Explaining New Phenomena in Terms of Previous Phenomena Rens Bod Abstract: It has become increasingly clear that natural phenomena cannot be formally deduced from laws but that almost every phenomenon has its own particular way of being linked to higher-level generalizations, usually via approximations, normalizations and corrections. This paper centers around the following problem: if every phenomenon has its own way of being explained, how can we -- or how can a theory -- explain any new phenomenon? I will argue that while particular explanations only apply to the specific phenomena they describe, parts of such explanations can be productively reused in explaining new phenomena. This leads to a new view on theory, which I call maximalism, according to which we are entitled to use all previously explained phenomena in understanding new phenomena. On the maximalist view, a theory is not a system of axioms or a set of models, but a dynamically updated corpus of explanations. New phenomena are explained by combining fragments of explanations of previous phenomena. I will give a specific instantiation of this view, based on a corpus of phenomena from classical and fluid mechanics, and show that the maximalist approach is not only used in scientific practice but that we cannot do without it in explaining real-world systems. This paper also aims to integrate three, somewhat disparate views: particularism, exemplarism and maximalism. Keywords: Philosophy of Science, Particularism, Exemplars, Maximalism, Data-Oriented Parsing