The textual priming of hard news stories

Recent corpus-linguistic work has shown that our lexical choices are governed by collocational, grammatical and textual considerations. Certain words can be shown to appear in (or avoid) paragraph- initial and text-initial position in specific genres far more often than can be explained by reference to random distribution. Although this phenomenon has been demonstrated for a number of words, there has been no large-scale study in one well-defined text type. In this project we explore in detail whether the words that journalists choose in the construction of their texts manifest the same kind of built-in bias for certain textual functions and positions