Social Causation

I would divide this book in three parts. The first part is a general philosophical analysis of causation, without any reference to social science. The author takes a stand against positivist views which were still dominant when the first edition of this book was written. Since then the tide has turned his way, but even though much of what he says is right, this first part of the book seems somewhat outdated.

The second part, which moves into the philosophy of social science, is much better. The author carefully sets out the criteria of meaningful causal investigation in the social sciences, emphasizing time and again the need to find specific differences between comparable situations. One of his main arguments is that a cause must be a show more relatively simple "precipitant" in an event - not an abstract social attitude or structure which lies beyond all observation. He accepts sceptical arguments, but points out that partial knowledge of social causation is still better than no knowledge at all. The discussion is rich in detail and shows a nice balance of theories and examples.

After concluding that statistical analysis only goes partway to causal explanation, the author uses the third part of the book to describe the idiosyncratic "dynamic assessment" by which the social investigator must finish the job. I don't think this part of the book was as interesting as the second part and at times I struggled to see its relevance. So this book contains some superfluous material, but it's still an original work that philosophically inclined social scientists should consider reading. show less