The question of characterizing academic vocabulary has often been framed in a context that is purely determined by questions of language teaching. The aim in such approaches is to come up with a list of words for learners of English for Special Purposes. This book approaches this question from a more general, empirical perspective, focusing on medical vocabulary. Its main contention is that the characterization of medical vocabulary is much more complex than is suggested by a simple list. In a list, a threshold determines the borderline on a one-dimensional scale between what counts as medical vocabulary and what fails to qualify as such. In analysing how such lists have been produced and how the cut-off point has been determined, the book shows a number of factors that have to be taken into account. It uses a comparison of two corpora to demonstrate the extent to which text type determines the outcome of frequency calculations. On the basis of such observations, it argues for a new methodology for the calculation of the degree of “medicalness” of lexemes.