Grounding and anchoring: on the structure of Epstein's social ontology

Brian Epstein's The Ant Trap is a praiseworthy addition to literature on social ontology and the philosophy of social sciences. Its central aim is to challenge received views about the social world - views with which social scientists and philosophers have aimed to answer questions about the nature of social science and about those things that social sciences aim to model and explain, like social facts, objects and phenomena. The received views that Epstein critiques deal with these issues in an overly people-centered manner. After all, even though social facts and phenomena clearly involve individual people arranged in certain ways, we must still spell out how people are involved in social facts and phenomena. There are many metaphysical questions about social properties, relations, dependence, constitution, causation, and facts that cannot be answered (for instance) just be looking at individual people alone. In order to answer questions about (e.g.) how one social entity depends for its existence on another, we need different metaphysical tools. Epstein thus holds that social ontological explanations would greatly benefit from making use of the theoretical toolkit that contemporary analytical metaphysics has to offer. He focuses specifically on two metaphysical instruments: grounding and anchoring. This paper examines Epstein's understanding and use of these tools. I contend that Epstein is exactly right to say that contemporary metaphysics contains many theoretical instruments that can be fruitfully applied to social ontological analyses. However, I am unconvinced that Epstein's tools achieve what they set out to do. In particular, I will address two issues: (1) How is grounding for Epstein meant to work? (2) Is anchoring distinct from grounding, and a relation that we need in social ontology?