Extending Explanatory Relevance

Several authors (Kvanvig, 2003; Elgin, 2017; Dellsén, 2020) claim that there is a mode of objectual understanding whereby an agent understands a phenomenon, field, or subject matter without any explanation. A key set of cases for which it has been argued that there can be no explanations while they can be understood objectually concern the contrasting behaviours of indeterministic systems: If mutually exclusive behaviours of such a system are equally probable, neither behaviour can be explained (Kvanvig, 2003). Following Hitchcock (1999), Khalifa (2013) objects that explanatory factors need to make probabilities change but not necessarily differ from each other. However, Dellsén (2020) proposes a scenario wherein even this intuition seems to fail. We here argue that the scenario actually shows that explanatory relevance needs to be extended: If a particular factor comes out as explanatorily irrelevant due to either (i) coarse-graining of the explanans or (ii) fine-tuning of the explanandum, it should still count as explanatorily relevant. We will support this intuition by realistic examples from quantum optics. We conclude that, to the extent that the relevant sorts of indeterministic cases can be understood at all, they can be understood by means of explanation.