This document discusses some of the conceptual challenges involved in big data practices, particularly regarding causality. It begins by questioning what causation means in the context of big data, where correlations are often taken as causal without a theoretical understanding. It then provides a crash course on traditional and pluralist philosophical views of causality. Several key questions for analyzing causality are identified, as well as levels of analysis. The document argues that conceptual challenges of causality have not changed with big data and advocates adopting a strategy like Bradford Hill's to make viewpoints on causality explicit based on data and other evidence, without treating them as rigid criteria. It suggests big data can help identify causes while resisting scientism and skewing of science and technocracy