Types of experiments and causal process tracing: What happened on the Kaibab Plateau in the 1920s

• Published in: Studies in history and philosophy of science. Part A Vol. 78; pp. 98 - 104
• Main Author: Millstein, Roberta L.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.12.2019; Elsevier Science Ltd
• Subjects: Causal process tracing; Causality; Causation; Community ecology; Deer; Ecological effects; Ecologists; Ecology; Experiment; Experiments; Methodology; Questions; Realism; Sequences; Social sciences; Tracing; Trophic cascade; Typological analysis; Typology; Kaibab Plateau; Ecology; Causal process tracing; Methodology; Experiment; Trophic cascade; Causation
• ISSN: 0039-3681; 1879-2510
• DOI: 10.1016/j.shpsa.2019.04.001

In a well-cited book chapter, ecologist Jared Diamond characterizes three main types of experiment performed in community ecology: laboratory experiment, field experiment, and natural experiment. Diamond argues that each form of experiment has strengths and weaknesses, with respect to, for example, realism or the ability to follow a causal trajectory. But does Diamond's typology exhaust the available kinds of cause-finding practices? Some social scientists have characterized something they call “causal process tracing.” Is this a fourth type of experiment or something else? I examine Diamond's typology and causal process tracing in the context of a case study concerning the dynamics of wolf and deer populations on the Kaibab Plateau in the 1920s, a case that has been used as a canonical example of a trophic cascade by ecologists but which has also been subject to controversy. I argue that ecologists have profitably deployed causal process tracing together with other types of experiment to help settle questions of causality in this case. It remains to be seen how widespread the use of causal process tracing outside of the social sciences is (or could be), but there are some potentially promising applications, particularly with respect to questions about specific causal sequences. •Causal process tracing is more of a practice than it is a specific method.•Causal process tracing can be used in conjunction with different experiment types.•Different cause-finding practices can be combined to increase causal confidence.•The wolf-deer populations of the Kaibab in the 1920s demonstrate these claims.•Causal process tracing can be profitably deployed outside the social sciences.