Hitchcock, Christopher - Conceptual Analysis Naturalized: A Metaphilosophical Case Study

02/23/2007

The Journal of Philosophy, Vol 103, #9 Sep 2006

Author distinguishes between conceptual analysis that tries to find necessary and sufficient conditions of concepts, and a conceptual analysis that tries to locate the 'theoretical role' of a concept-- that is, tries identify its role 'in facilitating inferences of various kinds'. Author claims that this second task of conceptual analysis can be informed by empirical studies in psychology and elsewhere. He uses an example of causation as his 'case study' of how the studies in psychology will bear on philosophy.

Section I
Gives a survey of the theories of causation:
1. Regularity theories- causation just is constant conjunction (Hume)
2. Mechanical theories- causation is the transfer of a substance or concept (energy, momentum, etc.)
3. Manipulability theories- casual knowledge comes from our ability to affect the world
4. Counterfactual theories- if not for p, no q e.g. counterfactual dependence and so on

Section II
This section provides the basis for the usefulness of an analysis of a concept using it's theoretical role. While this kind of analysis doesn't provide us with a solid explanation of the concept, it will give a deeper understanding because it helps us understand when the concept is appropriately applied.

Section III
Psychology about causation: it appears primates/chimps have trouble with simple tasks where a robust concept of causation might be necessary, but children around the age of 3-4 don't have the same trouble. These studies seem scattered. From them, author tries to construct a 3-part theory of causal reasoning:
1) full causal understanding, where knowledge gained from watching others manipulate can be applied to regularities/machinations in the world
3) Egoistic causal understanding, where the agent can use causal forces only when it is done personally, by the agent himself-- only using manipulation
2) Social causal understanding, where the agent's understanding of causal forces may be influenced by what other creatures do, but may not be able to see the mechanistic causal factors.

The key claim here is that the different proposals for analysis of causality in section I are considered to be modes, or modalities, and chimps are seen as understanding the modality of regularity and possibly manipulation, but perhaps not of mechanical causation. The author uses the different accounts of causation to construct possible theoretical explanations for the scientific data. Thus conceptual analysis of the concept of causation is brought into the empirical realm of theory-- conceptual analysis 'naturalized'. Author also argues that psychology can help philosophy.