Counting and measuring – a reply to Liebesman
Résumé
In a recent paper, Liebesman (2014) claims that we do not “count by identity”. Counting some objects by identity is giving the cardinality of the plurality that has these objects as members. Under this conception, some objects are two just in case these objects comprise one object distinct from another and no other object. Liebesman presents two arguments for his claim: the bagel argument and the liter argument.
In my reply, I focus on the bagel argument, which is the most important. Liebesman first purports to show that in (1) Two and a half bagels are on the table, we do not count by identity. In particular he rejects possibilities like the following: the expression two and a half bagels would really be two bagels and a half bagel. He then argues that in (2) Exactly two bagels are on the table, we do not count by identity either. Taking a uniform semantics
to be preferable, he concludes that we never count by identity.
As I explain in this paper, I agree that in some cases, we do not count directly and use some kind of measurement instead. But contra Liebesman, I argue that this is the exception rather than the rule. In normal cases, we just count (“by identity”). Moreover, I provide an explicit semantics of the relevant sentences. Finally, I argue that even when we measure, this
involves genuine counting.
Origine : Fichiers produits par l'(les) auteur(s)