The Coordination–Consensus Continuum: A Conceptual Framework and a Formal Identifiability Result for the Attribution Problem in Social Listening

Social listening systems are increasingly asked to answer a question they are not well equipped to address: when a community converges on a narrative, does that convergence reflect authentic collective opinion or orchestrated influence? This paper argues that the dominant framing — coordinated versus organic — is a false dichotomy that degrades both measurement and interpretation. Surface-level behavioral signatures (volume spikes, semantic similarity, temporal synchronization, accelerated propagation) are systematically ambiguous: modern organic virality routinely reproduces patterns historically treated as coordination signals, while sophisticated coordinated operations deliberately reproduce the appearance of authenticity. Treating coordination and consensus as endpoints of a continuum, we position any episode along five partially independent dimensions: intentionality, control centralization, participant agency, deceptiveness, and temporal–structural signature. We then make the central claim precise: the attribution problem is not one of measurement error but of non-identification. Using a transparent generative model whose ground truth is known by construction, we show that the latent attribution dimensions are non-identified from the behavioral signature — an out-of-sample learner recovers at most a tenth of the variance in the most identifiable axis, while the observable trace is dominated (over 80%) by an environmental factor that is not an attribution dimension at all. Identifiability of centralization is high at the clean extremes and collapses by 86% in the hybrid interior, exactly where consequential episodes live. We close by converting the result into an identification strategy that specifies which external evidence resolves which dimension. The contribution is twofold: a conceptual reframing from classification to positioning, and a formal demonstration, with quantified magnitudes, of why single-signal attribution structurally cannot succeed.