Two Kinds of Identity for "1/2" The Divide between Static Algebraic Equivalence and Dynamic LayerGenerating Structure
In the conventional treatment of number and figure, fractions such as 1/2, 2/4, 3/6 and 4/8 are reduced to the single scalar 1/2 and are then regarded as identical objects written in different ways. This paper questions a long-taken-for-granted premise: the “identity” established by an equivalence class is not the same kind of identity as that of “1/2 as an object that keeps growing along the denominator layer”. Two senses of 1/2 are distinguished. The first is the static algebraic scalar obtained after the reduction projection, which lies on the critical line Re(s)=1/2 and unfolds continuously along the imaginary direction. The second is the real generated curve that grows along the discrete denominator-layer direction inside the numerator–denominator plane. The two unfold in orthogonal ways and therefore cannot coincide geometrically. We further show that reduction is essentially a dimension-reducing projection: it erases the signed offset term together with the whole dimension of the denominator layer, leaving only the limit of the generating curve. An analytic illustrative model of the real trajectory, Y = 0.5X − 0.8X^(-0.6), is given, indicating that the ideal line is only the lower asymptotic limit of this curve rather than the curve itself, and the directionality of layer information under conjugation is discussed. Chapter 7 presents an independent modular-iteration demonstration: six rational points from one equivalence class, fed into the same discrete operator that reads both numerator and denominator, fall into period-2 limit cycles whose amplitudes are determined by the denominator layer and are not interchangeable across grids — reproducing, at the level of operation, the statement that algebraic equivalence does not entail dynamical equivalence. This operator is independent of the illustrative model in Section 3; they do not share a formula. The conclusion is that algebraic equivalence does not guarantee dynamical equivalence, and that substituting the “line through the marked points” for the “real numerical value” is precisely the essence of projection dissolving the layer structure. On this basis the paper locates the issue further: 1/2 is the only non-substitutable quantity in the ζ-framework (the unique fixed point of ξ(s)=ξ(1−s), i.e. of the involution s 1−s). It is therefore not one parameter among many but the single fulcrum on which the whole ↦ framework is positioned. Whether it is read as an alignable coordinate point or as the asymptotic limit approached by the whole family determines whether zeros stand to it in a relation of alignment or of convergence — and hence whether conclusions can only be proportional (density) rather than universal. The paper further distinguishes the point-coordinate perspective from the generative-spectrum perspective, and argues for the necessity of clarifying the ontological identity of this object: no amount of research confined to a single coordinate axis can automatically elevate conclusions obtained within that axis into a global proof over the full domain. This distinction also constitutes a methodological warning for current numerical practice around the ζ function: replacing primitive lattice points carrying denominator lineage by their reduced scalar is not an information-neutral substitution. The complete mapping from generative lattice points to the complex variable s is left for future work.
Authors
- Xiangsheng Yu
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-14
- DOI
- https://doi.org/10.5281/zenodo.24212544
- Primary Topic
- Analytic Number Theory Research
- Type
- preprint