On the types and tokens of Dutch dezelfde/hetzelfde

The main claim of this paper is that the Dutch word de-zelf-de ‘the same’ has much in common with superlatives, expressing maximal similarity or identity. The distinction between elative and maximal superlatives corresponds to a type and token interpretation of de-zelf-de. Like superlatives, -zelf-de must be bound by de or het. In attributive positions, the initial de of de-zelf-de can either be a degree element or a determiner. In postnumeral position, e.g., twee de-zelf-de auto’s ‘two identical cars’, its binder is the degree element de-. This gives rise to a type interpretation. In prenumeral position, as in de-zelf-de twee auto’s ‘the same two cars’, -zelf-de is bound by the determiner de. This gives rise to a token interpretation. When there is both a determiner de and a degree element de, as in de vrijwel de-zelf-de twee woningen ‘the almost the same two houses’, the token and type readings are combined.