Etymological Features Across Genres and Registers

Abstract

The English language, ‘a conglomerate of many different origins’, has borrowed extensively from other languages such as Latin, Greek and French. The borrowing includes both scholarly and everyday words and many text types and subject domains are characterised by their extensive use of such words. The study reported here is a survey of Latinate, Greek and French words (henceforth referred to as borrowed words) and their use in contemporary British English. The survey has the objective to chart the distribution of borrowed words across a set of different text types and subject domains. We report their frequencies of use and present a quantitative description of their distribution in different text categories (such as writing vs. speech and academic prose vs. nonacademic prose) and different domains (such as medicine and social sciences). The survey is significant in that it makes use of a large corpus of contemporary British English totalling 100 million words. To our knowledge, no similar survey has ever been performed on such a scale. It is also significant in that it measures the use of borrowed words both across different text types as an important stylistic feature and across a set of different domains as a subject-specific differentia. As our results show, such a study not only lends itself to our understanding of the impact of borrowed words on contemporary English lexicon but will also contribute to text typology in general and automatic text classification and genre detection in particular.