Interventions and Mechanistic Hierarchies

• Applicant: Professor Dr. Stephan Hartmann, since 8/2022
• Subject Area: Theoretical Philosophy
• Term: from 2020 to 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 441311834

Context: Sciences such as biology, medicine, and neuroscience provide explanations in terms of mechanisms: They explain how systems behave by pointing at causal interactions among their parts. The theoretical details of how the parts responsible for these systems’ behaviors (the constitutively relevant parts) can be identified on the basis of experiments is not well understood yet and controversially discussed in the philosophy of science literature.Challenge: The main problem is that the notions of experimental manipulations predominantly used in the literature are not suitable for part-whole systems. The challenge consists in developing a unified framework for probabilistic reasoning across part-whole hierarchies that allows for identifying the parts relevant for a system’s behavior on the basis of notions of experimental manipulations suitable for such systems. Key idea: The project builds on preliminary work and tools from the causal modeling literature. The key idea is to develop notions of experimental manipulations that can affect a system’s overall behavior in virtue of influencing its parts and to supplement these notions with different kinds of background knowledge the experimenter can have about how exactly the system is manipulated.Project overview: Part (i) of the project will develop notions of interventions suitable for part-whole systems. Part (ii) will develop methods for identifying constitutively relevant parts on the basis of these notions and part (iii) an approach for measuring the strength of constitutive relevance relations. Part (iv) will use the account as a tool for unifying and evaluating other recent approaches on the market. Parts (i)-(iii) will be supplemented by a case study.Output: The project will lift the mechanism debate to the next level via producing the first unified framework for probabilistic reasoning and constitutive/causal inference under experimental manipulations of mechanistic hierarchies. It will produce new methodologies for identifying a mechanism’s relevant parts, a quantitative approach to mechanistic explanation, and will contribute to overcoming the controversies in the recent literature on mechanisms.

• DFG Programme: Research Grants
• Ehemaliger Antragsteller: Dr. Alexander Gebharter, until 7/2022