Stylistic and Contextual Effects in Irony Processing

Stylistic and Contextual Effects in Irony Processing Akira Utsumi (utsumi@se.uec.ac.jp) Department of Systems Engineering, The University of Electro-Communications 1-5-1 Chofugaoka, Chofushi, Tokyo 182-8585, Japan Abstract Irony is perceived through a complex interaction between an utterance and its context and serves many social functions such as to be sarcastic and to be humorous. The purpose of this pa- per is to explore what role linguistic style and contextual in- formation play in the recognition of irony (i.e., assessing the degree of irony) and in the appreciation of ironic functions (i.e., assessing the degree of sarcasm and humor). Two exper- iments demonstrated that the degree of irony and sarcasm was affected primarily by linguistic style (i.e., sentence type and politeness), while the degree of humor was affected by both linguistic style and contextual information (i.e., context neg- ativity and ordinariness of negative situation). These results are almost consistent with the predictions by the implicit dis- play theory, a cognitive theory of verbal irony. Discussion of the findings also suggests that the implicit display theory can account for an indirect effect of context on the degree of irony. Introduction Irony is an interesting pragmatic phenomenon whose process- ing involves complex interaction between linguistic style and contextual information. There are also good reasons for prob- ing the mechanism of irony processing in cognitive science. First, irony offers an effective way of accomplishing various communication goals for maintaining and modifying social and interpersonal relationships that are difficult to do literally. Second, irony processing requires higher-order mindreading ability (Happ´e, 1993), which has been argued to play an im- portant role in the interpretation of ordinary utterances (Wil- son and Sperber, 2004). Third, as Gibbs (1994) argues, an ironic way of talking about experiences reflects our figurative foundation for everyday thought. Recently, many studies have paid much attention to irony processing (e.g., Gibbs, 1994; Sperber and Wilson, 1995; At- tardo, 2000; Colston, 2002; Giora, 2003). However, most of these studies focus only on the difference of processing be- tween ironic utterances and literal ones, in spite of the fact that irony is communicated by various kinds of expression (Kumon-Nakamura, Glucksberg, and Brown, 1995; Utsumi, 2000). For example, to your partner who stepped on your feet many times during a dance, you can say ironically in various ways: not only an opposition statement like “You’re really a good dancer”, but also a true assertion “I love good dancers”, a rhetorical question “Could you step on your own two feet?”, a circumlocutory utterance “I guess you have a broken leg”, and so on. The purpose of this study is to empirically ex- amine how irony processing differs among different kinds of ironic utterances and what role style and context play in caus- ing such differences. The issue of controversy in irony research is according to what features of irony people distinguish irony from non- irony. Beyond the fallacious view that irony is a meaning opposition or a mere violation, a number of studies have pro- posed a variety of views of irony: Irony is an echoic interpre- tation of an attributed thought (Sperber and Wilson, 1995), joint pretense (Clark, 1996), relevant inappropriateness (At- tardo, 2000), or indirect negation (Giora, 2003). However, these theories suffer from the same problem that they have attempted to provide necessary and/or sufficient properties for distinguishing irony from nonirony; there appear to be no such properties shared by all ironic utterances. To over- come this difficulty, I have proposed a more comprehensive view of irony, implicit display theory of verbal irony (Utsumi, 2000). The implicit display theory takes a comparative view that irony is a prototype-based category, which is the idea un- derlying cognitive linguistic research. Another point in which the implicit display theory radically differs from the previous views is that it claims a differential role of style and context, whereas the previous theories do not address such a difference or they confuse the different roles. According to the implicit display theory, style of an ironic expression is used to assess to what degree a specific ironic utterance is similar to the pro- totype of irony, while context motivates the addressee to in- terpret an expression ironically. The study I present in this paper empirically examined to what degree people perceive an utterance as ironic depending on style of the utterance and its context, and tested whether the claims of the implicit dis- play theory can explain the observed result. Another heated topic in irony research is the social func- tion of irony, which provides a plausible answer to why peo- ple use irony. The functions are divided into negative ones such as to be sarcastic and to criticize, and positive ones such as to be humorous. Previous studies (e.g., Dews and Win- ner, 1995; Colston, 2002) have compared the degrees of neg- ative effect between ironic utterances and literal equivalent utterances. However, these studies have not addressed how various kinds of ironic utterances differ in negative and pos- itive functions. My study thus examined both negative and positive effects of various ironic utterances by asking people to rate the degree of sarcasm and humor, and tested whether the obtained finding can be explained by the implicit display theory. Implicit Display Theory The main claim of the implicit display theory is threefold (Utsumi, 2000). First, irony presupposes ironic environment,