Argumentative euphemisms, political correctness and relevance

The account presented in the thesis combines insights from relevance-theoretic (Sperber and Wilson 1995) and neo-Gricean (Levinson 2000) pragmatics in arguing that a specific euphemistic effect is derived whenever it is mutually manifest to participants of a communicative exchange that a speaker is trying to be indirect by avoiding some dispreferred saliently unexpressed alternative lexical unit(s). This effect is derived when the indirectness is not conventionally associated with the particular linguistic form-trigger relative to some context of use and, therefore, stands out as marked in discourse. The central theoretical claim of the thesis is that the cognitive processing of utterances containing novel euphemistic/politically correct locutions involves meta-representations of saliently unexpressed dispreferred alternatives, as part of relevance-driven recognition of speaker intentions. It is argued that hearers are “invited” to infer the salient dispreferred alternatives in the process of deriving explicatures of utterances containing lexical units triggering euphemistic/politically correct interpretations. In the course of time, such invited inferences can lead to semantic change by becoming routinized relative to some context of use and reanalyzed as the defeasible default meanings of these locutions, presumed in the absence of contextual assumptions to the contrary. This conventionalization process is responsible for euphemisms becoming “contaminated” with negative connotations associated with taboos, which leads to their recycling in the vernacular or ‘euphemism treadmill’. It also explains why political correctness is effective only when it is novel and still capable of bringing people’s unconscious biases to consciousness. The biases are, arguably, brought to consciousness by metarepresenting the salient dispreferred alternatives as part of comprehension of utterances containing PC locutions perceived to be marked in the given context. It is suggested that the likelihood of the euphemism treadmill taking place is increased in cases of narrowing the lexicalized meaning of a concept to its taboo meaning, while it is less likely to happen in cases of conceptual broadening.

Thèse de doctorat : Université de Neuchâtel, 2014