Box 3.

Study 3—Historical studies to analyze the causal processes that created social–ecological traps

The goal of the study is to understand the causal production of poverty and other types of traps through historical and path-dependent processes (Boonstra and de Boer 2014). It is an example of causal reasoning that uses single, historical case studies and a qualitative approach to test whether specified mechanisms produce traps. The authors draw on the concept of path dependence (Mahoney 2001) and previous work about the importance of timing (Mahoney and Rueschemeyer 2003) to investigate if and how the timing of historical events contributes to the emergence and persistence of social–ecological traps. They use an ideal–typical representation of a path-dependent process to analyze and systematically compare the historical sequence of ecological, economic and political events that trigger self-reinforcing feedbacks in four exemplary cases using the method of process-tracing The authors pay much attention to specifying the causal configuration, i.e., the ideal–typical structure of a social–ecological trap, which is then tested in the four cases. They find that all four cases exhibit the causal aspects that are typical for a path-dependent process, namely the conditions that trigger a path-dependent process (antecedent conditions and a critical juncture) and a process through which the trap reproduces itself (Boonstra and de Boer 2014). These results across the cases support their claim that historical processes and the timing of mutually interacting events are critical for producing trapped situations. The causal reasoning involved in this study is rooted in a mechanism-based account of causation and claims are justified through demonstrating the effect of critical junctures of events for triggering a trap across all four cases. The authors highlight the importance of systematically investigating why the trap solidified in a particular point in time and using counterfactual reasoning to justify this claim. Furthermore, they argue that the identification of causal mechanisms facilitates generalization beyond an individual case, which helps avoid idiosyncratic “just-so” stories. However, more research is needed to further generalize findings and assess the importance of other processes, such as human agency for the emergence of a trap (Boonstra and de Boer 2014). Furthermore, the long-time horizon of this historical study facilitates studying the role of structural factors, but less so the role of micro-level processes such as social actions of specific individuals