The Causal Mechanical Model of Explanation

Citation: James Woodward (1989) The Causal Mechanical Model of Explanation. Scientific Explanation. Minnesota Studies in the Philosophy of Science, volume 13, eds. Philip Kitcher and Wesley Salmon, 357-383 (RSS)
Internet Archive Scholar (search for fulltext): The Causal Mechanical Model of Explanation
Download: http://www.mcps.umn.edu/philosophy/13 11Woodward.pdf
Tagged:

Summary

In the history of scientific explanation, Wesley Salmon’s ‘scientific explanation and the causal structure of the world’ (SE) became the most interesting model after Hempel’s ‘aspect of scientific explanation’. Salmon’s causality contains three basic aspects, causal process, causal interactions, and conjunctive forks. And he suggests ‘two-stage affair’ of SE, the stage of assembling information and of providing a causal account when it comes to process and interactions.

The author understands and agrees with some part of Salmon’s model but disagree with other parts. There are four main questions the author brings up. (1) Will SE meet the requirements of causal/mechanical model? (2) Does Salmon’s model capture all features for assessment of SE? (3) Should we take statistical theories and is Salmon’s statistical evidence relevant to the construction of explanation? (4) Should we abandon epistemic conceptions and follow Salmon’s ontic one?

And the author’s answers are following. (1) Salmon’s concept seems to fit and he has captured a lot of the central features of the notion of causation but it cannot explain some cases even in physical context which does not have causal process such as equation or geometrical structure and which is in elementary quantum mechanics. (2) The author thinks that it is doubtful that Salmon’s model captures all features for assessment of SE because there are exceptional cases such as individual causal processes and it does not emphasis on the rule of inferential consideration. (3) Salmon thinks that the statistical theories explain approximate relative frequencies in large numbers outcome and the author agree this part. But Salmon does not believe the possibility or validity of some statistical theory such as QM since he thinks that the paradigms of statistical theorizing involve low-level of phenomenological generalization but there are good examples such as Schrodinger equation. (4) The author insists that Salmon’s practical conception has two problems. There is a tendency which copes with lots of distinct questions together. And some failure to come to terms with apparent implication is found in his discussion.