Physics between math and biology — a formalization (with one correction)

A short note accompanying sciences_layers.py. It formalizes the sense in which physics sits between math and biology, corrects the word "proxy," and is explicit that it models a stylized and contestable picture, not a proven fact.

---

The correction: interface, not proxy

The request was to formalize physics as the proxy between math and biology. Built into the model is one honest change of word. In this toolkit a proxy is a measurable stand-in for an inaccessible truth, and its characteristic failure is decoupling from that truth (Goodhart). The math ↔ physics ↔ biology relation is not that. Physics is the layer where abstract mathematical structure acquires empirical content, and which is in turn the substrate biology is built on. That is an interface — a reduction intermediary, an abstraction layer — not a proxy. Physics does not stand in for math or biology; it is the place where the one becomes the other. The betweenness is real; its type is "interface," and calling it a proxy would import the wrong failure model.

What the formalization says

Three layers (math, physics, biology), each described by declared primitives (does it have empirical content? are its truths exactly provable? are its objects historically contingent?), are classified on four axes. On every axis, physics is the middle term:

Axis	math	physics	biology
abstraction	FORMAL	LAWLIKE	CONCRETE
verification	PROOF	MEASUREMENT	STATISTICAL
modal status	NECESSARY	NOMOLOGICAL	CONTINGENT
role	SOURCE	INTERFACE	APPLICATION

So "physics between math and biology" formalizes as: physics is the lawlike layer between pure form and organized matter; the measurement rung between proof and statistical inference; the nomological middle between the necessary and the contingent; and the interface where the formal source meets its living application.

The verification axis is the reachability spectrum

The verification axis is not a coincidence — it is the reachability-of-truth spectrum from the companion essay, laid across the sciences. Math is the exact pole: truths by proof, fully reachable. Physics is measurement: reachable to high precision against an independent reality. Biology is statistical and historical: contingent, harder to pin exactly (and one layer further — minds — is the unverifiable pole the qualia work addresses). So the layered sciences are, on their truth axis, ordered by how reachable their truths are, with physics again the middle rung.

Honest positioning — this is a stylized, contestable model

• It formalizes a picture, not a proven fact. The classical reduction hierarchy is disputed: Mayr argued for the autonomy of biology, emergence resists clean reduction, and Wigner's "unreasonable effectiveness of mathematics" names an unresolved puzzle about why math grips physics at all. The model asserts none of these are settled; it makes one common view explicit.
• The primitives are declared, coarse, and arguable. "Physics has no historical contingency" is itself contestable (the values of the constants, the particular universe). The classification is only as good as those declarations — which is exactly why it is stated, not smuggled.
• It classifies structure, not truth. It says which layer each domain occupies and why physics is the middle term; it makes no claim that any layer is more correct than another. Non-safety-critical scope; stdlib-only; deterministic; self-testing; built on taxonomy_builder.