Cycles of Generativity Exploitation: Stability, Distribution, and Procedural Evaluation

When exploitative relations close into a circuit, every participant can occupy both the affected and the beneficiary position. Aggregate capability can be unchanged or rising, roles can be symmetric, and the structure can reproduce itself indefinitely. This invites an inference that the paper's source discussion made explicitly and then withdrew: that a sufficiently reciprocal circulation cancels the exploitative character of its constituent relations, so that if everyone is exploited then no one is. The withdrawal was forced by a counterexample. In a closed cycle of robberies where each participant takes and is taken from in equal measure, aggregate holdings are unchanged and every constituent robbery remains wrongful. This paper reconstructs that dialectic and develops what survives it. It argues that distributive outcome, relational procedure, and dynamical stability are three independent evaluative registers, none of which determines another, and that stability in particular belongs to none of them: slave systems, monopolies, and predator–prey systems are all stable. It then dissolves the universal-participation paradox by locating an equivocation. The paradox assumes that exploitation requires a contrast class of persons, so that universal instantiation is self-defeating. The diagnostic's contrast class is a class of arrangements: each relation is assessed against an admissible feasible alternative, not against other participants. Universal instantiation is therefore coherent and describes a regime in which everyone would be better off under an admissible alternative while each still benefits through someone. Two formal results support the reconstruction. The edge predicate contains no aggregate term, so aggregate neutrality, symmetry, and closure are logically independent of it. And aggregate capability can strictly increase over time while every participant remains materially below an admissible alternative, because growth is measured against the past and the diagnostic against a feasible counterfactual. The paper then separates two structures that aggregate description conflates: a closed cycle in which every participant benefits through someone, and a systemic trap in which everyone is below a feasible alternative and no benefit-through witness exists anywhere. These are mutually exclusive and jointly non-exhaustive, so a third mixed class exists and is probably where most real configurations fall. They require different remedies: procedural intervention at edges for the first, and alteration of the feasible set for the second. The paper establishes no prevalence, classifies no observed regime, and supplies no aggregation rule for combining edge-level findings.