Published online by Cambridge University Press: 01 January 2022
A common strategy for simplifying complex systems involves partitioning them into subsystems whose behaviors are roughly independent of one another at shorter timescales. Dynamic causal models clarify how doing so reveals a system’s nonequilibrium causal relationships. Here I use these models to elucidate the idealizations and abstractions involved in representing a system at a timescale. The models reveal that key features of causal representations—such as which variables are exogenous—may vary with the timescale at which a system is considered. This has implications for debates regarding which systems can be represented causally.
Thanks to Jim Woodward, Lauren Ross, and Ken Kendler for inviting me to participate in the PSA colloquium Strategies for Dealing with Causal Complexity. Thank you to the Pittsburgh Center for Philosophy of Science for funding me during the year in which I wrote this article and the Humboldt Foundation for funding me during the period when I revised it. The following people read earlier drafts of the article and provided useful comments: Colin Allen, Janella Baxter, Thomasz Bigaj, Dan Burnston, Kevin Hoover, Chungyoung Lee, Edouard Machery, Antonella Tramacere, Bill Wimsatt, Jim Woodward, Liying Zhang, and two anonymous reviewers. Finally, I am indebted to Adam Edwards, Jonathan Livengood, Shannon Nolen, and Karen Zwier for years of fruitful discussions about causality and time.