A Draft Term Model for ρ-Arithmetic: First Steps toward a History-Preserving Arithmetic Framework
First three papers of ZFCρ completed a diagnostic chain: remainder exists (Paper 1), remainder has quantitative identity (Paper 2), remainder is conserved (Paper 3). This paper takes the first step from diagnosis to construction: a draft term model for ρ-arithmetic.
Ordinary arithmetic (PA) takes natural numbers as extensional objects; operations output only results and discard process. ρ-arithmetic takes history terms as objects — formal terms encoding the complete generative path from zero to the current value. Operations output not just a new value but a new history term; old histories are absorbed into internal structure of the new term.
This paper defines: (1) recursive grammar for history terms; (2) evaluation function val(h) = n; (3) two-level equality — history equality (h₁ = h₂ iff term isomorphism) and extensional equality (h₁ ≡_E h₂ iff val(h₁) = val(h₂)); (4) ρ-addition and ρ-multiplication with merge semantics; (5) ρ-conservation as built-in constraint on operations.
This paper proves no number-theoretic theorem. Its goal is to make ρ-arithmetic a formal object that others can independently examine and compute with.
1. Introduction
1.1 From Diagnosis to Construction
The first three papers of ZFCρ completed a diagnostic chain: Paper 1 established the ineliminability of remainder (ρ-thesis), Paper 2 provided the quantitative identity of remainder (π as the first complete specimen, Euler's formula as "act binding remainder"), Paper 3 proposed the ρ-conservation principle (remainder is neither created nor destroyed under extensionalization).
All three papers repeatedly invoked a central concept: extensionalization discards operative history. But "operative history" remained informal throughout — an intuitive concept used to explain what ZFC's extensionality structurally omits, itself never formalized.
This paper's task is to turn "operative history" into a formal object. Specifically, it provides a recursive grammar that makes "how a number came to be" into a term structure that can be written down, compared, and operated upon.
1.2 Mirror Relation
ZFC carved an axiomatic system from the paradoxes of naive set theory, then constructed PA on top. ZFCρ carves a conservation principle from ZFC's remainder, then constructs ρ-arithmetic on top.
2. History Terms: Recursive Grammar
2.1 Term Grammar
History terms h are recursively generated by the following grammar:
The four constructors:
0 (zero). Primitive constant. Carries no substructure and is the ultimate leaf node of all history terms.
succ(h) (successor). One step forward from history term h. Corresponds to PA's successor function S(n)=n+1, but retains "where this step came from" — h is preserved as a subterm.
add(h₁, h₂) (addition). A new term produced by additive merging of h₁ and h₂. Both are preserved as subterms.
mul(h₁, h₂) (multiplication). A new term produced by multiplicative merging of h₁ and h₂. Both are preserved as subterms.
2.2 Evaluation Function val
Recursively define val: h → N:
val is the forgetful projection from history terms to natural numbers — it discards all operative history and retains only the extensional value.
2.3 Two-Level Equality
History equality =: h₁ = h₂ iff h₁ and h₂ are isomorphic terms — identical recursive structure, identical constructor sequences, identical subterms.
Extensional equality ≡_E: h₁ ≡_E h₂ iff val(h₁) = val(h₂).
Core example: Let h_a = succ(succ(succ(0))) and h_b = add(succ(0), succ(succ(0))). Then val(h_a) = val(h_b) = 3, so h_a ≡_E h_b. But h_a ≠ h_b — the former is a three-step successor chain, the latter is an additive merge of 1 and 2; their term structures differ.
The gap between the two equalities is ρ. PA has only one level of equality (extensional). ρ-arithmetic's additional history-equality level preserves the operative history that PA discards. The gap between the two levels — how many distinct history terms share the same extensional value — is the structured manifestation of the ρ discarded by extensionalization.
3. ρ-Addition
3.1 Definition
For history terms h₁ and h₂, ρ-addition is:
The output is a new history term add(h₁, h₂). h₁ and h₂ are preserved as subterms within the new term's internal structure — they are merged into the new term, no longer existing as independent top-level objects.
3.2 Relation to PA Addition
3.3 Algebraic Laws under Two-Level Equality
Commutativity: Extensional: h₁ ⊕ h₂ ≡_E h₂ ⊕ h₁. History: add(h₁, h₂) ≠ add(h₂, h₁) (left and right subterms are swapped; different term structure).
Associativity: Extensional: (h₁ ⊕ h₂) ⊕ h₃ ≡_E h₁ ⊕ (h₂ ⊕ h₃). History: add(add(h₁, h₂), h₃) ≠ add(h₁, add(h₂, h₃)) (different tree shapes: left-leaning vs. right-leaning).
This is not a defect; it is the core feature. PA's commutativity and associativity are properties under extensional equality ≡_E — they still hold in ρ-arithmetic at that level. They fail at the history-equality level precisely because that level preserves the operative information PA discards.
4. ρ-Multiplication
4.1 Definition
4.2 Relation to PA Multiplication
4.3 Why Not Expand as Iterated Addition
An alternative definition would expand h₁ ⊗ h₂ as h₁ ⊕ h₁ ⊕ ... ⊕ h₁ (val(h₂) times). This expansion has two problems:
First, it discards h₂'s history — only val(h₂) is used to determine the repetition count; the internal structure of h₂ vanishes. This violates ρ-arithmetic's basic commitment to preserving history.
Second, nested addition causes exponential blowup in term size.
The mul(h₁, h₂) definition avoids both problems: both h₁ and h₂ are fully preserved as subterms, and the term grows by only one layer (a single mul root node).
5. Primes in the Term Model
5.1 Multiplicative Root Terms in Hist(n)
For a natural number n, define Hist_mul(n) as the subset of Hist(n) containing terms whose root node is mul(h₁, h₂) with val(h₁) > 1 and val(h₂) > 1.
5.2 Term-Model Characterization of Primes
n > 1 is prime if and only if Hist_mul(n) = ∅.
That is: no history term with a non-trivial mul root can directly evaluate to n. In other words, n cannot be generated as the direct result of multiplicative merging of two values greater than 1.
Honest acknowledgment. This remains logically negative — characterizing primes by excluding all non-trivial multiplicative roots. It is a faithful translation of the standard prime definition into ρ-arithmetic language, not an independently new definition. But it translates the negation from "no factors exist" to "Hist space lacks a certain type of root-node term" — a structural signature that may serve as a starting point for future positive characterizations.
6. ρ-Conservation in the Term Model
Paper 3's ρ-conservation principle manifests in the term model as:
where ρ(h) measures the operative complexity of history term h, and ρ(op) is the operative cost of the operation itself.
7. Roadmap to Complete Axiomatic System
For ρ-arithmetic to reach ZFC-level foundational completeness, it must pass through four levels of formalization:
M1 Syntax Layer (this paper): Define the recursive grammar of history terms, the evaluation function val, two-level equality (= and ≡_E), and merge-semantics ρ-addition and ρ-multiplication. Provides the basic objects and core constructions.
M2 Axiom Schema (next step): List the complete axiom set of ρ-arithmetic, making it a self-contained formal system. Requires: history-term generation axiom, structural induction axiom, ρ-conservation axiom, extensional projection axiom.
M3 Model Theory (future work): Construct the standard model of ρ-arithmetic and prove its semantic consistency. Requires developing the structural theory of Hist(n): growth rate, generating functions, asymptotic behavior.
M4 Relative Consistency (future work): Prove Con(PA) ⟹ Con(ρ-arithmetic's extensional-projection fragment), or the stronger Con(ZFC) ⟹ Con(ZFCρ).
8. Conclusion
Paper 1 named the remainder. Paper 2 gave remainder its face. Paper 3 established remainder's law. Paper 4 provides a draft term model for an arithmetic that respects remainder.
This draft defines the grammar of history terms, the evaluation function, two-level equality, and merge-semantics addition and multiplication. It makes ρ-arithmetic a formal object that can be written down, inspected, and computed with — not merely a philosophical claim.
The draft proves no number-theoretic theorem. It does not claim to solve Goldbach or twin primes. It does not even claim to be stronger than PA. It does not provide a complete quantitative definition of ρ.
What it does is this: it turns the intuition "a number is not just a value but a history" into the starting point of a formal system with precise grammar.
How far one can go from this starting point is left to mathematicians.
← ZFCρ Paper III: ρ-Conservation
ZFCρ Paper V: Generation Axioms and Structural Induction →
ZFCρ Series · Mathematical Foundations · Back to Papers
ZFCρ前三篇完成了诊断：余项存在（Paper 1），余项有定量身份（Paper 2），余项守恒（Paper 3）。本文从诊断转向建设的第一步：给出ρ-算术的项模型草案。
普通算术（Peano算术，下称PA）的对象是自然数，运算只输出结果，丢弃过程。ρ-算术的对象是历史项（history term）——编码了从零到当前值的完整生成路径的形式项。运算不只输出新值，还输出新的历史项，旧历史被合并吸收进新项的内部结构。
本文定义：(1) 历史项的递归文法；(2) 求值函数 val(h) = n；(3) 两层等式——历史等式（h₁ = h₂ 当且仅当项同构）和外延等式（h₁ ≡_E h₂ 当且仅当 val(h₁) = val(h₂)）；(4) ρ-加法和ρ-乘法的合并语义；(5) ρ守恒作为运算的内建约束。
本文不在框架内证明任何数论定理。它的目标是让ρ-算术成为一个可被独立检验的形式对象——别人可以接着算的数学对象。
1. 引言
1.1 从诊断到建设
ZFCρ系列的前三篇论文完成了一条诊断链：Paper 1确立余项不可消除（ρ命题），Paper 2给出余项的定量身份（π作为L₁→L₂跃迁的第一个完整样例，欧拉公式作为"行为绑定余项"的结构恒等式），Paper 3提出ρ守恒原理（余项在任何外延化中既不被创造也不被消灭）。
三篇论文反复使用同一个核心概念：外延化丢弃操作历史。但"操作历史"在前三篇中始终是非形式的——它是一个直觉概念，用来说明ZFC的外延性在结构上遗漏了什么，但它本身没有被形式化。
本文的工作是把"操作历史"变成形式对象。具体来说，本文给出一套递归文法，使得"一个数是怎么来的"成为可以被写下来、被比较、被运算的项结构。
2. 历史项：递归文法
2.1 项文法
历史项 h 由以下文法递归生成：
四种构造器的含义：
0（零）。原始常元。不携带任何子结构，是所有历史项的终极叶节点。
succ(h)（后继）。从历史项 h 向前走一步。对应PA中的后继函数 S(n)=n+1，但保留了"这一步是从哪里来的"——h作为子项被完整保留。
add(h₁, h₂)（加法）。从两个历史项 h₁ 和 h₂ 通过加法合并得来的新项。h₁ 和 h₂ 作为子项被保留在新项内部。
mul(h₁, h₂)（乘法）。从两个历史项 h₁ 和 h₂ 通过乘法合并得来的新项。h₁ 和 h₂ 作为子项被保留在新项内部。
2.2 求值函数 val
递归定义 val: h → N：
val 是从历史项到自然数的遗忘投影——它丢弃全部操作历史，只保留外延值。
2.3 两层等式
历史等式 =： h₁ = h₂ 当且仅当 h₁ 和 h₂ 是同构的项——相同的递归结构，相同的构造器序列，相同的子项。
外延等式 ≡_E： h₁ ≡_E h₂ 当且仅当 val(h₁) = val(h₂)。
核心例子。 设 h_a = succ(succ(succ(0)))，h_b = add(succ(0), succ(succ(0)))。则 val(h_a) = val(h_b) = 3，所以 h_a ≡_E h_b。但 h_a ≠ h_b——前者是三步后继的线性链，后者是1和2的加法合并，项结构不同。
两层等式之间的差距就是ρ。 PA只有一层等式（外延等式）。ρ-算术多出的历史等式层保留了PA丢弃的操作历史。两层之间的差距——同一个外延值有多少种不同的历史项——恰好就是外延化丢弃的ρ的结构化表现。
3. ρ-加法
3.1 定义
对历史项 h₁ 和 h₂，ρ-加法定义为：
输出是一个新的历史项 add(h₁, h₂)。h₁ 和 h₂ 作为子项被保留在新项的内部结构中——它们被合并进了新项，不再作为独立的顶层对象存在。
3.2 与PA加法的关系
3.3 代数律在两层等式下的行为
交换律：外延层成立：h₁ ⊕ h₂ ≡_E h₂ ⊕ h₁。历史层不成立：add(h₁, h₂) ≠ add(h₂, h₁)。
结合律：外延层成立：(h₁ ⊕ h₂) ⊕ h₃ ≡_E h₁ ⊕ (h₂ ⊕ h₃)。历史层不成立：add(add(h₁, h₂), h₃) ≠ add(h₁, add(h₂, h₃))。
这不是缺陷，是核心特征。 PA的交换律、结合律是外延等式 ≡_E 下的性质——它们在ρ-算术中仍然成立，只是成立的层级是外延层而非历史层。历史层不满足这些律，恰恰是因为它保留了PA丢弃的操作信息。
4. ρ-乘法
4.1 定义
4.2 与PA乘法的关系
4.3 为什么不展开为迭代加法
一种替代定义是把 h₁ ⊗ h₂ 展开为 h₁ ⊕ h₁ ⊕ ... ⊕ h₁（val(h₂) 次）。这种展开有两个问题：
第一，它丢弃了 h₂ 的历史——只用了 val(h₂) 来决定重复次数，h₂ 的内部结构消失了。这违反了ρ-算术"不丢历史"的基本承诺。
第二，嵌套加法会导致历史项的指数级膨胀。
mul(h₁, h₂) 的定义避免了这两个问题：h₁ 和 h₂ 都被完整保留为子项，项的大小只增长一层（加一个 mul 根节点）。
5. 素数在项模型中的刻画
5.1 根节点为乘法的历史项
对自然数 n，定义 Hist_mul(n) 为 Hist(n) 中以 mul(h₁, h₂) 为根节点（且 val(h₁) > 1，val(h₂) > 1）的历史项的子集。
5.2 素数的项模型刻画
n > 1 是素数，当且仅当 Hist_mul(n) = ∅。
即：不存在任何以非平凡 mul 为根节点的历史项可以直接求值为 n。换言之，n 不能作为两个大于1的值的乘法合并的直接结果被生成。
诚实承认。 这在逻辑形式上仍然是否定性的——通过排除所有非平凡乘法根来刻画。它是PA中素数定义（不存在非平凡因子）在ρ-算术语言中的忠实翻译，不是一个独立的新定义。但它把否定性条件从"不存在因子"翻译成了"Hist空间中缺乏某种类型的根节点项"——这是一种结构签名，可能为将来构造正面刻画提供出发点。
6. ρ守恒在项模型中的表现
Paper 3的ρ守恒原理在项模型中的具体表现是：
其中 ρ(h) 是历史项 h 的操作复杂度度量，ρ(op) 是运算本身的操作代价。
7. 走向完备公理系统的路线图
ρ-算术要达到ZFC级别的基础完备性，需要经过四个层级的形式化：
M1 语法层（本文）： 定义历史项的递归文法、求值函数val、两层等式（= 和 ≡_E）、ρ-加法和ρ-乘法的合并语义。提供基本对象和核心构造。
M2 公理模式（下一步）： 列出ρ-算术的完整公理清单，使之成为一个自足的形式系统。需要：历史项生成公理、结构归纳公理、ρ守恒公理、外延投影公理。
M3 模型论（未来工作）： 构造ρ-算术的标准模型，证明其语义一致性。需要建立Hist(n)的结构理论：大小的增长率、生成函数、渐近分布。
M4 相对一致性（未来工作）： 证明 Con(PA)⟹Con(ρ-算术的外延投影片段)，或更强的 Con(ZFC)⟹Con(ZFCρ)。
8. 结论
Paper 1命名了余项。Paper 2给出了余项的面貌。Paper 3确立了余项的定律。Paper 4给出了一个尊重余项的算术项模型的草案。
这个草案定义了历史项的文法、求值函数、两层等式、和合并语义的加法与乘法。它让ρ-算术成为一个可以被写下来、被检查、被计算的形式对象，而不仅仅是一个哲学主张。
草案没有证明任何数论定理。它没有宣称解决哥德巴赫或孪生素数。它甚至没有宣称自己比PA更强。它没有给出ρ的完整定量定义。
它做的事情是：把"数不只是一个值，还是一段历史"这个直觉，变成了一个有精确文法的形式系统的起点。
在这个起点上能走多远，留给数学家。