The aim of this article is to illustrate how highly rudimentary computer-assisted text analysis can aid the preliminary mapping of public discourse. The case that has been chosen is the use of a particular combination of words, ‘radical’ and ‘Muslim’, in a corpus of articles published in the Swedish printed news media which have been ac-cessed through the database Mediearkivet. Texts in the corpus were accessed and an-alysed using custom-made code in the programming language Python. The article shows how even limited skills in computer-assisted methods can aid scholars in dis-covering patterns and trends in a large collection of texts which is difficult, if not im-possible, to process manually.