This Version 1.0 preprint develops a theorem-driven structural account of dyadic drift under non-coercive regulatory-channel loss. The paper examines how the absence of a load-bearing but non-demandable dyadic function — such as safe proximity, tactile contact, reliable emotional availability, or other forms of co-regulatory access — may shift an intimate dyad from a buffered regulatory regime toward stress-dominant regulation. The argument is intent-agnostic and non-moralising. It does not infer motive from impact, does not treat any intimate function as legitimately coercible or demandable, and does not require villains as explanatory primitives. Instead, it asks what follows structurally when a regulatory channel is absent over time and when self-regulatory, coordination, inference, and repair burdens become increasingly concentrated within individuals or the dyad. The paper begins from a minimal set of structural commitments: dyads can function as coupled regulatory systems; some non-coercive functions are practically load-bearing without being normatively enforceable; the absence of such functions increases private regulatory load; time-weighted exposure — duration, frequency, and variability of absence — affects whether a dyad remains buffered or shifts toward stress-dominant regulation; and sustained exposure can deplete repair capacity, destabilise calibration, and produce drift toward withdrawal, audit, or exit. From these commitments, the article derives a theorem architecture concerning regulatory-channel absence, exposure-governed regime shift, repair depletion, calibration instability, adversarial appearance without adversarial intent, and exit as a feasibility response. It develops the regulatory-regime family in full, specifying observable implications, negative implications, temporal-ordering expectations, challenge conditions, boundary conditions, and downstream empirical instantiation logic. The contribution is structural and diagnostic rather than clinical, moral, legal, or policy-prescriptive. It does not address coercive control, violence, or abuse as its target domain; where coercion is plausible, protective and adversarial frameworks may be necessary. Its narrower purpose is to provide a descriptive grammar for non-coercive dyadic drift: not “who is right?”, but whether the system can still hold under the distribution of load, buffering, exposure, and repair capacity. Within the wider research programme, the article functions as a support and bridge theorem text connecting dyadic systems, co-regulation, social buffering, structural viability, burden transfer, repair depletion, and the distinction between normative evaluation, measurable surfaces, and mechanism-level feasibility.
J. E. Fröderberg (Tue,) studied this question.