The main objective of this paper has been twofold. First, I critically examined two standard views of irony proposed in traditional studies of irony: one in which irony conveys the contradictory of what is literally said, and the other in which irony is seen as a figure of speech where the intended meaning is the opposite of what is literally said. Secondly, I tried to see how intrinsic limitations in these views can be overcome within a new framework of irony; their intrinsic problems were critically examined in comparison with the echoic account of irony proposed by Sperber and Wilson. The echoic approach is more explanatory and predictive than traditional approaches in at least the following respects. The former approach can handle cases where ironic utterances convey the contradictory or opposite of what is literally said as well as various other cases of ironic utterances (ironical understatement, ironical quotation, ironical interjection, etc.). The former approach assumes that ironic instances are basically kinds of echoic mentions, which makes it possible to predict the existence of various ironic utterances of different degrees and types. These points should show that the echoic approach is basically the right direction for a natural and comprehensive account of irony. (Masa-Aki Yamanashi)