Conceptual analysis has a role to play in Translation Studies, but it is a means, not an end. An empirical paradigm gives central importance to testable hypotheses. Empirical research on translation profiles should result in a translation typology; one such typology is discussed. Translations have multiple causes, and we can already propose some possible causal laws. Three laws of translation effect are also proposed, and various parameters of effect are discussed, together with the associated problems of sampling and prescriptivism. The author argues that prescriptive statements are hypotheses about translation effects; as such, they should be tested like any other hypothesis.