Changes in the productivity of word-formation patterns: Some methodological remarks

Changes in the productivity of word-formation patterns are often investigated using hapax legomena. In this paper, I argue that at least in diachronic investigations of productivity, a measure based on first attestations is a viable alternative to hapax-based measures. I show that such a measure is a more direct proxy to new words than hapax-based measures - it measures what we want to measure, which is not always true for the latter. I present a method that deals with the common problem of varying subcorpus sizes (I suggest we randomly resample the subcorpora up to a predefined size), and to the problem of old words appearing as new at the start of the corpus (I suggest we take an earlier corpus and determine a point in time when almost all old words have registered). Armed with these instruments, we can determine the ratio of new types to existing types for a time span, which can be regarded as the renewal rate of the respective category.