The British, we are often told, are a people uniquely obsessed by 'class'. Not class as sociologists generally understand it (i.e. relationship to the means of production or structural inequalities in social capital and power), but 'class' as an all-embracing word for describing (and defining) distinctions based on perceived social differences. It is 'class' in this loose, vernacular sense that has been most prominent, and most influential, in modern British history -a point fully recognized in these three new books on the British sense of class. Indeed, it is a central argument of David Cannadine's stimulating and selfconsciously revisionist book, Class in Britain, that historians must rehabilitate 'class' as a legitimate subject of historical enquiry, but that they must do so without 'rehabilitating old-style class analysis'. 'Class', he insists, 'is one of the most important aspects of modern British history', despite recent tendencies to play down its significance (16-17). Quite so, but the question, of course, is whether this bold attempt at 'defining the subject afresh and envisioning it anew' can be judged a success. In large measure it can. According to Cannadine we must accept that 'class is best understood as being what culture does to inequality and social structure' (188). Cannadine is thus arguing for the inherently constructed nature of class and class identities (so that the distinction between 'class' and 'socio-economic position' would be similar to the distinction many historians now recognize between 'gender' and 'sex'). Not surprisingly, therefore, this is a book concerned primarily with the 'languages of class', or rather with representations of the social order -for Cannadine wishes to uphold a distinction between vocabularies of 'class' and 'models of society' (166). The former, he suggests, have done surprisingly little to shape social identities, because, unlike models of society, they cannot