Sociologists of education have not been slow to pursue these paths, but as much for methodological as theoretical reasons. A few historically informed studies of the teaching profession exist and there have been important demographic studies, but empirical fieldwork has concentrated heavily on the early professional socialisation of teachers. This theme is undoubtedly of great significance in its own right for, as Waller noted, 'Observation of personaUty changes in young persons just beginning to teach will help us to understand the process of occupational moulding.' But the virtual restriction of research to the study of the student, and to a lesser extent the probationer, teacher reflects, I suspect, the ready availability of 'beginning teachers' to the researcher, for whom during training they are effectively captive subjects, and the desire of the researcher, who is usually a teacher in a college or university department of education, for knowledge about

his own effectiveness as a teacher trainer. It is this ease of access which in part led researchers working in the subcultural and ethnographic traditions of symbolic interactionism to pay more attention to pupil, rather than teacher, cultures.