Being a monster Brian Rabern September 1, 2014 If I were to say to you, “I love you”, it would be true just in case the author loves the reader. But imagine we devised a clever way of speaking such that when we follow up a statement with the expression “swap!”, the contextual interpretation of speaker and hearer are switched. Under this supposition, if I were to say to you “I love you, swap!”, it would be true just in case the reader loves the author, i.e. just in case you love me. If natural language had means of expression such as “swap!”, it would be populated by linguistic devices that David Kaplan thought to be gross exceptions to the semantic norms. Kaplan (1989) bracketed off such devices as semantic freaks or semantic monsters. In essence, a semantic monster is an expression which when affixed to a sentence requires that the sentence be re-interpreted as if it were uttered in a different context. The topic of monsters has reared its head in a wide variety of philosophical and linguistic debates. The issue comes up the most with respect to the semantics of discourse about thought and talk, but the issue is at the very foundations of semantic theory, and has been implicit in the early discussions of semantic metatheory (e.g. Evans 1985: 375) . The thesis that natural language does not permit monsters has faced— in the intervening years—various empirical challenges (e.g., Schlenker 2003, Anand and Nevins 2004, Rabern 2012, Santorio 2012, among many others). But the issue remains controversial. In a recent article Predelli (2014: 389) claims that “this controversy has proceeded in apparent ignorance of the fact that Kaplan fails to provide a univocal definition of a monster and vacillates between at least three non-equivalent alternatives”. The aim of this note is show that there is a univocal notion of a Kaplanian monster. After rehearsing Kaplan’s framework, I will draw out the essence of a monster and lay down an explicit definition. Predelli’s first two notions are captured by this definition, while his third notion escapes it. This is as it should be, since the third notion of a “meaning-shifting” operator is not strictly speaking a monster in Kaplan’s sense of an “operation on character”. Recall the distinction between two kinds of meaning that Kaplan proposes, the character and the content of an expression. These two aspects of meaning play very different roles in Kaplan’s semantic theory: the content is the information asserted by means of a particular utterance, whereas, the character of an expres- sion encodes what any utterance of the expression would have as content. Characters . : Contexts → (Circumstances → Extensions) Contents . c : Circumstances → Extensions The general picture is this (see figure 1): the domain of the character function is a set C . Each c ∈ C is a tuple of content-generating parameters—these tuples are called “contexts of utterance”. Character func- tions map contexts of utterance to contents. The content of an expression is itself a function from a set I to extensions. Each i ∈ I is also a tuple of parameters, often assumed to be possible worlds (or worlds paired with times, locations, etc.)—these are called “circumstances of evaluation”. 1