The Deductive-Nomological Model of Explanation
Fred Eidlin, DEDUCTIVE-NOMOLOGICAL MODEL OF EXPLANATION, Albert J. Mills, Gabrielle Durepos, Elden Wiebe, eds., Encyclopedia of Case Study Research, SAGE Publications, 2010
3 Pages Posted: 3 Nov 2011
Date Written: November 3, 2010
Abstract
The deductive–nomological model (CLM), often called the covering law model (CLM), or the Hempel–Oppenheim model is a widely accepted, yet widely-criticized account of scientific explanation. According to this account, the explicandum is some phenomenon requiring explanation. The explicans of the explicandum removes the puzzling or problematic character of the explicandum by showing the phenomenon to be a particular case of a known general law. The CLM has, from early on, been subjected to serious criticisms. The CLM enjoys a curious status in the social sciences. It is presented in social science methodology textbooks as an account of scientific explanation. Yet, few explanations in the social sciences come close to fitting the model. Even social scientists who subscribe to the model usually admit that there are, as yet, no universal laws in the social sciences or that there are only a few that might arguably qualify as such. Some social scientists argue that, when the social scienes have reached maturity, their explanations will conform to the CLM. In the meantime, it is acceptable for social scientists to make use of “theories of the middle range.” In recent years, several writers have drawn attention to CLM's mistaken assumption that covariance or regularity implies causality. Karl Popper’s account of deductive explanation is sometimes mistakenly identified with the CLM, Like the CLM, in Popper’s account of a scientific explanation the explicans of the explicandum consists of laws and initial conditions, and the explicandum must follow logically from the explicans. However, although this is a necessary condition for Popper it is not a sufficient condition. First of all, Popper agrees with those critics of the CLM who point out that correlation or covariance does not entail causation. Something must drive the explanation, must account for why the explicandum necessarily follows from the explicans. An explicans may eliminate the problematic character of a phenomenon in various ways. Subsuming the phenomenon under a general law, as prescribed by the CLM, is one possibility. However, even in such cases the cause of an observed regularity may be unknown and requiring explanation. Moreover, the problematic character of a phenomenon may also be removed by repairing or filling out initial conditions, or by reconceptualizing the situation.
Keywords: covering law model, deductive model, explanation
Suggested Citation: Suggested Citation