How fictive names can refer

Manuel Englert

Universität Hamburg

A common objection to some realist views of fictive objects is that they
face the Selection Problem: authors cannot fix the reference of a fictive name such as ‘Sherlock Holmes’, ‘Snowy’, or ‘Arrakis’ because they cannot uniquely identify the object to which they are referring (see Sainsbury 2010, Chapter 3; Kripke 1980, 157-158, Kaplan 1973, 505-506; and Plantinga 1974, 154-155). I argue that the Selection Problem should be resisted. The central idea of my argument is that, even outside the context of fiction, fixing reference to an object does not always require singling out exactly one object. This point is illustrated by familiar examples of names that refer to vague objects, such as mountains, names introduced by arbitrary stipulations such as “Let Giulia be an arbitrary Italian”, and names for indiscernible objects. There are good reasons to hold that these names refer. I argue that the strategies developed to explain how these names can refer can also be applied to fictive names. This yields two theoretical options for how reference of fictive names works: the arbitrary reference theory (see Breckenridge and Magidor 2012) and the multiple reference theory (see Merlo (2017). The arbitrary reference theory denies that authors must know which object they are fixing reference to; instead, reference is fixed arbitrarily, with no possibility of knowing which object has been selected. The multiple reference theory rejects the assumption that a name must take exactly one object as its semantic value; instead, it allows a name to take several objects as its value. On this view, a fictive name refers multiply to several objects.

Chair: tba

Time:

Location: