My article, “Interpretivism, First-Person Authority, and Confabulation” is a first step in developing a weakly naturalistic interpretation theory for the social sciences, which is consistent with interpretivism. I have been interested in figuring out how a Davidson-inspired interpretation theory can incorporate psychological theories about the imperfections of cognition, which seems to fly in the face of his principles of holism, charity and the presumption of first-person authority. The project has prompted me to study philosophical theories of self-knowledge, psychological experiments that demonstrate confabulation, and dual-systems theory within psychology.
In the social sciences, it is widely accepted that an adequate description of social phenomena must include the participating agents’ own understanding of their actions. Social actors are “self-interpreting animals” (Taylor 1985), whose beliefs and desires, values and preferences enter constitutively into what they do. Adequate descriptions of social action therefore require that social scientists engage in a “double hermeneutic” where the object is to interpret how knowledgeable agents conceive of their own actions (Giddens 1976). This approach has been coined interpretivism.
Davidson’s contention that a presumption of first-person authority is a necessary presupposition of interpretation can be used to motivate interpretivist social science. According to the presumption of first-person authority, people’s self-attributions of psychological predicates should at first, be interpreted as being true, without need of supplementary evidence, even if error and correction is possible.
In favorable circumstances, interpreting social behavior is relatively straightforward. If an individual is sincere, and if his verbal behavior matches his non-verbal behavior, an interpreter can use the verbally expressed beliefs and desires to give meaning to the non-verbal behavior, and the non-verbal behavior to confirm the interpretations based upon the verbal-behavior. Without a presumption of first-person authority, fine-grained interpretations of social behavior is impossible
However, psychological studies on the imperfections of cognition challenge the presuppositions of privileged and authoritative self-knowledge, which interpretivism and Davidsonian interpretation theory share. I concentrate on studies that demonstrate that people have a tendency to confabulate explanations of their behavior, because their conscious selves do not know why they do what they do, and therefore create the explanations that make most sense (Nisbett and Wilson 1977; Gazzaniga 2011; Hall et al. 2010; Johansson et al. 2005, 2006, 2012). In the light of such studies, Davidson’s interpretation theory and interpretivist social science seem to rest on a bad psychology.
In my article, I attempt to explain why confabulation is neither a threat to interpretivist social science, nor a threat to the presumption of first-person authority in Davidson’s interpretation theory. My argument depends on Henderson’s (1993) distinction between developing a first-approximation scheme and refined interpretation. Whereas the primary goal of first-approximation schemes is to attribute meaning to verbal and non-verbal behaviors, the goal of refined interpretation is to offer explanations by uncovering the mental processes that are the causes of actions. When developing a first-approximation scheme, Davidson’s principle of charity and the presumption of first-person authority serve as rather strong constraints on adequate interpretation. In refined interpretation, the interpreter might use psychological theory, and explain some beliefs or actions as seriously mistaken or irrational.
The reason why confabulation is a problem neither for interpretivist social science, nor to the presumption of first-person authority, is that the interpretative endeavor, which is necessary in order to identify and provide evidence for confabulation is governed by a presumption of first-person authority. Explanations of confabulation thus depend on prior interpretations.