Flouting Grice's maxims: A study of gender-differentiated speech

This dissertation investigates the relationship between gender of the speaker and indirectness. There is a popular belief that women's speech is more indirect than men's, and it has been argued that women's speech is both more polite and less direct than men's speech (R. Lakoff 1975). While a number of empirical studies cast doubt on the correlation between sex of the speaker and politeness (Dubois and Crouch 1975, Brouwer, Gerritsen, and DeHaan 1979), and Brouwer 1982), as well as that of indirectness and politeness (Davison 1975, Blum-Kulka 1987), few studies question the premise that women are more indirect (cf. Gleason and Greif 1983).; In his investigation of indirectness, Grice (1975) suggested that an utterance has two parts: what is said and what is implied. In order to explain how people are able to imply more than they actually say, he proposed a theory of conversation which consists of a cooperative principle and associated maxims. Within Grice's theory, one type of indirectness can be achieved by deliberately violating or 'flouting' a maxim, where one expects the addressee to recognize this violation and thereby to infer something other than what was literally said.; The present study focuses on the type of indirectness represented by flouting the Gricean maxims. Analysis of conversations in seven families indicates that the fathers use this type of indirectness more often than the mothers do, both when talking to children and when in adult conversation.; Finding empirical evidence that men speak more indirectly than women do in certain situations confronts a societal myth that the reverse is true. In addition, Sperber and Wilson (1986) propose a theory of indirectness which differs from Grice's in that it has a cognitive rather than a social basis. Within this theory one would be less likely to expect social variables to interact with degrees of indirectness. The results of this study thus also have broader theoretical significance, since they show that the social variable of gender is a significant factor in the use of at least one type of indirectness, providing evidence in support of a theory with a social component.