Is causal bracketing an instance of non-associativity, or an interpretive gloss?

Speculative, and deliberately framed skeptically. This is the single most seductive claim in the first exploration — which is exactly why it gets the mechanism test rather than a place on the map. The mining pass admits it was never argued physically.

The claim. That "causal bracketing" — the grouping of events by temporal or causal order — is an instance of non-associativity, the failure of (ab)c = a(bc), with the associator measuring the causal reordering.

The mechanism test is the gate. Name the shared theorem or admit it is a gloss. Is there a genuine non-associative structure — an associator that is a class in H³, or an alternative/Moufang algebra — whose associator is the causal bracketing, so that causal reordering and the associator are one object by a theorem? Or is it a resemblance, two things that both "regroup," spoken of as one with no shared theorem?

Sealed skeptical prediction. Absent an exhibited associator tied to causal order by a theorem, this is a cross-reference — a gloss — not an edge: relation: undetermined, leaning gloss. The seductiveness is the reason for the caution, not evidence for the claim; a design story that feels complete is evidence you are good at design stories (docs/ENGINEERING-LEG.md, the anthropic trap). What would settle it: a non-associative algebra or an H³ obstruction whose structure is provably the causal bracketing — or the honest admission that the appeal was interpretive.