Researchers have developed a new approach to identifying the cause-and-effect relationships in complex systems using observational data. They propose a method for bounding the effect of interventions and counterfactuals by defining a canonical prior that has full support over the space of structural causal models with discrete observables. This allows them to estimate partially-identifiable causal effects, even when there is unobserved confounding. --- Why it matters: This work matters because it extends the capabilities of causal foundational modeling, enabling researchers to study complex systems where multiple factors contribute to an outcome. By developing methods for bounding counterfactuals, they can better understand how interventions might affect real-world systems. Source: https://arxiv.org/abs/2608.20841