tion, especially a successful one.
We can conclude, then, that the observed convergence of the female and male registers in the mixed-gender discussions was in all likelihood not an artifact of our experimental design. In fact, the pattern of men being more competitive than women in the same-gender situation and no gender difference in the mixed-gender situation corresponds to results reported by Aries ( 1976), Carli ( 1989), and others. Some of those studies (e.g., Bilous & Krauss, 1988, and Carli, 1989) also agree with the present study in observing that it is mainly the women who show a reduction in gender-specific interactional behavior when talking in a mixed-gender group (note, however, that Bilous and Krauss found more interruptions in all-female than in all-male discussions). The convergence pattern also parallels the findings of Ervin-Tripp and Lampert ( 1992) on the use of humor by women and by men. They found both women and men to shift their strategies towards a more equal balance in mixed-gender interactions.
There is an intriguing difference, however, between our results and those from other studies: The interactional style in the all-female group does not fully reflect what is assumed to be the typical female register, that is, an involved, collaborative style with frequent simultaneous talk and short turns ( Edelsky, 1981; Coates, 1989; Tannen, 1990). The women did use fewer interruptions in the all-female discussions, but they also produced less (instead of more) noninterruptive simultaneous talk and relatively long turns (for a similarly mixed pattern in Dutch women's talk see De Boer, 1987). We can only speculate about the causes of this discrepancy. First of all, neither the formal nor the informal discussions were truly informal with substantial FLOOR 2 interaction, whereas female register in the studies cited has been described for casual conversations or for casual episodes in task-oriented interactions. Our experience with Dutch and (WestCoast) American culture suggests that humor and small talk are more strictly separated from "shop talk" for the Dutch than they are for Americans (compare Ervin-Tripp & Lampert's observation that Americans are sometimes reproached for laughing too much and addressing serious problems with humor). The absence of an involved, collaborative female register in our data thus might be specific to the task-oriented talk we have been looking at, which in Dutch admits some, but not all of Edelsky's FLOOR 2 characteristics and thus only some of the register characteristics associated with FLOOR 2 interaction.
Aries, E. ( 1976). Interaction patterns and themes of male, female, and mixed groups. Small Group Behavior, 7, 7-18.
Aries, E. ( 1987). Gender and communication. In P. Shaver & C. Hendrick (Eds.), Sex and gender (pp. 149-176). Newbury Park, CA: Sage.
Beattie, G. W. ( 1981). Interruption in conversational interaction and its relation to the sex and status of the interactants. Linguistics, 19, 15-35.