Corpus data and theoretical implications
The chapter aims at showing the usefulness of quantitative, corpus-based investigations to get a better understanding of a word formation procedure, taking as a case study the Italian VN compounds found on a large newspaper corpus. Corpus data make it possible to compare the weight of the process with respect to competing derivational strategies, and to evaluate the relevance of tendential phonological and morphological restrictions. Moreover, the hapax data, being a powerful tool to distinguish empirically between production and lexicalization/storage, may offer a new insight into the long-debated issue of the output category of this word-formation rule. It is argued that while the Unitary Output Hypothesis may basically hold for the lexicalized items, it cannot be adequate to describe the formation procedure itself.