Definition
U28: “it’s more like a hierarchy of levels at which observables exist”. A75: O = ∪_ℓ O_ℓ, indexed by levels: linguistic (word choice, syntax, meter, register), conceptual (referents, causal relations), pragmatic (force, irony, presupposition), interactional (footing, alignment, commitments), dynamical (responses to framing, trajectories, adaptation). “These aren’t necessarily disjoint or linearly ordered. An observable can participate in several levels, and the levels themselves might form a partial order.” A75: “The hierarchy belongs to how we can observe and distinguish communicative phenomena, not necessarily to the phenomena themselves.”
What it is not
Not a measurement context. The chunk-5 reader notes that C6 §2 orders sheaf contexts (sets of jointly measured observables) by inclusion and calls that the partial order, which is a different order from the owner’s levels. The manuscript §2 keeps the levels without the order: “These observations are not ordered into one strict ontological ladder.” The dynamical level is not in the manuscript’s list.
Note of 2026-10-21: levels as a partial order, again
A7 §VII, A8 §4 and A9 §8 of the exchange of 2026-10-20 each separated orders that “levels” runs together. A7 named representational, predictive and operational relations. A8 named granularity, informativeness, computational complexity and explanatory scope. A8 concluded that “the organization of scientific explanations is better represented by a network or partial order than by a unique ladder of levels”. None cites this term, whose A75 already said “the levels themselves might form a partial order” and “The hierarchy belongs to how we can observe … not necessarily to the phenomena themselves”, or CLAIM-042. The rows differ each time, so nothing is filed.