The social science literature is full of discussions about causation and what the best method for causal inference might be. However, a relatively small percentage of them draw on the philosophical debate about causation. Certainly, there is a great deal of talk about philosophy of science on the ontological and metaphysical level, including, for example, engagement with the relation between neo-positivism and realism. What I have in mind are what are called theories of causation in philosophy of science that ask what causation is, what makes a causal statement true and how to detect a causal relationship in practice.
The book, Causation: A Very Short Introduction, by Stephen Mumford and Rani Lill Anjum, is a useful entry point for those who want to learn more about the theories that are on offer. The book has ten chapters that are presented under the headings of “The Problem”, “Regularity”, “Time and Space”, “Necessity”, “Counterfactual Dependence”, “Physicalism”, “Pluralism”, “Primitivism”, “Dispositionalism” and “Finding Causes”. Chapters 2 through 9 are conceptualizations of what causation is, while the last chapter discusses strategies for finding causal relationships.
Those who are familiar with the debate in social science will notice that the theories discussed in this book are more comprehensive, with regularity and counterfactual dependence (supposedly) capturing quantitative methods and physicalism (supposedly) covering process tracing (e.g., Beach/Pedersen in their process tracing book). Therefore, Causation is certainly appropriate for extending one’s horizon about theories of causation.
Readers who expect in-depth introductions to a specific theory are not well-served by the book, but this is not the goal of the “Very Short Introduction” series by Oxford UP. Beyond that, I had three concerns when reading the book. First, the authors could have been more generous with references to more specialized literature. At the end of the book, the authors discuss a few books for each theory that they discussed and offer their opinions on which is helpful, but for someone who is accustomed to in-text citations referring to specific aspects, this is not ideal.
Second, stylistically, the authors overstretch the use of questions throughout the book. This is particularly problematic because I lost sight of what questions were answered and which remained unaddressed. Fewer questions and the use of summarizing tables would have been beneficial in structuring the discussion.
Third, it was not always clear to me whether a theory was supposed to be ontological or methodological, i.e., useful for the detection of causal relationships. It became clearer when I reached the last chapter, which is on causal inference (not a term that is widely used in philosophy of science) and which mainly features Pearl’s account and Woodward’s interventionist theory. (Although the interventionist theory is close to the idea of an experiment, in my reading, Woodward conceives of it as a semantic, conceptual theory, which makes it a bit surprising that the authors present it as a methodological theory, a problem that underscores the importance of clarifying what type of theory one is talking about). Although the interventionist theory has some appeal, I asked myself whether all metaphysical and ontological theories are compatible with Woodward’s and Pearl’s account located on the type level that has much affinity with quantitative, large-n research. From a social science perspective, it would have been interesting to read more about the link between metaphysical, ontological and semantic theories on the one hand and methodological theories on the other but that might have exceeded the goal for a A Very Short Introduction.