Resonating with contextually inappropriate interpretations in production: The case of irony

Abstract

According to the graded salience hypothesis, salient meanings and salience-based interpretations are not only involved in language comprehension but also in language production (Giora 2003, 2011a; Giora and Gur 2003). This should be true of irony production as well. If, as predicted by the graded salience hypothesis, the ironist herself indeed activates utterance interpretations on account of their salience-based accessibility rather than solely on account of their contextual fit, this might be reflected in the ironies' environment. Given the crucial role of the salience-based interpretation of “what is said” in deriving and supporting the ironic interpretation, this interpretation should not be suppressed (Giora 1995). Such a view of irony production predicts that its environment will demonstrate dialogic resonance (à la Du Bois, this volume) with ironies' salience-based, but incompatible interpretations. To test this prediction, we studied a written Hebrew corpus including over 1600 ironies. Our findings show that 46% of the ironies, 10% of which are extended ironies, are addressed via reference to their salience-based contextually incompatible interpretations; resonance with the context-based , ironic interpretation occurred in only 8% of the cases; the environment of the rest either did not resonate with any of their interpretations (43%), or resonated with both their compatible and incompatible interpretations (3%). These results support the view that, like comprehenders (Giora et al. 2007), irony producers too activate and retain salience-based albeit inappropriate interpretations.