Finally, then, I return to Black's primary school data, drawing here on Black (2004b). As we have seen in Chapter 2, a major finding of this study is the difference in the quality of teacher-pupil talk when different groups of children are compared: middle-class boys tend to be involved in more "productive" exchanges with the teacher in the sense of dialogic interactions in which they are positioned and position themselves as genuine interlocutors in the learning project. There is also an indication that this basic distinction is cross-cut by ethnicity as suggested by the cases of Hasan, who is Gujarati, and Janet, who is Chinese (see Black, 2002a for a fuller account). In what follows, we see that there is a further layer of explanation to add in the case of Sian, a middle-class girl who is described by Black as an anomalous case" in that, with one exception, she is involved in far more exchanges than the other girls in the class. However, a close inspection of her participation in these exchanges shows that it is qualitatively different from that of the middle-class boys, and so she is anomalous in this sense too.