Those working in experimental philosophy have raised a number of arguments against the use of conceptual analysis in philosophical inquiries. But they have typically focused on a model that pursues conceptual analysis by taking intuitions as a kind of (defeasible) evidence for philosophical hypotheses. Little attention has been given to the constitutivist alternative, which sees metaphysical modal facts as reflections of constitutive semantic rules. i begin with a brief overview of the constitutivist approach and argue that we can defend a role for conceptual analysis, so understood, in ontological disputes against both the general skepticism about the relevance of intuitions, and against the specific worries raised by experimental results. Finally, i argue that even if the constitutivist view is adopted, experimental philosophy may still have quite a useful role to play, though purely empirical inquiries cannot in principle do the ontological work alone.
What are the proper methods for attempting to resolve ontological and modal disputes-taken as disputes about what ontological sorts of thing entities of various kinds (properties, numbers, statues, symphonies) are, what their existence and persistence conditions are, and so forth? To what extent are these questions resolvable at all?
One venerable method of approaching such questions-pursued in ordinary language philosophy and (arguably) phenomenology-is to go by way of conceptual analysis, considering, for example, whether a variety of actual and hypothetical cases would be situations in which a symphony was created or the same person survived.1 Although conceptual analysis became less popular in neo-Quinean metaphysics, there have been several recent attempts at reviving and defending conceptual analysis as a method for ontology (Jackson 1998, Chalmers and Jackson 2001, Thomasson 2007, McGinn 2011). But conceptual analysis, as a method for doing work in philosophy, has come under fire. Some have raised a sort of blanket skepticism that our intuitions-whether we find out about them through introspective means, or through the empirical methods pursued by 'positive' programs in experimental philosophy-can tell us anything about mind-external features of the world (see Pust and Goldman 1998,179; Knobe and Nichols 2008, 6). More specifically, those engaged in a 'negative' program in experimental philosophy have found evidence of certain types of variations in people's intuitive judgments which, they argue, undermines the idea that conceptual analysis may play a useful role in philosophical inquiry-or at least (what is often taken to be the same idea) undermines the idea that intuitions can count as evidence in philosophical debates.
In this paper I address a number of objections to the idea that conceptual analysis may give us philosophical knowledge. For simplicity, I will focus just on disputes that arise in ontology, regarding the natures, existence, identity, and persistence conditions of things of various sorts. Those arguing against conceptual analysis have typically focused on a model that pursues conceptual analysis by taking intuitions as a kind of (defeasible) evidence for philosophical hypotheses. Little attention has been given to an alternative model of conceptual analysis, which we might call the 'constitutivist approach', which sees constitutive semantic rules as determining the modal features of the objects, if any, to which we refer. I have argued for and defended the constitutivist view elsewhere (though not under that name)2 and I will not have space to argue for it again here. Instead, the current project is a defensive one.
In section 1,1 will give a brief overview of the constitutivist approach and its consequences for methodology in ontology. In sections 2-4, I argue that we can defend a role for conceptual analysis, so understood, in specifically ontological disputes against both the general skepticism about the relevance of intuitions and against the specific worries raised by experimental results. …