On the omission of articles and copulae in German newspaper headlines

This paper argues based on a corpus-linguistic study that both omitted articles and copulae in German headlines are to be treated as null elements NA and NC. Both items need to be licensed by a specific (parsing) strategy known as discourse orientation (Huang, 1984), which is also applicable in the special register of headlines. It is shown that distinguishing between discourse and sentence orientation and correlating these two strategies with λ-binding and existential quantification, respectively, naturally accounts for an asymmetry in article omission observed in Stowell (1991).