ and: $$ p>p_k $$ is prime, Then: $$ \gcd(p+h,M_k)=1. $$ ## T3 If there is also failure, Then: $$ \operatorname{lpf}(p+h)>p_k. $$ ## T4 A transport edge: $$ h\sim q $$ implies: $$ q\in\mathbb H_{p_k}(N_*). $$ ## T5 $$ D_{N_*}(p_k) \ge \nu_k(N_*). $$ ## T6 $$ \nu_k(N_*) \ge \left\lceil \frac{ |\mathcal H_k(N_*)| }{ \Delta_k^R(N_*) } \right\rceil. $$ ## T7 A minimal counterexample must satisfy: $$ \nu_k(N_*) \le U_{N_*}(p_k) $$ for any admissible capacity upper bound. --- # 94. What This Article Has Still Not Proved This article has not proved: $$ \nu_k(N_*) > U_{N_*}(p_k). $$ Nor has it proved that the Hall condition holds universally. Still less has it proved that right congestion must be sufficiently small. Thus: $$ \boxed{ \text{transport architecture} \neq \text{Goldbach proof}. } $$ --- # 95. Falsifiability Conditions ### F1 The right congestion of the transport graph can be large enough to make matching demand almost ineffective. ### F2 Smooth-shift rough complements bring no stronger provable lower bound than the original demand. ### F3 The AP degree bound is too loose. ### F4 Local capacity remains far greater than matching demand. ### F5 Hall expansion generally fails in the relevant family. ### F6 All shift designs are merely recodings of the original complement obstruction. If F1–F6 persist over the long term, the minimality transport line should stop expanding. --- # 96. Interface to the Next Article By the seventh article, RG-SCP already has: - counterexample state; - obstruction cover; - wheel geometry; - fixed-point dual-hole; - heuristic pressure; - demand–capacity inequality; - minimality transport demand. The next article should naturally perform: # RG-SCP 08 ## Reverse Closure Synthesis: Source-Class Audit, Formalizable Core, and First Experimental Protocol ### Reverse Closure Synthesis: Source-Class Audit, Formalizable Core, and First Experimental Protocol Rather than adding much more theory, it organizes: 1. which results are exact theorems; 2. which are merely equivalent reformulations; 3. which are heuristics; 4. which sources repeat; 5. which genuinely merit Lean; 6. what to run in the first computation round; 7. which stop conditions should lead us to new sources. If continuing is still worthwhile after the eighth article’s organization, then decide whether to open: $$ 09 $$ and: $$ 10 $$ for deeper analytic / spectral source escape. --- # 97. Conclusion The fourth article tried, from: $$ N_* $$ to transport failure downward, and found that the naive fixed-point shadow yields only smaller successful states. The seventh article changes direction: $$ \boxed{ \text{Do not move failure downward; turn lower successes upward into pressure.} } $$ If: $$ N_* $$ is the minimal counterexample, then every smaller even integer: $$ N_*-h $$ has a Goldbach representation. Every representation: $$ p+q=N_*-h $$ must satisfy: $$ p+h,\ q+h $$ are composite. when: $$ M_k\mid h $$ holds, If: $$ p>p_k, $$ we further have: $$ \boxed{ \operatorname{lpf}(p+h)>p_k. } $$ Thus, smooth-shift lower success generates, at: $$ N_* $$ deep obstruction demand. The transport graph organizes all smooth shifts. Minimality guarantees that every left layer is occupied, while matching controls how many distinct rough complements are forced to arise: $$ \boxed{ D_{N_*}(p_k) \ge \nu_k(N_*). } $$ If collision is too severe, arithmetic-progression congestion must also be paid: $$ p \equiv N_*-q \pmod{M_k}. $$ Thus, a hypothetical minimal counterexample is squeezed between: $$ \boxed{ \text{large transport demand} } $$ and: $$ \boxed{ \text{large AP congestion} } $$ these two. Finally, connecting with the sixth article: $$ D_{N_*}(p_k) \le U_{N_*}(p_k), $$ This gives: $$ \boxed{ \nu_k(N_*) \le U_{N_*}(p_k). } $$ Thus, the genuine minimality closure target is: $$ \boxed{ \text{transport matching lower bound} > \text{deep obstruction capacity upper bound}. } $$ This is the first time that RG-SCP has put: - minimal counterexample; - lower Goldbach success; - primorial smoothness; - bipartite matching; - arithmetic progression congestion; - obstruction capacity; into the same proof architecture. --- # 98. Shortest Version If only one sentence is retained: > **If a minimal Goldbach counterexample exists, all smaller even integers must already have prime pairs; when these pairs are moved upward along primorial-smooth shifts, the shifted prime endpoints must all become composites avoiding the first $k$ small prime factors, so minimality generates a deep-obstruction demand countable by matching.** The central chain: $$ \boxed{ N_* \text{ minimal failure} } $$ $$ \Downarrow $$ $$ \boxed{ \forall h\in\mathcal H_k(N_*), \quad \mathcal G(N_*-h)\neq\varnothing } $$ $$ \Downarrow $$ $$ \boxed{ \mathcal T_k(N_*) } $$ $$ \Downarrow $$ $$ \boxed{ D_{N_*}(p_k) \ge \nu_k(N_*) } $$ $$ \Downarrow $$ $$ \boxed{ \nu_k(N_*) \le U_{N_*}(p_k) } $$ The actual target to attack is: $$ \boxed{ \nu_k(N_*) > U_{N_*}(p_k). } $$ Once this is proved at some scale, the minimal counterexample cannot exist. --- # Appendix A: Canonical Transport Objects $$ \mathcal G(M) = \left\{ (p,q)\in\mathbb P^2: p+q=M \right\}. $$ $$ \mathcal R^{-}(M) = \left\{ q\in\mathbb P: q\le M/2, \quad M-q\in\mathbb P \right\}. $$ $$ \mathcal H_k(N) = \left\{ h: M_k\mid h, \quad h\ge M_k, \quad N-h>2p_k \right\}. $$ $$ h\sim q \Longleftrightarrow q\in \mathcal R^{-}(N-h). $$ $$ \nu_k(N) = \nu(\mathcal T_k(N)). $$ --- # Appendix B: Minimality Demand Chain $$ N_* = \min\mathcal C_G $$ $$ \Longrightarrow $$ $$ \forall h\in\mathcal H_k(N_*), \quad \mathcal R^{-}(N_*-h)\neq\varnothing $$ $$ \Longrightarrow $$ $$ \deg_L(h)\ge1 $$ $$ \Longrightarrow $$ $$ \nu_k \ge \left\lceil |L_k|/\Delta_k^R \right\rceil $$ $$ \Longrightarrow $$ $$ D_{N_*}(p_k)\ge\nu_k. $$ --- # Appendix C: Closure Certificate Schema ```yaml rg_scp_minimal_transport: version: N_star: minimality_assumption: scale: k: p_k: M_k: smooth_shifts: definition: count: bound: transport_graph: left_vertices: right_vertices: edge_rule: left_total: max_right_degree: degree_upper_bound: matching_lower_bound: matching_certificate: transported_demand: hard_survivor_subset: demand_lower_bound: obstruction_capacity: cutoff: p_k capacity_upper_bound: bound_source: closure: transport_lower: capacity_upper: strict_gap: minimal_counterexample_excluded: source_audit: external_theorems: formalization: provenance: checkpoint_hash: ``` --- # Appendix D: Formalization Roadmap ```text RG-SCP-07-Core ├── MinimalCounterexample ├── LowerEvenSuccess ├── GoldbachPairFiber ├── LiftOperators ├── LiftBlocking ├── ShiftSelfSuppression ├── SmoothShiftDeepObstruction ├── ReducedRepresentationEndpoint ├── TransportGraph ├── LeftTotality ├── EdgeToHardSurvivor ├── MatchingDemand ├── BoundedDegreeMatching └── MinimalityCapacityClosure RG-SCP-07-Interfaces ├── APDegreeUpperBound ├── HallExpansion ├── LocalCapacityUpperBound └── CrossScaleClosure ``` --- # Appendix E: Research Stop-Loss Conditions If subsequent verification shows: 1. $\Delta_k^R$ is almost of the same order as $|\mathcal H_k|$, with no gain in matching demand; 2. Hall expansion fails severely at every useful scale; 3. the transport lower bound is far below the demand lower bound already available in the sixth article; 4. the AP congestion upper bound cannot be improved; 5. smooth shifts only cause source reindexing, without new family-level constraints; 6. local capacity always swallows transport demand; then conclude: $$ \boxed{ \text{minimality-transport source stalled}. } $$ At this point, the eighth article’s source-class audit should formally decide: - to stop deepening this line; - or to turn to pair-spectral / generating-function / nonlocal sources.