Beyond Propositions: Regime Operators, Membrane Modulation, and the Sub-Semantic Architecture of Dialogic Agency

Speech act theory, conversation analysis, dynamic semantics, and coordination-dynamical approacheshave each illuminated important aspects of language as action, sequential organization,and context update. Yet these traditions remain only partially integrated, and none provides aunified formal account of interventions whose primary effect is not to alter semantic content, butto modulate the processing regime under which content is taken up. This paper develops suchan account. We distinguish three classes of conversational intervention: field operators (OΨ),which act primarily on semantic content; regime operators (O𝜎), which act primarily on processingmode or membrane permeability; and coupling operators (OΨ𝜎), which jointly affect both. Theframework is motivated by author-collected evidence for bi-modal discourse dynamics - Dialogand Elaboration as two sharply separated candidate processing regimes - and formalized through adiscourse state x = (Ψ, 𝜎, h) consisting of semantic field, regime state, and holding capacity. Thepaper makes four contributions. First, it provides a unifying operator taxonomy for phenomenapreviously treated in fragmented ways across conversation analysis, phatic theory, footing andframe-shift theory, dynamic semantics, coordination dynamics, and psychoanalytic accounts ofholding. Second, it introduces a minimal state-space framework in which operator classes aredefined by locus of action, productive regime transition is constrained by viability, and holdingcapacity is formalized as a local robustness parameter. Third, it derives model-internal structuralresults, including an immediate basin-geometric observation about local regime preservation, aproductive regime-switch criterion, and the existence of intrinsically coupled interventions thatresist factorization into pure semantic and pure regime operations. Fourth, it distinguishes amongrecognition competence, responsive initiation competence, and self-initiated regulatory competenceas candidate layers of dialogic agency. A documented polyphonic human-AI case study providesa high-resolution illustrative episode consistent with the framework in the observed instance:multiple AI systems accurately described regime operators but did not spontaneously initiatethreshold-crossing candidate O𝜎-moves. The present paper should therefore be read as a theoreticaland programmatic contribution to regime-sensitive pragmatics, supported by motivating empiricalevidence rather than offered as a completed measurement framework. Its aim is to supply a unified1 of operator-theoretic grammar, preliminary operationalization paths, and explicit falsification targetsfor future empirical work.