Same Meaning, Different Prose: Spine Preservation and Rendering Equivalence in Organizational Knowledge Work

This paper demonstrates Zharnikov’s (2026ao) Proposition P4 – rendering-equivalence under spine-preservation – in management theory, extending its Heisenberg–Schrödinger proof into strategy research. Two independently-authored pairs a dynamic-capabilities pair (Eisenhardt & Martin, 2000; Zollo & Winter, 2002) and a knowledge-based-view pair from the Strategic Management Journal Winter 1996 Special Issue (Grant, 1996; Liebeskind, 1996) – each return recombination metric Rec = 4; a random-graph null gives \(\Pr(\mathrm{Rec} \ge 3) \approx .000\) across 1,000 size-matched shadows. Three corpus-substrate renderings preserve 11/14, 4/4, and 12/15 items strictly within-operator and 14/14, 4/4, and 15/15 semantically, zero contradictions; under stronger cross-operator extraction (different model families), strict preservation is 9/14 and 10/15 on the two formal-apparatus renderings (4/4 on the focal-pair) and 14/14, 4/4, and 14/15 semantically, none. A bibliographic-hallucination audit of twelve AI-suggested anchors finds two verified, ten negative. Secondary β/δ estimates satisfy the cost-asymmetry ordering. Cross-language demonstrations span Russian (GigaChat Rec = 12, YandexGPT Rec = 11) and Chinese renderings across five LLMs from three training-corpus families, including an open-weights model running locally on a Mac mini (DeepSeek Rec = 12, Claude Opus Rec = 11, Qwen3.6:27b Rec = 12), with cross-extractor robustness (DeepSeek’s Chinese rendering re-extracted by Qwen3.6: Rec = 12). Inter-coder reliability is pre-registered for a future release. Includes zharnikov-2026ap-same-meaning-different-prose.yaml (Paper Spec v0.1.0) – a machine-readable specification of the paper's claims, assumptions, and dependencies. The paper's full machine-first bundle (the SPINE claim/dependency graph and the ONTOLOGY term module) lives in the public repository; see https://github.com/spectralbranding/paper-spec for the standard. This PDF is generated programmatically from that machine-first source under a research-as-repository model.