The focus of this article is on the social cohesion, or its lack thereof, in post-Brexit Britain. Theories of national identity and nation building emphasize how a nation can be conceptualized as an “imagined community,” with the idea that we feel a sense of comradeship with our fellow citizens—even if we have never met them—and that we accept greater responsibilities toward members of our imagined community than we do outsiders. Building on the earlier work of Renan (Reference Renan1882) and Kohn (Reference Kohn1944), Anderson (Reference Anderson1983, 3) proposed the idea that a nation “is imagined because the members of even the smallest nation will never know most of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of their communion.” According to Miller and Ali (Reference Miller and Ali2014, 238), national identity provides “the ‘cement’ or ‘glue’ that holds modern, culturally diverse societies together and allows them to function effectively…display generalized trust, and to show solidarity…to decide a wide range of matters by democratic means….”
Although this argument has been articulated primarily within political theory, it chimes with the social psychological theory of social identity. This theory holds that the construction of social identities is a fundamental social process, with people who share a social identity giving preference to members of the in-group and derogating members of the out-groups (Tajfel and Turner Reference Tajfel, Turner, Austin and Worchel1979). The national-identity argument also is appealing to British politicians, who have stressed the importance of teaching newly arrived migrants to share British values, introduced citizenship tests, and expanded the Britishness content of citizenship education more generally in schools.
The national-identity argument implicitly assumes that there will be an “imagined” or symbolic boundary to the imagined community. Following Lamont and Molnár (Reference Lamont and Molnár2002, 168), we define “symbolic boundaries” as “the conceptual distinctions made by social actors when categorizing people, practices, and even time and place. Symbolic boundaries…separate people into groups and generate feelings of similarity and group membership.” The substantive issue then arises about whether this sense of being part of an imagined British community has broken down, with competing visions of the nature of the society. In other words, who counts as “Us” and who counts as “Them” in contemporary Britain? Moreover, how permeable are the boundaries between groups? The rise of the movement for Scottish independence clearly suggests one emerging and strengthening symbolic boundary within the United Kingdom. However, the divisions within England over Brexit also suggest that—for some people, at least—the notion of “Us” is more circumscribed and fails to include “Them” in Westminster, and vice versa.
Issues of this type recently have been highlighted by former Prime Minister Gordon Brown, who argued that the United Kingdom is at risk of becoming a failed state and breaking up unless there are deep reforms in the way the country is governed. “I believe the choice is now between a reformed state and a failed state,” Brown (Reference Brown2021) wrote in The Daily Telegraph. He argued that the coronavirus pandemic had exposed divisions between different parts of the United Kingdom. “You’ve got not only the Scottish first minister but you’ve got the regional mayors saying they are not consulted and listened to, you’ve got the Welsh first minister saying their letters are not even replied to by Boris Johnson, you’ve got no mechanism, no forum for coordinating the regions and nations, and I think that the public are fed up” (Brown Reference Brown2021). In his article, Brown largely focused on the constitutional reforms needed to avoid Britain becoming a failed state. However, it could be argued that the underlying causes of the problem go much deeper. Perhaps constitutional arrangements have failed to keep up with the social and economic changes that Britain has experienced in recent decades, in particular with the centrifugal processes that have weakened Britain’s sense of comradeship.
Devolution in many respects was a consequence of the rise of nationalist movements in Scotland, Ireland, and Wales that sought greater independence within—or even from—the United Kingdom. Devolution often is believed to have further exacerbated these centrifugal tendencies within the United Kingdom and especially to have fostered a “little Englander” identity at odds with the previously dominant, inclusive British identity (Ford and Sobolewska Reference Ford, Sobolewska, Kenny, McLean and Paun2018). It is possible that Brexit is further exacerbating these centrifugal tendencies, making Scottish independence and Irish reunification more likely.
Brexit also in part may have been a consequence of and, as with devolution, perhaps also a catalyst for fragmentation within England. There certainly were large divisions across England (greater indeed than the divisions among the four constituent nations of the United Kingdom) in support for Brexit (and, by inference, in the distribution of Leave and Remain identities). Thus, in London, at one extreme, a substantial majority (60%) voted Remain, whereas in the Midlands and North-East, the percentages were reversed with approximately 58%–59% voting Leave (see online appendix table A1).
To be sure, these differences in support for Brexit mirror geographical differences across England in economic prosperity and therefore may not necessarily indicate fundamental divisions in social identities and new symbolic boundaries separating insiders from outsiders. Fisher, Kirby, and Macfarlane (Reference Fisher, Kirby and Macfarlane2021) also found that despite growing economic polarization among the different parts of England, attitudes and values remain quite similar. In contrast, Goodhart (Reference Goodhart2017) argued that underlying support for and opposition to Brexit was a fundamental division between “somewheres” and “anywheres”—a division between traditionally minded local communities with strong attachments to their local areas and a rootless cosmopolitan elite based in London increasingly out of touch with ordinary working people. Similarly, Mattinson (Reference Mattinson2020) described voter sentiments in the former “Red Wall” Labour Heartland seats that fell to the Conservatives in 2019. In these regions, many of which experienced decades of decline in traditional industry, people describe loss of regional identity and have high levels of resentment for London and “Southerners” generally. Ford and Sobolewska (Reference Ford, Sobolewska, Kenny, McLean and Paun2018) advanced a related argument focusing on a “culture clash” between progressive, liberal and multicultural, outward-looking elites and more inward-looking, chauvinistic “Little Englanders” opposed to immigration and diversity.
Our central question in this study, then, is where British citizens draw the symbolic boundaries defining membership of the “in-group” from the “out-group.” Do we continue to find—as the classic literature on the concept of the nation implied—that there is an “imagined community” effectively encompassing and uniting British citizens as a whole? Or do we find that there are internal symbolic boundaries between the different parts of and in different groups within the United Kingdom?
Our central question in this study, then, is where British citizens draw the symbolic boundaries defining membership of the “in-group” from the “out-group.”
DATA AND METHODS
To measure these sentiments toward members of one’s in-group and out-group, we asked a battery of questions about citizens’ perceptions of the different national and regional groups, along with those of the middle and working classes and a range of ethnic minorities (i.e., the classic outsiders). We focused specifically on three items that measured perceptions of whether groups “share my values,” “are people I could get on with,” and are “straightforward and honest.” Shared values have been a major element of the academic and political debates over the “glue” that helps a nation to cohere (Heath et al. Reference Heath, Garratt, Kashyap, Li and Richards2018). The item “people I could get on with” was designed to pick up on Anderson’s (Reference Anderson1983) idea of the imagined community as a broad horizontal comradeship. The item “straightforward and honest” relates to the notion of generalized trust in one’s fellows, which has played an important role in accounts of the way in which national identity can facilitate the provision of collective goods for members (Miller and Ali Reference Miller and Ali2014). These three items provide what we term a measure of “symbolic distance” between each specified pair of in-group and out-group (Lamont and Molnár Reference Lamont and Molnár2002). We then can infer the sharpness or blurriness of the boundaries by the perceived distance between in-group and out-group. The greater the distance and the more it is symmetrical, the sharper and more definite and defined we consider the boundary.
Because of our interest in recording perceptions about several regional and social groups, we took a “split-thirds” approach so that no single respondent gave their view on more than six groups (Richards and Heath Reference Richards and Heath2023). These data were collected online in the spring of 2019 as part of a larger project about Brexit-related attitudes in Britain. A reputable fieldwork agency was used (Heath and Richards Reference Heath and Richards2022) and considerable attention was given to obtaining a representative sample; weights were applied throughout the analysis. As a data-validation exercise, we compared symbolic boundaries between the White British and other national or ethnic groups that have been the focus of previous research in which European Social Survey data use measures of attitudes to different types of immigrant. Despite the difference in the measurement approach, the two types of measure seemed to tally well, thereby validating our measures and providing a useful yardstick for examining the internal boundaries within Britain. These results are reported in online appendix table A2. In tables 1–4, statistical significance is reported at the 95% level based on t-tests. All t-tests compared perceptions of the in-group and out-group from the same set of individuals—that is, within the relevant split-third sample. See the online appendix for details on data structure, missing data, and the sample used in each table.
Table 1 British Respondents’ Perceptions of the White British, English, Irish, Scots, and Welsh
Notes: See the online appendix for sample details. Figures in parentheses show the differences from the perceptions of the White British. N=335–420. We asked about Irish rather than Northern Irish but are aware that this is not a proxy for understanding perceptions of the people of Northern Ireland. Asterisks indicate that the difference is statistically significant at the 95% level.
The data collected in this project are stored as “safeguarded” data by the UK Data Service (https://ukdataservice.ac.uk SN8926) and are free to access with basic log-in details and by signing an end-user agreement.
RESULTS
This section presents our results on symbolic boundaries, beginning with perceptions of the British, English, Scots, and Welsh, and followed by analysis of social class and North–South boundaries within England.
Symbolic Boundaries Among the Nations
First, we examined the perceptions of sample members who self-described simply as British. A British identity historically has been an inclusive identity embracing people from all four of the home nations. We therefore expected this analysis to show relatively small symbolic distances among people who think of themselves as British and the members of the four individual nations of the United Kingdom.
As shown in table 1, there is virtually no symbolic distance between the White British and the English, with percentage-point differences being uniformly small and nonsignificant. This adheres to what Cohen (Reference Cohen1995) termed the “fuzzy frontier” between Britishness and Englishness, and it is hardly surprising given the number of people who hold dual British and English identities (Kenny, Heath, and Richards Reference Kenny, Heath and Richards2021).
The symbolic boundary between the British and the Welsh also is fuzzy and blurred, with British respondents perceiving the Welsh to be more straightforward and honest than their fellow British. This suggests that the majority group has some misgivings about their fellow members, which are repeated in subsequent tables. There is more of a boundary, however, with regard to the Irish and Scots. On the shared-values indicator, for example, the percentage-point gap vis-à-vis both Scots and Irish reached 19 points.
We also can examine the symbolic distances between the three nations of English, Welsh, and Scottish people—not only English respondents’ perceptions of members of the other two nations but also Scottish and Welsh perceptions of the English.
Table 2A illustrates the symbolic boundary between respondents who self-identified as English and those who self-identified as Scottish. In both cases, there is in-group preference, and the mutual perceptions are broadly symmetrical. Thus, from an English perspective, the distance on the “share my values” item between in-group and out-group is 22 points, compared to 27 points from a Scottish perspective.
Table 2A English and Scottish Perceptions of the In-Group and Out-Group
Notes: See the online appendix for sample details. N=223–264 (English identifiers) and 36–41 (Scottish identifiers). Asterisks indicate that the difference is statistically significant at the 95% level.
There are two other interesting features of table 2A. First, we can see that the Scots rate members of their own in-group considerably more favorably than do the English. Thus, 94% of Scots believe that they could get on with fellow Scots, whereas the comparable figure of in-group preference among the English is 10 points lower, at 84%. The differences are even more substantial on the “straightforward and honest” item, at 19 points. These are hints that the Scottish nation is more cohesive, with higher opinions of fellow Scots, than the English nation.
Turning next to the symbolic boundary between the English and the Welsh, there is only a small sample of respondents who self-identified as Welsh; therefore, we need to be especially cautious. Nevertheless, as illustrated in table 2B, the boundary appears to be very blurred with only minor (although significant) percentage-point differences between perceptions of the in-group and the out-group. Perhaps the most noticeable finding was the Welsh lack of trust in the English: only 55% of Welsh respondents agreed that the English are straightforward and honest.
Table 2B English and Welsh Perceptions of the In-Group and Out-Group
Notes: See the online appendix for sample details. N=223–264 (English), 19–21 (Welsh). Asterisks indicate that the difference is statistically significant at the 95% level.
Symbolic Boundaries Within England
We also examined differences within England between Southerners and Northerners as well as the classic distinction between the middle class and the working class, which traditionally has been one of the great dividing lines in British society. However, it is now believed, at least in political contexts, to be much weakened (Evans and Tilley Reference Evans and Tilley2017).
An interesting feature of table 3 is the asymmetry between middle-class and working-class perceptions of one another. Thus, the middle class perceives a relatively blurred symbolic boundary between themselves and the working class, with rather small distances similar to those between the English and the Welsh (and reaching statistical significance only in the case of “straightforward and honest”). In contrast, members of the working class perceive rather sharper boundaries between the two classes, with significant gaps on all three indicators. It is tempting to interpret this asymmetry as related to the relationship of subordination and superordination between the two groups. Experiences of subordination may lead working-class people to be more aware of the differences, whereas the superordinate group may take the subordinate group for granted—and, indeed, may patronize them.
Table 3 Middle-Class and Working-Class Perceptions of Their In-Groups and Out-Groups
Notes: See the online appendix for sample details and definitions of middle class and working class. N=200–230 (middle class), 143–153 (working class). Asterisks indicate that the difference is statistically significant at the 95% level.
Table 4 illustrates the equivalent analysis of Northerners’ and Southerners’ perceptions of one another. Because we do not have a direct measure of Northern or Southern identity, we defined Northerners as those who were residents of the three most northerly regions of England (i.e., the North-East, the North-West, and Yorkshire and Humberside) and who also felt close to their region. We defined Southerners as those who were residents of the core southern areas of London and the South-East and who also felt close to their region. We found marked in-group preferences among both groups but also a marked asymmetry, significantly more so than for the class divide. This asymmetry emerged partly because Northerners had more positive perceptions of their in-group than Southerners (i.e., Northerners’ perceptions of their in-group were almost as positive as those of the Scots toward their in-group) and partly because of their more negative perceptions of the out-groups (i.e., significantly more negative than the Scots’ perception of the English out-group).
Table 4 Southerners’ and Northerners’ Perceptions of Their In-Groups and Out-Groups
Notes: See the online appendix for sample details and definitions. N=148–193 (Northerners), 121–141 (Southerners). Asterisks indicate that the difference is statistically significant at the 95% level.
In other words, Northerners appear to demonstrate rather high levels of in-group comradeship alongside unusually negative views of the Southerner out-group. Why do Northerners have such negative perceptions of Southerners? One possibility is that they differ from Southerners (on average) in their individual attitudes and values. For example, Northerners might be less highly educated than Southerners and, perhaps in consequence, less tolerant of outsiders, or they may be more likely to share exclusively English identities, which tend to appeal to people with a chauvinistic or nativist attitude. However, a multivariate analysis of Northerners and Southerners found that none of these factors can explain the difference in perceptions (see online appendix tables A5–A7).
An alternative possibility is that rather than reflecting differences between Northerners’ and Southerners’ personal characteristics, the explanation may be related to how people are treated or how they feel that they are treated. This is clear among Mattinson’s (Reference Mattinson2020, 28) study participants in Red Wall seats who spoke about the North–South divide “as though they were witnessing a theft…funds were being taken away from them and transferred to wealthier parts of the country, places where the political sun had been shining—London came in for special mention again and again. The resentment cut deep.” In our survey, we asked: “In general, which regions, if any, does the government look after best, and which the least well?” Figure 1 uses the 12 UK regions as the units of analysis and plots the relationship between how well residents in a given region feel that the government looks after their region and their perceptions of whether Southerners share their values.
Figure 1 Perceptions of Sharing Values with Southerners Correlates with Perceptions of How Well Government Looks after Region
As might be expected, people who live in the four southerly regions of London, the South-East, the South-West, and East of England tend to have more positive views of Southerners’ values, but we also can see that there are gradients within both the bloc of four southern regions and the other eight regions. In general, the more negative a region is about how well government treats it, the more negative it is toward Southerners. This is especially true with respect to sharing the values shown in figure 1; we found positive but weaker correlations for “get on with” and “straightforward and honest.” To be sure, causation could run in either direction. However, this suggests that the possibility that perceived power imbalances and unfair treatment may lead to negative attitudes should be taken seriously.
CONCLUSIONS
Britain does not appear to exhibit a deep horizontal comradeship encompassing all areas and regions of the country. Whereas some boundaries, such as those between the English and the Welsh and between the working and middle classes, appear fuzzy and permeable, the symbolic boundaries between England and Scotland are relatively sharp. Whereas the class boundary is fuzzy from the perspective of the middle class, this relationship is asymmetrical. The working classes perceive sharper boundaries with the middle classes than the other way around. Most striking is a sharp but also asymmetric boundary between Northerners and Southerners within England. Northerners in particular have highly negative views of Southerners, with only one third thinking that Southerners “share my values” or are “straightforward and honest.” As such, on the North–South divide, we found that the resentment in the former labor heartlands (Mattinson Reference Mattinson2020) holds with nationally representative data.
We have no data that would enable us to state whether this boundary is a long-standing one or a more recent one reflecting the cleavages that led to Brexit. It well may be relevant, however, that Fisher, Kirby, and Macfarlane’s (Reference Fisher, Kirby and Macfarlane2021) detailed analysis of change over time (using the annual British Social Attitudes surveys) found both widening economic gaps between North and South—especially London since the mid-1990s—and a growing divergence on liberal values. These two factors have been linked to patterns of support for Brexit. In particular, the growing divide on liberal values—wherein London residents have become significantly more liberal than residents in the North (i.e., a gap of approximately 20 percentage points in 2018)—suggests that perceptions that Southerners do not share Northerners’ values in this respect is accurate.
More generally, our results suggest that issues of political integration might apply to the relationships between the Westminster government and the residents of the Northern regions as well as between the government and Scotland. Our results further suggest that popular sentiments and feelings of horizontal comradeship within the United Kingdom are far from united. The legitimacy of top-down rule from Westminster might become increasingly questioned.
DATA AVAILABILITY STATEMENT
The data used in this research are available at the UK Data Service, Study No. 8926, to registered users after signing an end-user agreement. See https://beta.ukdataservice.ac.uk/datacatalogue/studies/study?id=8926. Research documentation that support the findings of this study are openly available at the PS: Political Science & Politics Harvard Dataverse at https://doi.org/10.7910/DVN/YGVLRC.
Supplementary Materials
To view supplementary material for this article, please visit http://doi.org/10.1017/S1049096523000240.
ACKNOWLEDGMENT
We thank Steve Fisher for helpful comments on an early draft of the article.
CONFLICTS OF INTEREST
The authors declare that there are no ethical issues or conflicts of interest in this research.
Open access
