In a summary of the main early sociolinguistic studies carried out in Australia (McGregor, 1979; Horvath, 1985; Guy et al., 1986; Guy and Vonwiller, 1989), McGregor (2005: 34) comments that the linguistic situation at the time of those studies provided a typical starting point for a change in progress based on social distributions. In particular, uptalk was at that time a recent innovation and ‘a feature of the speech of young adolescent females … generally associated with the low prestige, broad variety of Australian English’. Similarly commenting on the spread of uptalk, Bradford (1996: 23) writes about how uptalk was ‘initially a peer group activity, creating a speech community’ among young women, before it began ‘permeating the speech of men and older members of society only after becoming well established in a community’. There is in fact a broad consensus that uptalk is primarily and initially associated with younger and female speakers. In addition, variation in uptalk use has been linked with sexuality/sexual orientation, ethnicity, socioeconomic grouping and a range of other speaker-related factors, as well as with discourse-related factors such as text type. These factors have not been studied equally for all varieties of English, so the following sections will draw from sources from a number of regions. Until further research has been carried out, it will remain unclear whether each of these factors affects each variety in the same way.

Speaker sex

The most commonly made claim concerning the distribution of uptalk is that it is a feature of female speech. This claim needs to be considered however in the context of more general intonational differences between the sexes.

Differences between women and men in their patterns of intonation have been widely studied (for reviews see, among others, McConnell-Ginet, 1978; Henton, 1989, 1995; Daly and Warren, 2001). For physiological reasons involving the relative size of the larynx, females have higher and wider pitch ranges than men, though the two ranges overlap, covering a range of approximately 50–250Hz for men and 120–480Hz for women (Laver, 1994: 451).