dimensionlessOnlyClaim
plain-language theorem explainer
Documents the honest scope of Recognition Science in dimensionless-only calibration: all dimensionless ratios (inverse fine-structure constant, mass ratios, and kin) are claimed from φ alone, while any SI numerical value needs an external anchor such as CODATA ℏ. Verification and constants authors cite it when stating what the formalization does and does not assert. It is a fixed string definition, not a proved theorem.
Claim. Policy string for dimensionless-only mode: Recognition Science derives all dimensionless ratios ($\alpha^{-1}$, mass ratios, and similar) from $\varphi = (1+\sqrt{5})/2$ alone; SI numerical values require an external anchor (for example CODATA $\hbar$); no SI constants are claimed as internally derived in the current formalization.
background
The CalibrationPolicy module separates two layers of RS claims. Dimensionless predictions are ratios, exponents, and algebraic relations fixed by $\varphi=(1+\sqrt{5})/2$ with no external scale. SI-anchored predictions are ordinary SI numbers that need at least one external constant to set absolute units.
Dimensionless-only mode is the conservative policy: claim only what follows from $\varphi$, and treat every SI constant as an external calibration input. Single-anchor mode would fix exactly one SI constant (for example $\hbar$) and push the $\varphi$-ladder relations through to the rest. The module notes that Constants currently uses placeholders ($\hbar=1$, $G=1$, $c=1$), which is dimensionless-only practice: ratios are meaningful, SI numerics are not.
In RS-native units one often writes $c=1$, $\hbar=\varphi^{-5}$, $G=\varphi^5/\pi$, with $\alpha^{-1}$ in a narrow band near $137$; those native expressions still do not by themselves export CODATA SI digits without an anchor choice.
proof idea
Not a proof. The declaration is a string definition: three concatenated English sentences fixed at compile time. No tactics, no lemmas, no hypotheses. Sibling names in the module (DimensionlessPrediction, CalibrationMode, singleAnchorClaim, CalibrationCompliance) are the surrounding policy vocabulary; this binding is only the human-readable claim text for the dimensionless-only branch.
why it matters
It records the epistemic boundary the verification layer is willing to advertise: zero-parameter content is the $\varphi$-forced dimensionless sector (mass ratios on the $\varphi$-ladder, $\alpha$ band structure, eight-tick and $D=3$ structural facts upstream), not a claim that SI meter-kilogram-second digits fall out of the ledger with no external input.
That matches the module status note: either accept an external anchor (and drop pure "no external input" rhetoric) or formalize an internal absolute-scale mechanism, which is not done yet. Downstream use count is presently zero; the string exists so compliance and documentation surfaces can quote one canonical sentence instead of ad-hoc wording. It pairs with the single-anchor claim sibling as the other half of the policy fork.
Switch to Lean above to see the machine-checked source, dependencies, and usage graph.