PSA2016: The 25th Biennial Meeting of the Philosophy of Science Association

Full Program »

Prospects for a Theory of Evidence for Non-Quantitative Methods

Much of the recent methodologically oriented work in philosophy of the social sciences has focused on experimental methods and formal modeling, and is thus directly relevant for only a minority of social scientists. Much progress has been made regarding the foundations of statistical causal inference, but these developments do not help social scientists working with non-quantitative methods. Existing philosophical theories of confirmation are not really helpful for the practicing social scientist either, since they only describe what should happen to rational beliefs once evidence has been identified as relevant and its strength evaluated. I outline two inter-related challenges, together with some preliminary answers, that would make philosophy of social science more methodologically relevant. First, the discussion on social mechanisms should move from ontology and explanation towards testing and research heuristics (Little 2015). The causal ambitions of case-based qualitative research are now taken seriously in the methodological discussion within the social sciences and principles of process tracing for (mostly within-case) causal inference are being formulated and debated. Although much valuable philosophical work on process tracing in terms of causal mechanism has already been done (Reiss 2015; Runhardt 2015; Ruzzene 2014; Steel 2008), and there is a common understanding of the goals of case-based inquiry and of the need for formulating stronger epistemic principles for fulfilling those goals, the logic of process tracing has not really advanced in ways that would have provided new practicable methodologies. I briefly suggest ways of further regimenting the evidential role of the concept of social mechanism. Second, the empirical study of these causal relations and mechanisms almost always has to include a strong interpretive component, but the empirical study of meanings and normative practices is wrought with methodological issues largely untouched by current mainstream philosophy of science. Beyond a broad acknowledgment of the truth of causalism, how exactly should the relationship between folk-psychological attribution of intentional states and causal explanation be understood? More importantly, how can the reliability of these attributions, both those made by the researcher and those made by the studied subjects, be measured and improved? How should the explanatory role of the various meaning-concepts routinely attributed to larger social units such as practices, rituals and other social institutions, be understood? In what way can such attributions be seen as objective and how should their reliability be evaluated? I do not present any concrete answers to these questions, but instead argue that such answers can only be found from a thoroughly naturalistic perspective and that this perspective implies forgoing the idea of a specifically sociological theory of action.

Author Information:

Jaakko Kuorikoski    
University of Helsinki

 

Powered by OpenConf®
Copyright©2002-2015 Zakon Group LLC