Self Reflections in the Mass

Savage’s ambitious book offers a seductive association between, on the one hand, the rise of a new middle class which, in the name of a technological modernity, sharply distinguished itself from the ‘gentlemanly’ culture of the Britain’s established intellectual elites and, on the other hand, the postwar flourishing of the social sciences in Britain’s universities, ousting their previous subordination to the ‘gentlemanly’ pursuit of the arts and humanities. While neither phenomenon can be reduced to, or entirely explained by, the other, Savage argues that there are causative links running in both directions. Sociologists, Savage argues, should pay attention to the ‘footprint’ of their own activity which was itself implicated in the reconstruction of popular identities.

This brief summary does not do justice to the range of Savage’s argument or to the flood of ideas which spill, sometimes confusingly, over the main lines of the exposition. The four reviewers in Sociology’s symposium on the book found different things to praise and to criticize but all were agreed that it was an exceptionally difficult text to summarize. As Sheila Rowbotham put it, ‘there is a compressed density in the writing which led me to conclude that there are several books rather than one squashed between the covers’.¹ In what follows I will discuss only one of these books – not the history of sociological method, but the argument that between the 1930s and the 1990s:

powerful new kinds of middle-class identities were generated, ones which broke from old notions of ‘status’ and ‘gentility’ and which emphasised instead the technocratic and scientific capacities of the middle class, and hence saw them as key parts of an efficient and modernising nation.²

A good deal of the evidence on which this argument is based has been drawn from the archives of Mass Observation (MO), some of which I have worked on myself. MO, both in its early years (1937–49), and in its later reincarnation as the Mass Observation Project (MOP) since 1981,³ provides wonderfully rich material for the investigation of popular – and particularly middle-class – identities. Here I will focus on Savage’s use of this material in order to address some important issues about the value of the MO archive to historians concerned with identity and social change.

Savage uses a 1990 MOP ‘directive’ – MO’s term for the open-ended questionnaires that it sends to its panel of correspondents – on ‘social divisions’ to demonstrate ‘the ubiquity of social science narrative ... now routinely invoked in the statements of the literate middle classes’ (p. 238), sociology, he argues, having effectively displaced ‘older modes of knowing the social associated with the arts and humanities’ (p. 243). By way of illustration he offers three extended quotations in which respondents made explicit reference to sociological debates about class. These certainly provide clear evidence of sociology’s ‘footprint’ in the culture, and, although none of the respondents lay claim to technical or scientific expertise as central to their middle-classness, they do reveal a self-reflexivity which Savage sees as characteristic of the new middle-class identity. Rather than presenting themselves as ‘exemplifying typical patterns’, the mass observers resist easy classification, insist on the fluidity of their identities and offer ‘life narratives designed to exemplify the singularity of the correspondent’s life ... Class is presented as a matter of agency, rather than as something handed down. ... ’ (pp. 240, 242).

My own sample of the 1990 material (on which I am currently working) reveals a similar sense of agency, but, unfortunately for Savage’s argument about change over time, much the same is true of the responses MO elicited from its 1939 directive on class. When Savage first put forward these arguments, in an article published in 2007, based on a reading of MO material from 1948 and 1990, he was careful to acknowledge the danger of relying on two snapshots – ‘it is tempting to read the accounts as symptomatic of a broad period, rather than...