An empirical investigation of definiteness

The debate between those who believe that uniqueness is the defining property of definite noun phrases (Russell, 1905; Loebner, 1987; Kadmon, 1987; Abbott, 1999; Roberts, 2001) and those who instead argue that the determining factor is familiarity (Christophersen, 1939; Heim, 1982; Prince, 1981, 1992) is still ongoing (see also (Birner and Ward, 1994; Abbott, 2000)). We report the results of a study in which this question was investigated by means of corpus analysis. Perhaps the most novel feature of this work is that for our study we used a large number of annotators together with statistical tests of reliability to ensure that our results could be replicated, whereas most previous studies were carried out by a single person on the basis of her own intuitions.