Developments in British Sociology as Shown in British Sociology Journals
by Charles Crothers
Sociological Research Online, 16 (3) 13
Received: 1 Apr 2011 Accepted: 17 Aug 2011 Published: 31 Aug 2011
To provide a factual foundation for understanding of the trajectory of the development in British sociology a content analysis of the journal articles in the main generalist British sociology journals is provided. This contributes both an overall picture, and allows an account of contrasts between the journals. Attention is focused on the extent to which the content differs between British and other authors (and more detailed geographical breakdowns and other aspects of authorship) and how content has changed over time. Finally, how the research outcomes are related to some of the characteristics of producers and producing departments are assayed.
Keywords: Authorship Patterns, Bibliographical Databases, British Sociology, Fields Within Sociology, International Sociology, Journal Articles, Regional Differences
Introduction1.1 To engage with issues of where sociology is going, it is helpful to look carefully at where it has come from. How the literature of sociology in general, or a specific national sociology (in this case 'British' sociology), has developed over time, how it compares to other disciplines or other national sociologies, are important descriptive questions in sociology's monitoring of, and reflection on, its own histories and futures. A descriptive account of developments in British journal-article sociology needs to attend to the research problems examined, and to the theories, methods and data deployed, as well as the extent to which different components of research infrastructure have supported the outputs. This, in turn, should be buttressed by explanatory accounts of why the literature has taken the shape it has. An overview of some evidence relevant for developing a perspective on British sociology is provided in this article.
1.2 When sociology's disciplinary future, current situation or history is discussed, we sometimes risk engaging in its analysis without adequate consideration of relevant theoretical and/or methodological issues to which we would give more attention if studying other areas. The data provided here are offered as a bulwark against speculation, and to offset the necessary limitations of insider accounts, as well as to raise interesting questions that deserve answers. A considerable body of descriptive material is examined for the light it throws on such issues as subject-area content and how this has varied over time, and the national, regional and organisational affiliations of authors. The extent to which the three main journals have differed from each other is also considered, and some potential causes of the empirical trends found are discussed.
The Literature on the Literature of Sociology2.1 Some similar analyses have been made by other writers. In the heyday of the 'sociology of sociology' such analyses were prevalent in America, partly because of journals devoted to this topic, although much less so in, and for, British sociology. But a foundational study was carried out by Collison and Webber (1971), who analysed contributors and contributions to the three core British sociological journals for 1950-70, and related data on membership of the British Sociological Association and its Executive Committee to these publication patterns. They found that contributions to journals came mainly from men, older sociologists, university staff rather than non-academics, and those of higher academic rank. Whereas contributions from the London School of Economics and Liverpool University were of great but decreasing relative importance, those from Cambridge and Leicester Universities were of growing importance. They also noted that "..the contributions of American and other foreign authors are strikingly numerous". Using the categories of Sociological Abstracts, they found that the largest topic area was 'sociology: history and theory', followed by 'social differentiation', 'complex organisations (management)', 'sociology of education' and 'political interactions'.
2.2 What could be seen as a recent partial update (Halsey 2004) of this early study is limited by its small sample size, but is of some interest. A sample of articles from the three general journals in the first year of each decade was analysed in terms of the influentials cited, and then of patterns over time in popular areas, methods used and ideological approaches. Halsey finds that only eight topics from a 38-field categorisation of sociology stood out: "stratification (including social differentiation, class, status, mobility and occupational scales), social theory (papers that do not use or discuss sociological data), social policy, political sociology, religion (including paranormal behaviour and beliefs), education, economic organisation (including industrial relations), occupations and gender" (2004: 184). The time series shows that the key topics fluctuated in importance across the near-century studied. However, he ponders whether the pattern shows that empirically oriented "...old orthodoxy is being replaced by social and cultural theory" (2004: 186). In terms of approach, "The evidence is of a gradual displacement of empiricism by interpretivism. At lower levels there was also a decline of functionalism from around 1970, a rise and subsequent fall of Marxism and Weberianism in the 1980s, and the appearance of feminism in the 1970s, followed by postmodernism in the 1990s" (2004: 194. cf. Matthewman & Hooey 2006).
2.3 Other journal content analyses have documented what their writers see as limitations in sociology, in effect addressing policy issues for sociology. Methodological themes have been emphasised, with studies - including those of Bechhofer, 1981 and 1996; Bulmer, 1989; and Payne et al. 2004 - which note and deplore the shortage of quantitative work by British sociologists. Platt (2007) shows a steadily rising tide of women authors and articles on 'female' topics – increasingly contributed to by women authors - accompanied by a modestly increasing dominance of qualitative articles – from both men and women. A scatter of other studies (many in rather distantly published journals) appeared in the 1970s and 1980s (for example, Armer 1987; Chauhan 1979; Garnett 1988; Glenn 1972; Hedley 1984; Korfias 1970; Oromaner 1970). Vanderstraeten (2010) has contributed one on journals and practices in the Low countries which further builds up opportunities for comparison. These studies have shown, inter alia, that British journals (usually compared to American ones) are more open to overseas authors, but have been broadly similar in the range of topics investigated compared to the sociology journals of other jurisdictions. Where possible data from these sources are drawn on for comparison with this, or to extend its range.
2.4 The recent celebrations of several 60th anniversaries have led to interesting reviews which contribute more interpretive material. The BJS (Heidensohn and Wright, 2010) republished several important articles from each decade, together with updating commentaries, and provided an overall rather broad periodisation (which emphasises images of sociology's continuing sense of crises) as follows:1950s: firm foundations; 1960s: a discipline in ferment; 1970s: continuity and crisis; 1980s: a decade of eclecticism; 1990s: disintegration and disarray?; 2000s: sociology in a new century.
2.5 Other broader commentaries (varying in the extent to which they are grounded in systematic evidence) include those of Morgan & Stanley (1993) which reports on the major debates in Sociology in its first 25 years and offers an alternative view of collective patterns in sociological knowledge production. A special issue was edited by Holmwood & Scott (2007) on 'Sociology and its Public Face(s)', with a later rejoinder by Crompton (2008). Osborne, Rose & Savage (2008) discuss 'Re-inscribing British sociology', which Savage (2010) extends using a broad framework in which he shows some of the interactions between social research methods and social identities, as well as charting the supersession of gentlemanly by professional sociology. Holmwood (2010) laments sociology's unfortunate fallen status compared to other disciplines, its embedding in interdisciplinarity and the impact of audit culture. Other studies of sociology/sociologists have used demographic data, as does Wakeling (2007) to examines the racial composition of UK sociology. Disciplinary histories (e.g. Platt 2003) are also valuable sources for indicating aspects of the institutional framework within which sociology has been pursued.
Methods3.1 This article uses data on the contents of the main general British journals since 1952 to throw light on developments in British sociology and its journals. This starting point is set by the availability of data from Sociological Abstracts (now Proquest's CSA Sociological Abstracts and sometimes referred to as Sociofile.). This bibliographical data-base, established in 1953, abstracts and indexes the full range of the international journal literature in sociology and related disciplines. It covers over 1,800 serials publications, and more recent extensions also provide abstracts of books, book chapters, dissertations, and conference papers although this coverage is uneven. Major advantages of Sociological Abstracts (SA) are its broad coverage, which determinedly covers all journals within the boundaries of the discipline, and also its classification of article content according to a formally-developed scheme, and also to a thesaurus of subject terms.
3.2 The journals studied are the British Journal of Sociology (BJS), Sociological Review (SR) and Sociology. Clearly these do not cover all of British sociology, or even all of British general journal sociology. (Later additions include Sociological Research Online, which has also become important, as well as the more specialised Work, Employment and Society, and the recent Cultural Sociology.) But equally clearly they have been very important to that, and their changing character has a wider significance which justifies using them as a measure of the development of British sociology. The full run of all three journals is covered, thus allowing decade-by-decade comparisons. Recently, the ownership of Sociological Abstracts has passed on to Proquest, and their new formatting prevents easy updating for more recent articles of their classification of fields. Hence the analyses stop in 2006, and are based on some 3500 articles - some years earlier than the summary data, which could be more readily updated up to 2010.
3.3 The changing context of these journals places potential limits on this study. The more recent decades of British sociology have been quite different from the decades immediately after World War Two (cf. Savage, 2010). In earlier phases of its development the three journals represented the UK in a world set of journals which overwhelmingly had a national orientation, and where inter-disciplinary fields were also usually represented by a single national journal for each field. More recently, there has been a huge increase in the number of more specialist journals, many of these claiming to be international in their orientation. A wider range of specialist sociology and interdisciplinary journals now compete for writers' and readers' attention. But, despite the widening range of outlets, the core UK sociology journals are where one might expect any expression of a UK national sociology, with a concern for general societal and sociological issues, to continue to be expressed. That function of the core journals can continue, even if special-interest content migrates to other journals. Since it would require a far wider scan of journals to encompass the whole journal output of UK-based sociologists (and of other national sociology communities which contribute to UK core sociology journals) this research has had to draw its boundaries more narrowly.
3.4 The extent to which focusing on journal content provides coverage of UK Sociology can be put into context using the 2008 RAE data (HERO, n.d.). It is possible to calculate that, for the 34 units involved in the 'Sociology' category, 22% of 8300 research outputs reported were books, 27% chapters, and 43% articles. (The remainder comprised conference contributions and other material). Overall, for all disciplines combined the proportions were 11%, 15%, and 59%. These figures are affected by the limitation that the data represent only the 'best four' contributions put forward in the RAE exercise, but it is clear that much of the higher quality research output is captured and only a small proportion of lower quality. There may have been changes in the output range over time; this needs to be further investigated. However, it is apparent that journal article sociology is of high importance.
3.5 The convenience of downloading data from SA allows the whole population of articles to be studied rather than the smallish samples and limited time-frames, of almost all the previous studies. (The SA's definition of 'journal articles' includes some critical comments and symposia that might not be included in other studies; some 100 of the articles included here appear to be of this type.) The subject-matter content of each article is coded into a 50-category classification scheme, which provides some understanding of how sociology has been distributed among fields and topics. However, the scheme is awkward, partly because it has developed in a somewhat ad hoc way over a long period, so it has been reworked into a more conceptually coherent framework for this study (see Table 3). The classification is ordered from broader topics to more specialised ones. It begins with codes concerning the discipline as a whole, including its theories and methods, and then deals with broadly categorised approaches. The main body of the classification system consists in the various specialty fields, and the final grouping concerns social policy applications of sociology.
3.6 SA provides two codes for each of just over one in ten articles over the whole period, with the proportion double-coded over the last two decades covered twice that in the earlier two decades. It is likely that this is largely a product of different coding practices over time, but it may also reflect increasing complexity of material, or material which cuts across traditional boundaries, requiring multiple codes. The three journals do not differ markedly in the number of codes assigned to their articles. A problem hard to deal with in any such classificatory work is that key terms, or the understanding of key terms, can change markedly over time, while perhaps retaining some more general continuity. For example, as Outhwaite and Platt (1987) point out, '..The obvious key words to use in subject classification change: 'embourgeoisement' fades away, 'labour process' appears on the scene'..
3.7 The second data field extensively examined for this paper concerns author affiliation. Unfortunately this is usually provided only for first authors, so here we are limited by that. The country given in the data has been coded and then grouped into geographical regions. Scotland, Northern Ireland, and Wales as author affiliations have been separately identified from those for England. It is more difficult to code the immediate organisational context of authors, since the form in which their addresses are stated varies considerably. However, given the frequent concern about differences between the productions of those working outside academia, and of those working in 'mainstream' or in cognate departments or in research centres, these contexts have been separately identified. It has been possible to track the publishing activities of departments over time, which adds an interesting organisational dimension.
3.8 Several immediate issues concerning the fit between the 'sample' and the UK journal article 'population' must be confronted. On the one hand, British journals provide a site for publishing by a wider range of sociologists than those domiciled in the UK – though identifying their contributions and their characteristics provides a useful opportunity for investigating the international role of UK journals and differences between British and other sociological research outputs. On the other hand, British (or UK-based) sociologists also publish overseas – a topic for further investigation, especially given the opening up of publishing opportunities compared to previous decades, but a priori less central to discussion of British sociology as such. Gingras and Warren (2006) show, for a few key US and Canadian journals, that over the 1981-2003 period the number of papers UK-based authors published was 31 in the American Journal of Sociology, 18 in the American Sociological Review, 13 in Social Forces, 15 in the Canadian Journal of Sociology and 11 in the Canadian Review of Sociology and Anthropology, as opposed to 454 in BJS. Had they included the other two UK mainstream journals they would have found nearly 1500 articles published there: the 88 published in these key North American journals constitute some 6% of this overall figure. Earlier, Collison and Webber (1971) report that UK contributions to core US sociology journals ran at under 1%. This reinforces an argument that British sociologists concentrate much of their publishing with UK journals, but that their publication range is widening over time.
3.9 Another issue is who counts as 'British'. Many British-born sociologists have had spells overseas, sometimes as permanent migrants but sometimes retaining intellectual links or eventually returning, while numbers of sociologists of non-British origin are employed here. Further work could break open the 'organisational affiliation' information to trace the careers of sociologists inside and outside of jurisdictions. Sociologists based overseas could also be seen as contributing directly to 'British sociology' if defined as sociology about Britain; some have indeed worked on British data, and this has become easier with the increasing availability of data archives permitting secondary analysis of survey data. Despite these difficulties, our operational definition here of 'British' work has to be that it comes from a British address. Finally, not all the articles studied have been written by professional sociologists, although some of those who appear not to have that identity may represent the spread in the penetration of sociologists through a wider range of departments and other organisations. It has, nonetheless, been assumed here that articles which appear in what are formally British sociology journals can be taken (at least broadly) as representative of British sociology.
Background: UK sociology journals4.1 The Sociological Review has a long history, beginning in 1908 with its new series, based at Keele, starting in 1953. The British Journal of Sociology was founded at LSE in 1950, and the BSA's official journal Sociology started in 1967. More recently, a range of other, usually more specialist journals - although leavened by one or two general ones - has been added to the British publication order (see Table 1). There are now at least five UK generalist sociology journals, and well over a dozen broader or more specialised ones with considerable sociological content. (See Halsey 2004: pp183ff. on UK sociology journals, and also Platt 2007.) The UK contingent is flanked after 1991 by the Irish Journal of Sociology, while the Scottish Journal of Sociology published some 55 articles from 1976 through to 1980. These are not included, but it is possible that their availability may have affected what is published in the journals covered.
4.2 According to the Journal Citation Report (JCR) for 2010 (the latest year available), British sociology journals (again, see Table 1) are in the middle to top grades of the international pack. The three mainstream journals (especially BJS) have a significant 'impact factor' (and for BJS also high 'immediacy') and also continue to enjoy readers' attention (citation half-life) over moderate stretches of time. We may note that a result which can be derived from Table 1 is that the three journals covered here provide just over one-sixth of the over 600 'places' per annum for articles in this set of journals.
|Table 1: Journal Citation Index Results 2009|
Definition: Impact factor: citations in 2009 to recent articles -2007 and 2008 - divided by number of recent articles.
Source: Journal Citation Index
(i) Numbers5.1 Table 2 shows that according to the Social Science Citation Index (SSCI) just over 4000 articles have been published in UK sociology journals. The 10-year periods shown included fit the original time-frame of the data-analysis and the founding of Sociology. Since the advent of Sociology, numbers of articles published increased substantially in the second decade, but have remained fairly steady since (at 75-80 p.a.) with a sudden sharp increase since the mid-2000s. The three journals tend very much to level-peg and their share of the output amongst the set is even and constant, apart from the most recent decade in which Sociology increased its output whereas BJS fell away in number of articles.
|Table 2: Numbers of Articles (by Period and Journal)|
Source: Sociological Abstracts
5.2 Overall, just over one quarter (28%) of the articles (across the whole period of this study) are authored from outside Britain. Apart from a sharp drop in the most recent period (which may be an effect of the RAE exercise, with UK authors striving to maximise their publications by the deadline?) this proportion has been constant over time. But the three journals are markedly different: BJS is flooded with non-British authors (39%) compared to SR (26%) and especially Sociology (17%). This may owe something to the (anecdotally apparent) common belief abroad that 'The British' BJS must be the leading, or official BSA, journal. Among the authors not domiciled in the UK we find that the largest concentration is from North America, with significant minorities from Western Europe, Australasia and Israel. Within the UK, 85% of contributions come from England, a proportion which corresponds very closely to its proportion of university staff in 1997.
5.3 Data on authors also show the production of articles from authors with varying organisational affiliations: those in institutions with more applied traditions, in research centres as opposed to teaching departments, and outside academia. It was possible to classify those outside academia as from Government agencies, other organisations (e.g. hospitals), stand-alone research institutes, or those giving private addresses, and those inside academia as from research centres, units etc., those in cognate departments (e.g. social work, education), those affiliated with a polytechnic (when those existed) or a college of higher education, and as a residual category - those not falling into any of the above who have been assumed to be in 'mainstream sociology departments'. Given the particular history of sociology within UK polytechnics (up until their absorption in the overall University sector from 1992) this dichotomisation is problematic. Their role in sociology education was large, and many staff were heavily involved in the development of UK sociology; this topic is another needing further attention. It was not possible to carry out this exercise with sharp precision, as there is insufficient information in many cases, and there are other difficulties such as organisations changing their names and/or type over time. Further, it seems likely that some categories may be inflated - in particular those claiming affiliations from university research centres, some of which may be merely nominal.
5.4 The overall picture is dominated by contributions from 'mainstream' departments. Nevertheless, it can be seen that there have also been substantial inputs from polytechnic staff, cognate departments, research units and stand-alone organisations. Each type of affiliation is linked to other characteristics. Across most of these categories the proportion of articles from UK authors is similar over time, apart from a higher proportion of polytechnic staff and a lower proportion from stand-alone research organisations in earlier periods. The three journals differ slightly; SR has larger shares from polytechnics and stand-alone research institutions, whereas Sociology has a much higher share of its contributions from research centres. 'Mainstream' departments have been steady in their proportionate contribution (after the first decade covered here), while polytechnic contributions (and to a lesser extent colleges) have fallen off more recently as they have become absorbed into the main university system. Contributions from government organisations and those giving private addresses have steadily declined. On the other hand, contributions from staff in research centres have sharply increased, as has the share from cognate departments. Contrary to some beliefs, there has been only a slight tendency for those from research centres to write within larger authorial teams. Some of this expanding share undoubtedly is a result of the increasing fragmentation of sociology, as sociologists teach and write within a wide range of types of department.
5.5 Table 3 provides data on the content of the articles. It is conventional to reduce such tables by collapsing categories, but this over-simplifies the range of topic areas covered, so a full presentation has been included. The proportions in each category are reported in relation to both total number of codes (column 3) and total number of articles (column 4). Big categories are those found in earlier studies: history and theory, economic and political, stratification, and family and cultural sociologies.
|Table 3 Classifications (multiple responses) : up to 2006|
Source: Sociological Abstracts: 3 Main UK sociology journals
5.6 The earliest decade (shown in a breakdown of Table 3's data not included here) looks quite different from the subsequent period, when the categories in which sociological output was produced seem to have shaken down. Growth areas include a huge take-off of feminist sociology and group interrelations (e.g. ethnic relations) together with networks, theory/history of ideas, macro-sociology, environmental sociology and the new area of sports/leisure sociology. There has been a fall over time in welfare topics, sociological practice, organisations, communities, social change, education, and military sociology. Radical sociology and sociology of art/language seem to have peaked in the late 1970s and late 1980s respectively; on the other hand, cultural sociology and political sociology seem to be recovering a strength they held in earlier periods. Obviously the advent of new UK specialist journals may have affected the distribution of articles in relation to cultural analyses and work-related studies.
5.7 Comparing the journals, the patterns are in general remarkably similar (as Collison & Webber also found). However, 'methods' as a topic area seems captured by Sociology, whereas more theoretical and related pieces (including coverage of macro-sociology and social change) have appeared (relatively and absolutely) more frequently in BJS. Sociology has not attracted cultural sociology to the extent of the other two journals, although it has run more articles on networks. Economic sociology is under-represented in BJS. UK-based authors' articles are more often on methods or on leisure and sport, whereas overseas authors concentrate relatively more on history/theory, organisations or social change, but the differences are very small.
5.8 There is a sociological literature treating differences between England and other countries within the UK, but the interesting issue of how far this might be reflected in the sociologies produced in these regions does not seem to have been raised. A partial exception may be the literature on the characteristics and causation of the Scottish enlightenment 'proto-sociology', and more recently on the need for a sociology of Scotland. Data from the Scottish and the Irish Journals of Sociology might have altered this pattern if it had been included. Unfortunately the smaller numbers for the 'peripheral' countries preclude sharp analysis; in some cases a single prolific author can have a dramatic effect. Overall, there are no significant differences.
5.9 While there has been movement in journal topic coverage over time, this has been markedly small in its extent: there are some drifts, but few dramatic changes at the level of broad categories. It is possible that alternative ways of conceptualising categories would reveal more dramatic shifts, but it seems more likely that (as in some other areas) the dramatic shifts apparent to some observers seem to fade once subject to evidence-based scrutiny.
5.10 Who has been writing these articles? The names of all first authors (within and across articles) were pooled, and the list was extensively corrected for the various factors which may interfere with correct linkage of authorships over time. Potential errors which were examined include change to a married name, different formats for first names and/or initials, and some spelling and /or layout issues. While extensive errors were found and corrected, the effect on the representation of the overall authorship pattern was not great: partly because most highly-published authors had at least a main form of their name, and because most of the error accumulated amongst lower publishing authors.
5.11 Altogether there have been some 3620 authors involved in producing these 4500 articles. However, the distribution is markedly skewed, with both mean and mode at one article per author, and the upper quartile is two. No more than the top 1% have produced as many as nine articles in this period. On the other hand, the sheer range of contributors is amazing. It is important to remind ourselves again that publications in books, or in journals other than these three, are not included in these figures.
5.12 The three dozen high producers (with ten or more articles in the journals studied) were extracted for further analysis. At the top of this group is Bryan Turner, with 24 articles; the average is 12.5. The high producers have nearly all had long publishing careers - the average length between first and last publication to date in these journals was over a quarter century. Almost all have been publishing into the last decade covered. A modest proportion (about 20%) are younger scholars, while the rest are either in retirement or nearing it. Few were women (just under 25%), and a smaller proportion (under 10%) were either non-British or, while British, had been involved in career moves that had taken them out of the UK for substantial periods, in some cases maybe permanently. We need to bear in mind, in interpreting such figures, that the composition of the underlying UK sociology community has changed. Its cohort representation has varied at different periods because of recruitment patterns, and its gender composition has changed (see Platt, 2007).
5.13 The distribution of articles published by each author across the three journals was noted. Overall there is a remarkably even spread, but it is clear that no individual author's output divided up evenly, or even came close to doing so. We also recorded average involvement in authorial teams. Individual writers show rather different styles of team membership, but the higher producing authors in this listing are no more likely than the average author in the whole study to have been involved in multi-authored papers.
5.14 Although the pattern is difficult to summarise because of its complexity, it was possible to identify some differences of content in terms of authors' organisational affiliations. For example, research units and government departments were more likely to contribute on methods, and less likely to contribute on theoretical topics. There are particular clusters of topics associated with research centres and stand-alone centres which tend to be visible topics (e.g. family, the work environment, gender issues and stratification) which perhaps are more likely to capture funding and policy interest. Polytechnics, cognate department and government research authors have tended to show similar interests. It might seem that the basic foundations of general sociological work are carried out by people from mainstream departments, with this being added to by the specialised work of the more specialist organisations.
(v) The Rise and Fall of Departmental Publishing Profiles
5.15 With some difficulty it was possible to assign to departments their publishing profiles over time. Again this exercise could not be carried out with sharp precision, as there is often insufficient information, and there are other difficulties such as organisations changing their names and/or type over time. Further, it seems likely that some categories may be inflated - in particular those claiming affiliations from university research centres. The profiles identified were then related broadly to the size of the various departments (1997 was taken as an indicator year), and to Collison and Webber's earlier study. Given the difficulty of longitudinal study, only broad results are reported here. The modal department size in 1997 (with one third of the total) was in the 10-14 range, with about a quarter each in the immediately smaller (five to nine members) and the immediately larger sized groupings. (The largest department had 23 staff). Over the 50 year period covered, those from higher producing departments published up to 75 articles in the three journals covered. Of the 30 larger producing departments the average ratio of publications to staff was 2.5, with a few departments producing five times their number of staff at 1997.
5.16 To provide a summary picture, we can list as high producers LSE, Essex, Oxford and Lancaster, all also with high output ratios in relation to their 1997 staff numbers. Then come Keele, Cambridge, Queens Belfast and Surrey, also with high ratios. The next group has produced an output of around 30-40 articles but with variable output ratios (high relative producers are marked with an asterisk): Kent *, Warwick, Sheffield*, Manchester, Leeds, Cardiff*, Edinburgh, and Sussex*. The final group mentioned here includes Glasgow, Leicester, Edinburgh, Bristol, Manchester, York, Bristol, Salford, Open and Brunel. In comparison to the earlier league table, LSE retains its high slot together with Oxford, but is joined by two departments which had barely been established in 1971: Essex and Lancaster. Apart from Cambridge and Manchester, the next two groupings are also upwardly mobile. It is not surprising that when so many new departments have entered the scene, and their size and composition have fluctuated over time, there have been important changes in their research output rankings. The extent to which those represent strictly departmental characteristics, or are related to topic interests and related publication strategies, deserves further study.
Interpretations:6.1 Halsey (2004: 194) suggests that the different approaches he found in the three journals owe much to the decisions made by individual editors. But before articles can be submitted for consideration the sociology has to be produced. Moreover, to make confident assertions about the relative importance of the editorial role in selecting the content areas of what is published one would need to know what work had been submitted, and data on that is not publicly available. However, there is some evidence that journals (especially during particular editorial tenures) may be seen as more or less favourable to certain topics or approaches, even if it has been quite common for the editors to publish denials of this, and there is also some recruitment of particular content through special issues. Sociology's recent advertising for proposals of special issues could be seen as an attempt to make such decisions more responsive to community interest. However, this is another topic needing more careful separate attention. Editorial boards, insofar as their responsibility is for refereeing, might be seen as at least equally relevant, and here one could draw attention to the dominant role of departmental members in the boards of the two institutionally-based journals (BJS and SR); however, it is not easy to show that this has in practice made a difference to the outcomes (for example, Platt (2003: 63) found that there was no clear pattern of articles by women appearing in proportions corresponding to the proportion of female members of editorial boards); the broad similarities on various points found here suggest otherwise.
6.2 General trends, some specific to sociology, which have affected UK academia may be glimpsed more opaquely behind the changing patterns of journal outputs. These would clearly include changing institutional arrangements amongst universities (including the 'promotion' of the 1992 universities), and within them (e.g. some restructuring of sociology departments into wider configurations, and the widespread development of sociological interests within interdisciplinary fields of study), funding possibilities (with the ESRC and its predecessor undoubtedly enabling the 'lifting' of sociological research, and also some shaping of its attention-span), public administration ideologies (e.g. an increased interest in evidence-based policy, with some government funding support), by the ways incentives to publish have been reshaped (especially by the RAE exercises) and by trends within the social sciences and sociology in terms of the increased array of publication outlets, and the changing popularities of topics, theoretical approaches, methodologies and methods. Many of these points could be explored further in more detailed work; at this stage we cannot offer data on the relative influence of such factors.
6.3 Should these perhaps be seen as international journals? It is difficult to assess the extent to which the project of a UK national sociology can be glimpsed through their pages. Globalisation has clearly directed British attention to wider spheres - and in the journals this is reflected in the increasing recruitment from abroad of editorial board members - and there has clearly been some interest in the cosmopolitan task of contributing to the general stock of sociological knowledge. Cosmopolitan ambitions could be seen as implicit in the naming of the core journals, only one of which has a UK-linked term in its title - as do none of Sociology's sister BSA journals.
6.4 Much of the content of the three core UK journals is couched in a mix of local and cosmopolitan terms, with the local data being used to draw more cosmopolitan conclusions, but even more is couched at particularly local levels, with tightly focused interests which are clearly sub-national in character. However, some commentators have considered that the UK has only partially achieved the articulation of a national sociology, though clearly some of the ingredients are there: for example Perry Anderson (1992) argues that the rise of UK-related historical sociology in the books of Mann, Runciman, Giddens and Gellner is an indication that sociology has retrieved a role as a major contributor to UK academic culture. Levine (1995) has also traced a more historical lineage of national social thought. But few UK sociologists have provided a self-conscious explicitly UK sociology (except in introductory textbooks), though some at earlier periods certainly had the general ambition to provide a sociology that was not unduly dependent on US models. Reflection on the range of topics covered does suggest that the ingredients for a national sociology have been laid down - for instance, it has been suggested that British sociology has been especially preoccupied with class and that this reflected national characteristics - but few articles address this explicitly. It seems likely that national sociologies will resemble each other to the extent that nations face similar issues (economic development, social cohesion, family changes, etc), even if the concrete specificities are different. The questions implied can only be answered by bringing to bear comparative data from other countries.
Conclusions: Interpretations in Search of Further Empirical Investigation7.1 This article has sketched a broad outline of some important aspects of UK sociology and its changes from the 1960s. From the evidence presented here, there are few differences in British journals between British authors and those non-British authors publishing in British journals, and few between authors from different parts of the UK. Separating British from non-British contributions to the three journals allows the opening up for investigation of the question of the place of British journals within the international division of labour of world sociology. Clearly the UK journals receive considerable attention from North American, European, Australasian and Israeli sociologists. In some part this may be because authors see the UK as an alternative focus of sociological leadership, where 'non-American' sociological material can be safely submitted and published; for some in the ranks of this international contingent, British sociology journals are undoubtedly a step higher up the hierarchy of journal prestige than those from their own country. There is undoubtedly also a supply-side push, with publishers suggesting to editors that they accept more overseas-sourced articles in order to increase the international attraction of their journals.
7.2 The data presented provide a factual summary of several of the key features of leading British sociology journals and their articles, which disconfirm some common perceptions, and need to be taken into account in future work on the development of British sociology. They also draw attention to a valuable resource of broader quantitative data than have commonly been used, as well as raising many interesting points which merit further and more detailed work.
AcknowledgementThis paper was written as part of work on a joint project with Jennifer Platt, who is thanked for her assistance. To avoid conflict of interest, Prof. Platt did not participate in the editorial decision-making concerning the inclusion of the article in this section.
ReferencesANDERSON, Perry (1992) English Questions, London: Verso
BULMER, Martin (1989) 'Theory and Method in Recent British Sociology: wither the empirical impulse?' British Journal of Sociology vol. 40 pp. 394-418
CROMPTON, Rosemary (2008) 'Forty Years of Sociology: Some Comments'. Sociology vol. 42 pp. 1218-1227
GARNETT, Richard (1988) 'The Study of War in American sociology: an analysis of selected journals 1936 to 1984', The American Sociologist, vol. 19, issue 3 pp. 270-282.
GLENN, Norval (1972) 'American sociologists' evaluations of 63 Sociology Journals', American Sociologist vol. 6 pp. 298-303.
GINGRAS, Yves and Jean-Philippe Warren (2006) 'A British Connexion? A quantitative analysis of the changing relations between American, British and Canadian sociologists', ms.
HALSEY, A. H. (2004) A History of Sociology in Britain: science, literature and society Oxford: Oxford University Press
HEDLEY, Alan (1984) 'Social generalizations: biases and solutions' International Journal of Comparative Sociology vol. 25, issue 3-4 pp. 159-172.
HEIDENSOHN, Frances and Richard Wright (2010) 'Special issue on The British Journal of Sociology at sixty'. British Journal of Sociology: Supplement vol. 61 pp. 1-6
HERO (Higher Education Research Office) 1996 Research Assessment Exercise Database <http://www.hero.ac.uk/rae/rae96/database/index.html>
HOLMWOOD, John and Sue Scott (eds.) (2007) 'Special issue on Sociology and its Public Face(s)' Sociology vol. 41 pp. 779-783.
HOLMWOOD, John (2010) 'Sociology's misfortune: disciplines, interdisciplinarity and the impact of audit culture' British Journal of Sociology vol. 61, issue 4 pp. 639–658
KORFIAS, Michele (1970) 'Aspects of Sociological Thought in Great Britain' Annali di Sociologia vol. 7 pp. 27-32.
LEVINE, Donald (1995) Sociological Traditions Chicago: University of Chicago Press
MATTHEWMAN, Steve and Douglas Hoey (2006) 'What Happened to Postmodernism?' Sociology, vol. 40, issue 3: 529-547
MORGAN, David and Liz Stanley eds. (1993) Debates in Sociology Manchester: Manchester University Press.
OROMANER, M.J (1970) 'Comparison of Influentials in Contemporary American and British Sociology: a study in the internationalisation of sociology' British Journal of Sociology vol. 21 pp. 324-332. [doi:://dx.doi.org/10.2307/589137]
OSBORNE, Thomas; Nikolas Rose and Mike Savage (2008) 'Editors' Introduction Reinscribing British sociology: some critical reflections'. Sociological Review vol. 56 issue 1 pp. 519-534,
OUTHWAITE, William and Jennifer Platt (1987) 'Introduction', Sociology: 20 year Index, British Sociological Association
PLATT, Jennifer (2003) The British Sociological Association: a Sociological History, Durham: Sociology Press.
PLATT, Jennifer (2007) 'The Women's Movement and British Journal Articles, 1950-2004' Sociology vol. 41 pp. 961-975
SAVAGE, Mike (2010) Identities and Social Change in Britain since 1940 Oxford: Oxford University Press
WAKELING, Paul (2007) 'White Faces, Black Faces: Is British Sociology a White Discipline?' Sociology, vol. 41 pp. 945-960 (2007).