A source of variation: A corpus-based study of the choice between π and κ in the NT Greek Gospels

Abstract

Using a quantitative methodology based on extensively annotated corpus data from the PROIEL corpus, we examine the distribution of απο and εκ in the NT Greek Gospels. The original semantic opposition between these two prepositions in terms of an ablative-elative distinction started fading during the historical development of Greek and has been argued to be already much weaker at the time of the New Testament. To explore this we generate a semantic map without semantic pre-analysis on the basis of four parallel language samples. We then use statistical techniques to interpret this map. We ind that there is still a fairly clean separation between εκ and απο largely based on semantic role. However, απο is quite frequently used in elative contexts. A lexical analysis clarifies that the use of απο in this environment amounts to the preposition specialising with certain lexical items, some of them with variable interpretations, as seen in the case of toponyms. © 2012 BRILL.