Causal explanation, intentionality, and prediction: Evaluating the Criticism of "Deductivism"

Carsten Allan Koch

    Research output: Contribution to conferencePaperResearch

    Abstract

    In a number of influential contributions, Tony Lawson has attacked a view of science that he refers to as deductivism, and criticized economists for implicitly using it in their research.

    Lawson argues that deductivism is simply the covering-law model, also known as the causal model of scientific explanation (section 3). In the paper we refer to that model as that of causal explanation. Since this is the most generally accepted model of scientific explanation, this criticism has quite far-reaching implications.

    We argue that Lawson’s criticism of the causal model of scientific explanation is mistaken (section 4-5). However, it is admitted that one type of laws (whether singular or universal) are implicitly assumed in causal explanation in order to make them applicable to situations with more than one effect. The need for this kind of statement is exemplified by a simple physical experiment.

    Lawson also critisizes the use of universal laws in social science, especially in economics. This view cannot be as easily dismissed as his general criticism of causal explanation. We argue that a number of arguments often used against the existence of (correct) universal laws in the social sciences can be put into question.

    First, it is argued that entities need not be identical, or even remotely alike, to be applicable to the same law. What is necessary is that they have common properties, e.g. mass in physics, and that the law relates to that property (section 6).

    Second, one might take the so-called model of intentional explanation to be a candidate for being a universal law in social science. It is argued, against Popper himself, that this model fulfils Popper’s famous criterion for the demarcation of science and metaphysics, the falsifiability of the former (section 7).

    A third point of discussion concerns the question of whether the existence of free will excludes the possibility of prediction of behaviour by scientific or other methods. It is argued that, at least for an example, free will does not necessarily imply that the possibility of prediction of behaviour is ruled out. This section is, however, much less developed than the other results of the paper.
    Original languageEnglish
    Publication date2001
    Number of pages21
    Publication statusPublished - 2001

    Cite this