July 25, 2017
Arguments against the Russellian theory of definite descriptions based on cases that involve failures of uniqueness are a recurrent theme in the relevant literature. In this paper, I discuss a number of such arguments, from Strawson (1950), Ramachandran (1993) and Szabo (2005). I argue that the Russellian has resources to account for these data by deploying a variety of mechanisms of quantifier domain restrictions. Finally, I present a case that is more problematic for the Russellian. While the previous cases all involve referential uses of descriptions (or some variations of such uses), the most effective objection to the uniqueness condition draws on genuine attributive uses. The research that led to this paper has been supported partially by the Spanish Ministry of Economy and Competitiveness, EXCELENCIA programme, project no. FFI2016-80636-P. I would also like to thank Dan Zeman and Adrian Briciu for comments on a previous version of the paper.