Abstract: Usually, the label “Nominalism” evokes sorne kind of ontological parsimony. Ockham’s Razor epitomizes this understanding of Nominalism: “it is pointless to do with more things what can be done with fewer”—or in a later formulation “entities should not be multiplied beyond necessity”. You can be a nominalist about all sorts of things—e.g., universals, abstract items, states of affairs, possible or impossible worlds, and so on. As far as Nominalism goes, William of Ockham was probably as hardcore as it gets, both in his ontology and in his semantics.

Ockham’s semantic account became the baseline picture shared by a group of later authors. These logicians embraced Ockham’s broad outlook and approach, without shying away from even major doctrinal disagreements. Along with Ockham’s general perspective, they also inherited Ockham’s problems.

Here, I am especially interested in the problems that 14th-century nominalists had with handling empty terms, i.e., those terms that do not have—and sometimes cannat have—a referent in the world. Such terms are problematic in medieval logic in general and for post-Ockham nominalist logicians in particular. Sorne of these logicians seemingly do away with ontological parsimony, buying into all sorts of nonexistent but imaginable referents (imaginabilia)—even those that are intrinsically impossible because they are self-contradictory. The most renowned proponent of this type of view was Marsilius of Inghen. Marsilius’ semantics of empty terms became the standard term of comparison against John Buridan’s just as famous alternative doctrine.

In this talk, I explain the problematic background of empty terms in Aristotelian logic and the additional struggles faced by nominalist semantics. Then, I compare Buridan’s and Marsilius’ alternative accounts of necessarily empty terms. Is Marsilius committed to a whole lot of nonexistent entities? Does he do so beyond necessity, pointlessly doing with more things what could be done with fewer? What would this commitrnent amount to? How can it be compatible with being a nominalist?