Giulia Schirripa

The summit of Mt Everest is determinately part of it; parts of the Khumbu Glacier are determinately not. In between are a range of candidate parts whose mereological status with respect to Everest is unsettled. Among the various accounts of vagueness, supervaluationism has received a great deal of attention. But very little attention has been paid to supervaluational theories of mereological relations themselves, and formal mereologies set within supervaluational logics are largely unexplored. In this paper, we provide two formal semantics for a vague mereology. We start with a simple supervaluational system using overlap as a primitive under a global consequence relation, but argue that it is unsatisfactory. We then develop a richer framework based on the idea of borderline connections, and expand it to a modal semantics of precisifications with a determinacy operator under a local consequence relation. We then prove a range of results, including the equivalence between borderline connection and indeterminate overlap, the vagueness of minimal and maximal fusions, and the distinction between mereological and compositional vagueness. The framework is intended to be compatible with a range of philosophical interpretations, including epistemic, semantic, and ontic views of vagueness.

The mereological account of social groups based on Classical Extensional Mereology (CEM) is notoriously associated with the reductionist stance that social groups are regular fusions identified by mereological conditions alone. This approach has many limitations and has suffered knock-down objections due to its transitive and coextensional features. The reaction to this impasse has been twofold. Philosophers have either moved away from CEM-based accounts altogether or attempted to resist these objections by appealing to extra-ontological explanations while retaining the reductive qualification of social groups. In this paper, I develop a non-reductive yet CEM-based framework that employs the notion of φ-parthood to restrict the interpretation of parthood simpliciter through extra-mereological, socially relevant conditions. Social groups turn out to be specified fusions and, depending on how the extra-mereological conditions are interpreted, the resulting account can be extensional or intensional. I show that both accounts successfully handle transitivity and coextensionality objections, and argue that long-standing objections from modal and temporal inflexibility are misguided for any CEM-based account.