Book Review: Identities and Social Change since 1940: The Politics of Method

The argument unravelled

If there is any danger of self-satisfaction or a run on self-congratulation amongst contemporary sociologists, then Mike Savage's message should perhaps create some unease. His book is complex, involving a number of interweaving themes not easy to summarize briefly and the densely-written Introduction may deter some readers. He begins by suggesting that people's identities, once embedded in specific landscapes, have changed as new identities were created, based on the emerging abstract language of social science. He outlines his notion of ‘elective belonging’, which he has already explored in the various publications carried out with colleagues in the Greater Manchester area.

Coupled with this shift in the construction of social identities, Savage identifies the emergence in the post-war period of a new category of technical expert, based on meritocratic skills: the analysis of this development provides him with a historical grounding for an analysis of a different kind of post-war social science. The practitioners of this new mode have not been studied systematically and since, according to Savage, I am included in such a putative category (being in the 1960s a participant and observer as both researcher and teacher and now, as a result of recent interviews, a respondent), I declare more than a passing reviewer's interest. But more of that later.

Luckily, it seems, I just missed what Savage describes as the end of ‘gentlemanly social sciences’, which he suggests continued well into the 1950s. Certainly, the final battles between the ‘gentlemen’ and the technical experts were particularly bitter and damaging: Halsey has described the ‘uncharitable and rancorous atmosphere’ at the LSE in the 1960s (Halsey, 2004: 78). From what I heard from my supervisor John Westergaard in 1961 and 1962 of the vicious infighting between David Glass and Donald MacRae (sic Not MacCrae as misspelt by Savage, 110), most other Departmental squabbles seem like vicarage tea parties.

This new specialist sociology Savage recognises as ‘a social movement, in which social science became vested with unprecedented hopes and aspirations’ (20). Crucially these ‘Pioneers’, as they are called in Paul Thompson's ESRC, Data Archive Project, were not ‘innocent observers of change, but imported it as an essential part of their self-identity’ (ibid). This, then, was a project, different from what had come before and, one might suggest, different from the liquid sociology of today.

Once this new breed of social scientists has been delineated, Savage moves on to consider what might be their reformulated, scientific object. The answer, he suggests, was to define an ‘ordinary, average, national society’ (ibid). He makes much of the way a new tradition of community studies, pioneered by Welsh trained geographers and anthropologists, ‘questioned the assumed normality of English Life’ (ibid). Evidently this is a somewhat quirky and unconventional stance to which I return below.

Much of the rest of the book is about the politics of method, focussing on the interview and social survey research. Some of this material has already been published and has been the focus of critical comments (Crompton, 2008). Finally, Savage suggests that the new social science infrastructure was used to mine down into everyday life to show ‘how people's ordinary situations shaped their lives, actions, thoughts and endeavours. This was an intimate, critical and usually compassionate sociology’ (237).

The key contention, therefore, is that something distinctive and special developed in the 1950s and 1960s providing a sociological project that helped to make what Savage describes as ‘the golden age of the academic social sciences’ (249). He now believes that the social sciences have lost this project and he is, at best, ambivalent about whether it can be recovered.

This, then, is the bare outline of what I take Savage to be seeking to demonstrate. I now turn to consider whether the evidence and the sources he adduces to support his arguments are strong enough to convince us that his thesis is built on sound foundations. There are certainly some grounds for uncertainty.

The evidence for the argument assessed

Savage begins the search for empirical evidence by drawing on his own previously published research. His suggestion that ‘We can discriminate, in the broadest terms, between those who choose and are vested in place, and those who are thrown into, and dwell in it ‘(33) is not, of course particularly novel: Savage sees this as a main theme of post-war social science (37). He refers to Jackson's study of Huddersfield that reported on the ‘dwelling of those “born and bred”, where family affiliations define one's place’ (36), unlike the majority of those interviewed in Luton for the Affluent Worker Study, who were mostly immigrants from elsewhere in the UK.

This concern with ‘elective belonging’ and ‘the enchanting of the landscape’ in the 1960s was not perceived then to have its present putative significance in a wider cultural change. Now Savage claims that:

These new conceptions of the landscape, he argues, were generated by the practices of the social sciences: their methods and expertise evoked new social divisions based on ‘choice’ (the middle class) and ‘rootedness in the landscape’ (the working class). When I described this process in 1965 I did not see the full significance of this in the way Savage does now (Pahl, 1965).

It seems that researchers in the 1960s were unwittingly complicit agents ‘with the very business of change and identity’ (47). One factor that then greatly influenced my personal perceptions of change, came from working in Hertfordshire, where there were four designated New Towns rapidly filling up with working-class immigrants from London. Savage does not consider the implications of this and similar substantial working class re-settlement for working class identity.

Turning to a discussion of the British Intellectual and Highbrow culture, Savage now clearly distances himself from Bourdieu's Gallic approach to intellectuals, based primarily on the leisured aesthetics of the humanities. In Britain, by contrast, he sees ‘technical identities’ challenging and ultimately breaking from a previous fusion of modernism and gentility of which the Bloomsbury group is taken as a prime example. But, perhaps surprisingly, Savage's delineation and analysis of this new intellectual group, which many might associate with the WEA, and the intellectual infrastructure of The Left Book Club and the early Pelican paperbacks, is limited to the social class and activities of those who volunteered to work for Mass Observation.

This may seem a rather bizarre choice: the assertion that the Mass Observers ‘were identifying themselves as a kind of technical, “social scientific” intellectual’ (60) is odd. Other readings of the personal diaries of these Observers might suggest that whatever their self-rated class, many of their accounts were self-indulgent and self-pitying (Garfield, 2004). Descriptions of what they were preparing for their suppers or how they made a quick profit in the antique trade don't exactly support the notion that this is ‘the early crystallization of a scientific and technical aesthetic’ (79). Of course it depends on what quotations one picks, but Jonathan Rose's account of The Intellectual Life of the British Working Classes, with its emphasis on the WEA or such institutions as the Welsh Miners' Libraries, provides a convincing and grounded thesis with considerable empirical depth that at least deserves to be more seriously considered (Rose, 2001). The emerging technocratic elite had deep roots and did not spring up out of nowhere between 1940 and 1960.

I find it particularly puzzling that Savage should rather cavalierly assert that the post-war intellectual current was not the simple product of social changes associated with ‘the expansion of new industries, bureaucratic structures or the welfare state, or as the product of new kinds of political or cultural regime,’ but prefers to emphasize how it ‘emerged through dispute with older cultural formations’ (91). This is a bold statement that should, perhaps, be more subtly nuanced.

Much of the rest of the book is concerned with new kinds of research methods ‘by which social scientists claimed access to, and jurisdiction over, the social’ (94). Information is abstracted from individuals and then defined in terms of social relationships – hence the idea of jurisdiction.

There can be little dispute that what passed for sociology in the inter-war period in Britain was very naive and parochial. If one compares, say, the ‘community study’ of Hertford published in the Sociological Review in 1937 by a local solicitor (Roper-Power, 1937) with the work of the Chicago School in the 1920s (Carey, 1975) or Middletown (Lynd, 1929), the contrast could hardly be greater. The old gentlemanly approach to sociology, mostly based on an evolutionary model of society, was epitomised, perhaps, by L T Hobhouse at the LSE. Collini's description of the origin of this social type is worth recording:

The moment of sociology – 1962?

It should be remembered that the BSA, founded in 1950, ‘took its cue very much from the moralizing gentry project’ (110) and papers published in the British Journal of Sociology in 1950s were eclectic and, if anything, hostile to the work of empirical sociologists. Thus we come to 1962 and what Savage celebrates as ‘The Moment of Sociology’.

The year 1962 is chosen partly because that was when the weekly journal New Society was launched. I think it is right to recognise its importance. ¹ The other main theme was the importance of the new ‘plate glass’ universities, founded in the post Robbins Report era of the early 1960s, which did much to displace the gentlemanly model. Savage takes issue with Halsey's diffusionist model of the development of post-war British Sociology from the LSE, seeing it as ‘seriously flawed’ (125). From this pivotal period of the early 1960s emerged the formation of ‘new discourses of the social as an imminent site of change (133). In the second half of the book, Savage unpacks the social science apparatus, before concluding with a discursive essay on the politics of method. In some ways this is potentially the more innovative part of the book but it suffers from weaknesses in interpretation and some misunderstanding of the nature of his sources.

It may seem eccentric for Savage to begin his analysis of the social science apparatus in the post-war period by focussing on community. However, it suits his purpose, in that what he claims was a failure to produce a ‘scientific community study’ (138) (whatever that might mean) helped to turn attention away from the landscaped vision of social analysis, which Savage himself later adopted.

Arguably, the community study tradition came more from geography and anthropology than sociology and from Wales rather than England. Professor Daryll Forde of the University College of Wales at Aberystwyth encouraged Alwyn Rees to begin a study of a parish in Montgomeryshire in 1938 and work continued for the next eight years. Professor E. G. Bowen and Professor H. J. Fleure also encouraged Rees, who lived in the parish for periods between 1944 and 1946. The book, Life in a Welsh Countryside was published in 1950. It was out of this anthropogeography at Aberystwyth that led Bill Williams to his study of Gosforth in Cumberland six years later. Savage does not consider these geographical origins, although those, such as myself, reading geography at Cambridge in the 1950s, were also lectured on Forde and Fleure and were deeply immersed in the local holistic ‘pays’ studies of the French human geography school.

Savage's emphasis on Williams leads to a neglect of alternatives. Many would see Littlejohn's community study of Eskdalemuir, carried out between 1949 and 1951 and focussed firmly on issues of social class, as a more appropriate example. Littlejohn's base was the Department of Sociology and Anthropology at Edinburgh but he obtained valuable advice from Dr W. Watson at Manchester (Littlejohn, 1963). Many would see this as a truly pioneering piece of work but Savage ignores it, perhaps because of the long delay in its publication.

Be all that as it may, it does seem that Manchester was taking over from Aberystwyth: Ronnie Frankenberg has consistently proclaimed his debt to Professor Max Gluckman for his support and guidance in doing his study in Wales. I find it puzzling that Savage should focus on the sociology of the Welsh Marches in the discussion of these issues, as if the place itself was significant. He might just as well have taken the Scottish Borders, referring to Gosforth in Cumberland and Eskdalemuir in The Cheviots. To my mind, a much more plausible thesis would be to contrast the traditional anthropogeography of Fleure and Forde at Aberystwyth with the social anthropology, developed out of the Rhodes-Livingstone Institute and the African tradition by Gluckman at Manchester. Focussing on locale rather than on intellectual tradition does not appear to be fully justified.

It is when Savage turns to Margaret Stacey's study of Banbury that his argument becomes more convincing. She came from the unpromising background of the moralistic evolutionary sociology of the LSE and an emphasis on social problems. From 1949–1951, the same years that Littlejohn was working in the Scottish Borders, Stacey was doing her fieldwork with the help of a WEA class. It took a further nine years for the book to be published. I remember The Times newspaper addressed a first leader to a discussion of its implications in 1960, which helped to inspire my own research.

Savage reports on other studies but does not refer to Klein's Samples from English Cultures (1965), which might have served him as a useful benchmark for earlier ‘traditional’ studies of working class life. His main thesis appears to be that locality studies failed to illustrate effectively the nature of change by making comparisons over time and in this, I believe, he was over-influenced by the empty and unsatisfactory nature of the Banbury re-study. He asserts that ‘sociologists increasingly wanted to lever change out of any fixed location’ (163). That may have been true for some but the commitment to locale did not get universally forgotten. At the end of the 1970s it re-emerged with studies of de-industrialisation in Sheffield, South Wales, the Isle of Sheppey and elsewhere. Surely it wasn't the end of community; rather it was putting locale into parking orbit as the battalions of sample-surveyors marched on beneath.

Sample surveys: For better or worse?

The research practices deployed by sociologists operating on a different scientific object (typically ‘Head of Households’) helped to produce a new kind of social science knowledge and, by extension, personal identity. By abstracting individuals from their households ‘key informants’ or ‘respondents’ could not be perceived and understood in the round. The modern and rational study of individuals took them away from the complexities of family and locality. The evidence Savage adduces in support of this is not completely secure.

Much of the section on interviewing (178–185) is based on material Savage has consulted in the ESRC Data Archive at the University of Essex, from which he has quoted extensively. In the case of my own archived material, Savage has got himself into a muddle. He recognised (5) that some of this was teaching material but later in the book (183) these teaching relics are misattributed as provocative and non-neutral interviews carried out by me. This is a mistake. The problem is that in addition to being a full-time graduate student at the LSE in the early 1960s, I was also a full time extra-mural tutor for the University of Cambridge. Part of my responsibilities was to ensure that students did some kind of written work. However, since most of my students had no experience or training in writing essays, I had to devise alternative strategies. One such was to give them some broad questions and get them to interview other local people or each other on those topics that interested them (See Pahl, 1970: 70–8). Inadvertently a box of this old teaching material was deposited in the Archive with my doctoral research files. How many readers of this Review would like their students' coursework later attributed to them as author? A further confusion is that Savage quotes from one of the interviews I provided for the Pioneers of Social Research Project along the lines of a life story. Unfortunately he quotes from this without making clear (as the transcript certainly does) that I was talking about the problems of getting women, with very little formal education, to engage in discussion in a WEA class. I was urged by my more experienced colleagues in the WEA to adopt a fairly robust Socratic style of teaching. This is what I described. To refer to my teaching method as reflecting a ‘masculine’ style of interviewing, as he does, is misleading and is not justifiable. It was very difficult in those days to persuade my more patriarchal colleagues that a daytime class for women was a serious educational innovation, so I was very anxious that it should succeed. My early struggles as a teacher do not seem to me to bear very directly on what Savage is attempting to demonstrate.

More seriously, perhaps, on pages 37–39 Savage discusses the interviews on the managerial spiralists (a term coined by Watson at Manchester in 1960 and not by me in 1965 as Savage generously thinks (Watson, 1960)) that formed part of the data for Managers and their Wives, (surprisingly omitted in his Bibliography) (Pahl and Pahl, 1971). I did not do these interviews. I was fortunate in being able to persuade a skilled and well-trained social anthropologist – Marie Corbin – who had just returned from fieldwork in the South of Spain – to do the extensive in-depth interviews. Marie Corbin's interviews are exemplary and her published account of the nuances of working closer to her own class culture is, in my view, a classic contribution to qualitative sociology (Corbin in Pahl and Pahl, 1971: 286–306). Whilst it may be flattering to have these marvellously subtle interviews of the middle class attributed to me, it is unacceptable to deny Dr Marie Corbin the credit which she richly deserves. Furthermore, it adds insult to injury to imply that she was in some ways implicated in Savage's thesis of Masculinising the Interview Process in the 1960s.

Another more minor criticism relates to Savage's interpretation of the comments appended to formal interview schedules of the period. (See for example, 1, 3–5, 97–98). As is now well understood, researchers prepared pre-coded interview schedules which a number of other different interviewers could ‘administer’. The principal investigators generally had other duties and responsibilities that prevented them from engaging in extensive field work themselves. However, to get some flavour of what the context was like, there was generally a space at the end of the schedule for the interviewer to add some notes in the car or the pub afterwards. Interviewing can be lonely and demanding work and one way to keep up morale is for interviewers to meet up for a chat in the pub afterwards. It may be an opportunity to show off one's Footlights' wit or to create some work satisfaction and solidarity with colleagues. To my mind these comments should not be read as serious relics of the emerging professional sociology. I suspect that from Charles Booth and Beatrice Webb to Pierre Bourdieu and Mike Savage, interviewers have gossiped informally about their respondents when off duty. I also knew anecdotally that the principal investigators at Cambridge were more concerned to get coded data on their new computing kit than to read what Michael Rose and others had written at the bottom of their interview schedules. Relics that tell us about the mood of interviewers late at night have been given a significance by Savage they do not merit. Qualitative comments might help those doing the quantitative analysis to get a ‘feel’ for their Luton respondents but they might equally well be simply ignored.

I recognise that these preceding remarks may appear negative but they do not necessarily undermine Savage's arguments. Certainly what appears to be hasty writing does reduce the force of his main theme, and this is unfortunate; what he wants to argue is important and still worthy of very serious consideration. To put it precisely: ‘The idea of the modern nation, which could be subject to rational planning, was itself, dependent on the mobilization of sampling methodology’ (201). There is something of circular causation here: the new sample survey techniques were seen as central to both civil servants and sociologists. The consequence of this marriage of convenience was what Savage calls a ‘flat, homogenous, bounded nation … rather than social relations being inherently located in households and localities’ (211).

His view that new kinds of social knowledge, developed from a new social science apparatus, produced new social identities may be somewhat overstated but the argument in general seems to me to be substantially valid.

Whilst time-series analysis of quantitative data is commonplace in the social sciences, the re-analysis of qualitative interview material relating to the same topic over time is not well-established. The efforts of Paul Thompson and others to establish a Qualidata Archive has encouraged researchers to go back to the original interviews and to make their own selections and interpretations. These may then be at variance with the conclusions and analysis of the researchers who gathered the material who were, of course, working in different contexts and may have had different ideas of what appeared to them the sociological puzzles of the time.

Savage focuses on two main caches of interviews: the Affluent Worker project of the early 1960s and the study of shipbuilding workers by Richard Brown later in the decade. He examined 227 household interviews in the former study (together with a further 38 pilot interviews) and 193 interviews with Brown's shipbuilders. We are not told what system of formal analysis, if any, Savage adopted when engaged in this fairly substantial piece of empirical research. Quotations from the interview are given without any clear indication of how typical they might be. Hence, it is difficult to judge the representativeness of the quotations he chooses as evidence to support his own reading of class identities.

The argument assessed and its message revealed

It is Savage's contention that ‘The working class was not associated with particular kinds of employment relations or places of residence, but was seen as normal, authentic people’ (221). Certainly there are many striking examples of people expressing views ‘articulating a strong naturalistic ethic’ (222). This primordial individual is not a ‘social cipher’ (224) but someone who is ‘normal’, compared with a distant and aristocratic elite. It was the emergence of the new technical and meritocratic cadre establishing itself between male manual workers and the remote elite that led to the distinctive cultural resources of these manual workers to be undermined.

Whether or not Savage's thesis is fully justified by the evidence and research strategy he adopts, his final discussion of the remaking of social class identities is a fine demonstration of the best sociological imagination. He strongly asserts:

Thus it is that the removal of the resources of technique and skill from the domain of manual workers, has made the educated middle classes ‘the quintessential autonomous and reflexive individuals of contemporary capitalism … who command the key terms of contemporary normalcy hence masking their privileges and powers’ (235).

This, then, is the nub of it. The pioneers of the 1960s are shown to have demonstrated ‘their true originality, even greatness, in terms of marking a break with powerful currents of gentlemanly social science with its preoccupation with mapping and moralizing’ (238). This, crucially, was the basis of a critical social science that now deserves to be remembered and celebrated in this BSA Anniversary Year of 2011. Mike Savage should be congratulated for raising this issue at such an opportune time.

Having now disentangled Savage's main thesis and explored some of the strengths and weaknesses of the evidence he adduces to support it, what other evidence might he have considered? In passing, as we have seen, he distances himself clearly from Bourdieu and Halsey, but he does, surely, neglect other important ingredients involved in encouraging the emergence of the post-war technical and meritocratic elite. His emphasis on Mass Observation is, to my mind, not fully justified; perhaps just as significant was Tom Harrison's other project, namely the publication of Picture Post throughout the 1940s. The recurring theme of ‘When We Build Again’ and of renewal and change entered millions of homes every week. The idealistic ethos of the post-war period with the New Towns Act and the conscious search for ‘community’, inspired the newly socially mobile to look positively towards change, innovation and technical skill. New Society was important but so also was Picture Post in helping to create, in an earlier period, the fertile context in which a new type of technical rationality could later grow.

Secondly, I believe Savage underestimates the importance of the impact of American sociology at Cambridge and LSE in the 1950s. High-fliers in political science, history or geography typically won Harkness scholarships or similar to the most prestigious American universities. As W G Runciman has said ‘I learned my sociology not in Britain but in the United States’ (Halsey, 2004: 219). The first book I was encouraged to read by an American graduate student when I arrived at the LSE was R K Merton's Social Theory and Social Structure. The University Library at Cambridge, with its easy access open shelves had a remarkably large stock of classic and contemporary American sociology. The undergraduates I supervised at Cambridge at that time were encouraged to read American journals by their lecturers and examination questions were likely to be focussed on American debates. Post-graduate sociology was predominantly American sociology and few of us took the well-meaning British studies of the 1950s that seriously. What could compare with C Wright Mills?

Thirdly, Savage could, perhaps, have paid more attention to the social origins of the 1960s sociologists. Those rejecting the lower-middle class backgrounds of suburban respectability or the cloying matriarchal self-satisfaction of the respectable working class, created their own kind of scientific identity. This thesis, was, I think, implicit in the article by Box and Forde (1967) and deserves more serious attention.

I found many small typographical and bibliographical errors too numerous to mention, suggesting a rush to publish. In the more leisurely days of the 1960s manuscripts were passed round in typescript to get colleagues' comments, sometimes delaying publication for up to a year. This is a luxury that no doubt Savage could ill-afford, but it might have made it a much better book. Also, a reader not knowledgeable about the full corpus of Savage's work might have difficulty in recognising the true significance of what he is trying to convey. Nevertheless, on balance, Savage should be congratulated for the development of his challenging thesis, full of perceptive, original, but sometimes highly controversial, insights, that now deserves to be widely discussed and developed.

One is left with the question of who will read this book? Students working in their little pools may not think it has much to do with them and, I fear, many of their teachers may see it as not much different from exploring an old family photograph album. I do hope that I am wrong. This book deserves to be poured as a marker fluid into the full flood of contemporary liquid sociology and read by the ‘autonomous and reflexive individuals of contemporary capitalism’ (235). Surely one message of this convoluted, and yet highly challenging book is – ‘Sociologist, know thyself … the unexamined sociologist is not worth having’!