'Migrants and Nationals' Research Project: Summary of Findings in England

download as PDF

Susan Condor, Lancaster University

Overview of findings from the England strand of the project: 'Migrants and Nationals', funded within the Leverhulme Trust Constitutional Change and Identity programme (Grant number: 35113).

© Susan Condor, 2005

1. General Background to the research

1.1 This was a 5-year longitudinal study seeking to document and to map patterns of national identity in England in the immediate aftermath of UK constitutional change. Data were collected between 2000 and 2004

1.2 The original research proposal (see Appendix 2) drew attention to three key problematics relevant to the research conducted in England:

(a) The general question of the ways in which people construct national identity, and the possibility of this changing over time.

(b) The consequences for people born in Scotland now living in England. (c) The responses of members of ethnic minority groups.

1.3 In addition to addressing the specific questions detailed in the original research proposal, we also considered the following issues relevant to the England context:

(a) How people living in England understand the process of constitutional change.

(b) The possibility that members of minority national and ethnic groups in England (such as Scots, or people of Pakistani ethnic heritage) may draw upon a variety of cultural and rhetorical resources when constructing notions of identity or community, and consequently may not always be simply reacting to 'majority' viewpoints.

(c) People's orientations to other potential or actual changes to the UK constitution, in particular proposed Regional Assemblies and EU integration.

1.4 The general objectives in conducting the interviews were:

(a) To collect data which would allow meaningful comparison between respondent groups, and which would allow comparison of an individual respondent's perspective over time.

(b) To collect data in such a manner as to allow for an appreciation of the nature and extent of flexibility within an individual respondents' repertoire (what social psychologists sometimes term the 'latitude of attitude').

(c) To avoid biases in respondent recruitment and problems of 'reactivity' by ensuring that we did not prime respondents as to our specific research objectives, and to avoid contaminating the panel by ensuring respondents remained as far as possible unaware of our particular interests.

(d) To collect, as far as possible, vernacular accounts which would map reasonably well onto everyday discourse.

(e) To allow interviews to be customised in such a manner as to avoid imposing forms of vocabulary upon respondents, to allow for the possibility that different respondent groups may orient to different concerns, and to allow for differences in political understanding, interest and verbal fluency among research respondents.

(f) To collect data which would afford analysis of people's 'in principle' self-reports (i.e. what they say they do or say in theory) and their 'in practice' behaviour (i.e. what it is they actually do or say). In addition, we attempted to collect data that would allow us to gain insight into attitudes and identities operating at different levels of discursive consciousness (what social psychologists often distinguish as public, private and implicit attitudes and identities).

1.5 As outlined in the research proposal, this study utilised a longitudinal panel design. Panel respondents were interviewed twice, with an 18 month-2 year period between interviews. The third round of the research focussed on a strategically selected sub-sample of respondents, who were interviewed as couples or in small groups.

1.6 In addition to the panel interviews, a range of supplementary material was used to ensure the validity and reliability of emergent findings:

(a) To check for re-interviewing effects, new respondents were recruited and participated in round 1-style interviews throughout the duration of the project.

(b) Two satellite studies were conducted in Greater Manchester to cover issues difficult accommodate within the panel design: a study of 45 people of Pakistani ethnic background, and a study of 15 active members and associates of the BNP.

(c) In order to check for possible generational (as opposed to individual) changes in identity and attitude, we conducted focus group studies on young people from our two research sites over the 5-year research period.

(d) In order to assess change over time, the findings from the panel interviews were compared with findings from a qualitative interview study on national identity in England conducted prior to the change in the UK constitution (1992-7).

(e) In order to establish generalisability of conclusions, and to establish the extent to which our panel study was providing us with a saturated sample of responses, we compared our findings against those of some other qualitative interview studies conducted in England at the same time.

(f) The kinds of response patterns identified in the qualitative interview data were routinely checked against survey data.

2. The England-born panel ('English Nationals')

In this section we summarise some key aspects of the research strategy and findings relating to the English nationals panel (that is, people born and currently resident in England).

2.1 Locating the research problem

Questions related to English national identity have typically been raised in the context of various forms of moral panic. At the time that we started this research, there were four common concerns relating to national identity in England that informed political, media and academic debate, which had, in part, informed the original specification of the research problem in our research proposal.

(1) That a sense of national identity amongst the indigenous English had hitherto been lying dormant but was about to be 'wakened'.

(2) That when this happened, a sense of English identity would displace British identity

(3) 'British identity' is the psychological condition on which the legitimacy of the British state rests. Ergo, as a consequence of a decline in British and a rise in English identity, the population of England will start to question the legitimacy of the British state, and will develop a nationally specific political voice. This could take one of two forms, both of which would threaten to unsettle the devolution settlement. First, 'the English' could claim rights to national self-governance. Second, 'the English' could express resentment concerning the devolution settlement in Wales and Scotland.

(4) Assertions of English identity have been conventionally associated with the discourse of right wing political groups. In addition, members of ethnically and racially defined 'minority' groups often use the term 'English' to designate the white majority. Consequently, any rise in 'English identity' may bring in its wake a rise in racially and ethnically exclusive understandings of belonging and political community.

None of these presumptions had any established empirical foundation. Consequently, we approached all of these concerns as empirical questions, rather than a priori frames of reference.

2.2 Study Details

Research Sites The research proposal specified that two key sites would be used, one in the North and one in the South of England. The two research sites eventually selected were as follows:

(1) Greater Manchester. This is an urban conurbation in North West of England, with a relatively large (pop 2.5 million) and ethnically diverse population.

(2) The Rother region of East Sussex. A number of small towns and villages on, or close to the South Coast of England, selected partly on the basis of their not being within commutable distance of London or any other major city. The area is mainly rural (mixed farming and fishing) and since the loss of light industry during the 1980s, relies heavily on tourism, the local area being marketed by the English tourist board as in terms of its quintessentially English character ('1066 country').

Panel Samples Within each site, i respondents were selected through a combination of open and principled sampling with a view to three considerations:

(1) In line with standard recommendations concerning sample selection for the purposes of qualitative research, panel members were recruited with a view of maximizing the likelihood of achieving a saturated sample of accounting practices by ensuring the diversity of respondents, and, in particular, recruiting respondents in such a manner as to actively seek negative instances;

(2) In order to afford analytic leverage on questions concerning national identity etc., an attempt was made to recruit respondents who might, on a priori grounds be anticipated to hold particular sorts of views (e.g. political affiliation, social class) and on the basis of other factors identified during the course of ongoing analysis (e.g. people with more 'local' or 'cosmopolitan' personal networks; people who were more or less engaged in civil society or formal political activity).

We endeavoured to ensure that any factor of potential analytic interest was represented by a range of cases, and that there were no obvious confounds.

Round 1 (2000-2001). A total of 108 respondents took part, of whom 85 were subsequently selected to constitute our eventual working panel. is represented by a range of cases, and that there are no confounds between factors.

Round 2 (2002-3). A total of 90 respondents from the first round were reinterviewed.

Round 3 (2004). Sixteen strategically selected pairs or groups of respondents (40 respondents in total) took part in the third round interviews.

Satellite studies

Although most of the research objectives laid out in our original proposal could be adequately addressed through panel methodology there remained some residual questions which are not easily addressed within the general scope of the panel studies, and for which we consequently designed two satellite studies that ran alongside the main programme of research

People of Pakistani background living in Greater Manchester

Part of our research remit concerned the possible effects of changes to the UK constitutional on the attitudes towards, and experiences of, members of nationally, ethnically and 'racially' defined 'minority' groups. Although attitudes towards minority groups may be monitored through the English national panel, and the experiences of Scotland-born individuals monitored through the Scots Migrant panel, there is relatively little scope within the panel study for a meaningful analysis of the experiences of members of other minority groups. There were three reasons why the research questions concerning ethnic minority experiences in England cannot be simply accommodated within to English nationals panel.

The criteria that members of the English nationals panel be born in England made it hard to address the specific question of how people may attempt to 'negotiate entry into new national communities'. The birth criteria of panel membership also meant that there would be a disparity between the status of those respondents designated Scottish residents in England (who would, by definition, all be immigrants) and the status of any other national or cultural minority respondents (who would all, by definition, be native-born).

Were members of ethnically or racially defined minorities to be studied simply through their inclusion in the English nationals panel, this would of necessity result in tokenism. With a target panel size of 90, it would not be possible adequately to represent any, far less a reasonable proportion, of the various ethnic minority peoples currently living in England. Again, there a problem of lack of symmetry presents itself. In constructing the panels, extreme care was taken to ensure the heterogeneity of the English nationals and (as far as possible) of the Scottish Migrants panels. Given that the experiences of Scots in England were to be monitored through a panel of 60 members, balanced in terms of gender, age profile and length of residence, and varying in terms of educational qualifications, socio-economic status, political beliefs and so forth, it would have been inappropriate to attempt to reach conclusions about the attitudes or experiences of members of ethnic groups [sic] based on one or two individuals.

Social scientific research has generally found recruitment of members of at least some ethnic minority communities to be very difficult. In view of the fact that the researchers in England are all white, and given the general nature of the research topic, we might reasonably expect the recruitment of an adequate sample of ethnic minority respondents to be more time-consuming than the recruitment of the English national respondents, and, moreover, it would be reasonable to suspect that the task of panel-maintenance would be more difficult.

For these reasons, it was decided that the research questions pertaining to ethnic minority experiences and attitudes would be best dealt with using a satellite study, running in parallel to the main panel study. Since, by design, only one of our research sites (Greater Manchester) had a sizeable ethnic minority population, the satellite study was only conducted there. Rather than attempt a generic study, we chose to focus on those people whom, according to the latest census, constituted the largest ethnic minority group in Greater Manchester (Pakistani-origin Muslims), and to limit the scope of the study still further, we confined our study to people aged between 17-45. Within this category of target respondents, an attempt was made to recruit as wide a range of people as possible. However, the time taken to contact and recruit respondents meant that a panel methodology would not be feasible. Consequently, we conducted this part of the research by interviewing respondents either individually or in focus groups throughout the period 2001 Ð2004. In the event, the period of our research coincided with a series of moral panics concerning the young Pakistani 'community' in Greater Manchester (promoted not only by responses to 'September 11', but also by the highly-publicised 'race riots' in 2001).

Organised racism

One aspect of our research problematic related specifically to the possibility that new, ethnically-defined, constructions of national belonging may arise in England in the aftermath of constitutional reform. The panel study of English nationals provided a good vehicle by which to document and monitor patterns of racism in general, and ethnic nationalism in particular. However, social scientific work on racist discourse has often documented how arguments justifying the exclusion of, or discriminatory action against, particular need not be based on appeals to blood or to the inherent inferiority of the outgroup. Second, that research which has looked in detail at the strategies that Far Right ('racist') political organisations do use has shown these to be highly flexible and responsive to changing political circumstances. On the basis of what we knew of the dynamics and strategies of racist and of anti-racist social movements, it would have been be na•ve to assume that any response to constitutional change would occur in the form of a one-off transformation in social attitudes. Rather, it is more likely that issues relating to the proper basis of national self-definition will themselves become the bones of ongoing contestation.

For these reasons we considered that our general research problematic might also be addressed by reversing the specific questions as laid out in our original research proposal. Rather than simply asking if ethnic versions of nationhood are arising and gaining credibility in England, we could also pose two alternative questions:

1. To what extent are arguments for the social and/or civic inclusion or exclusion of particular groups justified with reference specifically to national belonging in general, and blood criteria in particular?

2. How, precisely, does explicitly exclusionary rhetoric work? When people do explicitly challenge the entitlements and citizenship status of particular ethnically-defined groups, or to restrict immigration on the part of ethnically, nationally or racially defined outgroups, how precisely do they justify these arguments?

Although the main panel study provided a unique opportunity to monitor forms of everyday racism and anti-racism over a period of time, the general problematic as laid out in our original proposal would also require that we include within our research remit some study of people who are already known to use national categories as a basis for determining social inclusion and exclusion. For this reason, we have planned a study monitoring changing forms of racist discourse within the British National Party to run alongside our main panel study. Specifically, this involved a core mini-panel of 15 BNP members or associates.

2.3 Summary of general conclusions

The general finding from our research on the identities and political attitudes of people born and resident in England is that none of the fears popularly expressed at the time of the devolution settlement have proved to be well founded. Although our research has only spanned a relatively short period of time, we do not see any reason to believe that things are liable to change dramatically over the next few years.

Our general conclusions from the main panel study, combined with supplementary data, are that moral panics that were apparent at the time we started our research were based on an insufficiently nuanced understanding of the way in which notions of national identity operate in England. In addition, they were often based on a simple inference from the situation as it existed in Scotland, without taking account of the way in which ostensibly similar forms of discourse may have substantively different functions within different cultural and political contexts.

Specifically, our findings lead us to question the way in which previous theory and research has assumed the following:

á That the population of England recognise the Scotland-England 'union' as the keystone of the British constitution, and that consequently any changes to this arrangement will impact heavily on their sense of British identity.

á That identity is a relatively simple matter, such that it possible to infer the salience and strength of subjective identity from public acts of self-labelling, and such that particular political motives can be read off national identity claims in a relatively straightforward manner.

á That the categories English and British have a singular, fixed meaning, which remains constant independent of context of use.

á That national identity represents a form of independent variable, such that any changes to national self-definition would be likely to have direct and predictable knock-on consequences for an individual's political attitudes and social action.

In contrast to these assumptions, it is clear from our research that understandings of nationhood in England - whilst still under-stated relative to the situation in Scotland - are highly complex and nuanced, and consequently capable of considerably more flexibility and adaptability than had often been previously supposed.

Evidence of complexity should not, however, be treated as necessarily indicative of confusion. It is as well known in psychology and in linguistics that the most useful (self) categories are often those that afford the most flexible customisation to accommodate various situational contingencies, and that hence preclude simplistic, rigid, categorical judgement.

Complexity (and the variability which accompanies this) should also not be confused with incomprehensibility. This is not to say that people in England necessarily have full, reflexive, discursive consciousness of the ways in which they use national categories, such that respondents' accounts can be taken as direct testimony. Analysis of the details of accounting practices, however, reveals fairly robust patterns in the way in which people 'do' national identity as a matter of practical, if not discursive, consciousness.

This capacity for flexibility and variability renders it difficult to acquire a nuanced understanding of national identity in England through the use of methods that do not afford an appreciation of the role of local context in the meaning and interpretation of national identity claims (as is often the case in survey research, for example).

However, an appreciation of the flexibility of national accounting according to local context need not come at the expense of either an appreciation of reliable individual differences in attitude and identity, or of general cultural patterns. In this respect, it is important to note that despite evidence of variability in the meaning and use of national categories within our interviews (the extent of which was strategically maximised by sampling and interviewing procedures) it was nevertheless possible to identify general, underlying, patterns in response which account for virtually all of the accounts collected from white people born and resident in England.

These patterns become particularly apparent when contrasted with other populations of respondent: the Scottish nationals, (many of) the Scots Migrants, people from Pakistani ethnic backgrounds and people with far right political beliefs, all of whose orientations to national identity differed qualitatively in certain respects from those of the normal pattern of responding in the English nationals sample.

2.4 Summary of key findings

2.4.1 Generalisability of reported findings

For reasons of simplicity of exposition the following summary of substantive findings has been expressed in relatively categorical terms, at the expense of some measure of nuance. However, all of the specific findings highlighted here have been selected on the basis that they tolerate generalisation reasonably well. In all cases, the findings reported pertain equally to respondents of all ages, from both research sites, of both genders and most political leanings. In no instance were there many deviant cases, and in almost all instances the deviant cases identified all fell into one or more of the following categories:

1. People with some direct experience of living in another country in the British Isles, or with close family members from another British Isles country. These people often presented accounts of nationhood and national identity which much more closely paralleled findings from the Scottish nationals panel in the present study.

2. People with experience of serving (non-conscripted) in the armed services. This seems to have had two consequences. First, these respondents have available to them a technical vocabulary of British statehood (and are consequently, for example, more aware of the constitutional status of the monarchy). Second (from the respondents' own self-reports) it seems that they have often had experience of working and living in close proximity with people from Scotland who encouraged them to describe themselves primarily as English.

3. People with far right political attitudes.

2.4.2 Evidence of change over time

A comparison of data collected in 2000-4 with analogous interview data collected a decade earlier in England generally indicated little in the way of substantive change over time.

The major difference was that respondents interviewed post-devolution were generally more willing to call themselves English. In addition, respondents were, after 2000, increasingly able to engage with questions concerning national identity choice (i.e. whether one prefers to call oneself English or British), although few people (mainly those with right wing politics) did so entirely seriously, and very few people regarded this as a particularly important issue. The categories British and English were still being used, after 2000, to refer to the same constellations of construct as they had in 1992.

Otherwise the major difference between accounts of nationhood and national identity collected 2000-4 and those collected in the early-to-mid 1990s appeared to have little to do with contemporary constitutional politics. The overwhelming difference involved a shift away from an orientation to nationhood that was heavily coloured by discourses of decline and associated overwhelmingly with pessimistic imagery of impeding catastrophe and insecurity.

This is not to say that orientations to nationhood in England collected from 2000 onwards were positively optimistic (there was little evidence of a sense of national progress). Although respondents interviewed post-devolution were less inclined to be overtly critical of their country or display positive shame in their national identity (whether England/English or Britain/British) neither was there evidence of any renewed sense of national pride. Rather, respondents typically oriented to a normative injunction according to which one should accept one's national identity as a fact of personal biography and of social and political life.

In summary: In the early-mid 1990s, interview respondents in England generally adopted a 'things are getting worse all the time' stance. In our interviews conducted at the start of the twenty first century, respondents in England were generally adopting the position that 'things could be worse'.

2.5 English and British identity

2.5.1 Willingness to self-identify as English.

In common with other research we found a rise in the extent to which white people in England were willing to claim an identity as English compared to the pre-devolution context. The particular moment of change appears to have taken place at around the start of 2000. People in our pilot interviews (conducted in early 2000) generally preferred the label British to English, and justified this on much the same grounds as people in England tended to prior to devolution (e.g. that British was the more inclusive term). However, by the start of our round 1 interviews 'proper', people seemed more inclined to call themselves English (although not necessarily at the expense of British). Thereafter there was remarkable consistency over time in the extent to which particular respondents reported preferring to call themselves English or British (and their reasons for so doing).

2.5.2 Criteria for nationality attributions and national identity claims.

One aspect of our proposed research concerned the perceived criteria for national inclusion, specifically: 'how are factors such as blood, birth and ancestry weighted against factors like residence, choice and commitment?'.

In practice it was analytically quite difficult to pin down this issue, since the criteria that respondents invoked could vary according to context, and there was often a distinct difference between the criteria that respondents specified mindfully (e.g. when answering a question on the subject) and those that they assumed implicitly.

It was clear that the white respondents oriented to a perceived taboo against racialised criteria of nationality, and tended to regard birth warrants as comparatively normatively acceptable. Comparative analysis did not suggest that this had changed since the early-mid 1990s, nor that there was any evidence of change during the lifetime of this particular project.

The one exception to this normative preference for birthplace criteria on the part of white respondents could be found in instances in which respondents were orienting to the possibility that their claim to English identity might be regarded as boring. In these cases, respondents could refer to their lack of foreign ancestry to explain why they did not have any choice but to call themselves English. Alternatively, respondents could mobilise (exotic) foreign ancestry as a means by which to differentiate themselves from, and elevate themselves above, their ordinary English compatriots (Irish ancestors appeared particularly in vogue).

It should be noted that the question of national identity markers (i.e. whether nationality is a matter of birth, blood or belonging) does not generally constitute a matter of explicit public debate in England, and interview respondents often treated it as an odd (and abstract) question when raised as an explicit conversational topic. Moreover, although respondents were generally inclined to regard accounts that

attribute nationality to blood as indicative of racism, people with far right political beliefs (including members of the BNP) were in fact inclined to use more rather subtle (cultural racist) arguments.

2.5.3 Interest in national identity

One factor which remained relatively constant since the mid 1990s, and which did not appear to change over the duration of our current project, was that respondents in England still did not appear particularly inclined to view national identity as an interesting or noteworthy topic of conversation in its own right.

This necessarily posed problems for our research process, and we went to some length to ensure that we did not bias our sample by flagging the issue of national identity at the point of recruitment, and that we used strategies of interviewing which would allow us access to people's banal assumptions about nationhood even if they were not inclined to engage in lengthy explicit debate about this issue.

In our panel study, most of the English national respondents treated the topic of national identity in general, and their own sense of national identity in particular, as of little interest-value. One consequence was that respondents were disproportionately inclined to pick up on the non-national aspects of any conversational topic offered. Consequently, we had long conversations in which elderly people spoke of their experiences of 'the war' in considerable detail, whist never making use of explicit national categories. People also often managed to dodge the national angle of such apparently promising topics as the Jubilee or the football World Cup.

Occasionally, during the course of the five years over which the interviews were conducted, a political or media event would occur which respondents would use as a conceptual peg when answering our questions concerning national identity. For example, respondents could refer to debates over the categories used in the census, the apparent rise in national identity during the World Cup tournament, and occasional media reports of survey evidence on national identity and so forth as something that had made them think the question of national identity 'might be interesting'. However, these moments of interest were generally short-lived, and tended to be soon forgotten.

Respondents with post-compulsory education, and in particular those with liberal political views, tended to treat a concern with issues relating to national identity as a parochial and petty distraction from the 'big issues'. Less educated people, those with less interest in 'politics', and those who lived in the area in which they had been born, were inclined to treat national identity as overly abstract and academic topic, with little resonance for their own everyday lives and concerns.

Devolution Ð in so far as it was understood within a domestic political context Ð was similarly regarded as a relatively uninteresting political event (by those who were politically active, of liberal political persuasion and/or relatively highly educated) and as of negligible relevance to the concrete realities of their own everyday lives on the part of the great majority of respondents.

There were two contexts in which issues relating to national identity were generally deemed to be of potential personal or social interest to the white respondents: (a) When this was cast as a matter of immigration, multiculturalism and racism (b) When this was cast within the context of EU membership. Significantly, devolution as a topic was also commonly oriented to as personally or socially significance in so far as it was cast within one of these two problematics.

People from Pakistani ethnic backgrounds living in Greater Manchester, unsurprisingly, did generally treat the general topic of national identity as an interesting and important matter. However, in practice, in this context national identity was normally understood as a proxy for matters relating to ethnicity or citizenship, or was understood to be important in so far as it was imbricated with connotations of generation, westernisation and social class.

2.5.4 Devolution as a warrant for English identity claims.

In general, respondents did not treat devolution as an especially salient context for talk about national identity, even when it was currently figuring in 'the news'. At the end of our round 3 interviews (after respondents had been questioned closely on three separate occasions about their views on the Scottish parliament) we asked them to discuss the 'crisis in British identity'. Some respondents suggested that they had never thought about the issue (despite having discussed it with us twice previously). Hardly any respondents interpreted this as a question about UK constitutional change. Approximately half interpreted this as a question about immigration and multi-culturalism, and a half interpreted this as a question about the EU

When asked specific questions concerning their personal preference for national self-labelling, respondents often spontaneously suggested devolution now made it permissible to claim an English identity (the implication being that it had not been in the past).

In contrast, the BNP supporters were quite distinctive in so far as they associated claims to English identity to claims to political independence for England. These respondents typically expressed positive attitudes towards the Scottish parliament, and the prospect of Scottish independence, and often spontaneously cited the fact of devolution as justification for their own assertion of English nationalism.

2.5.5 Changing images of Englishness?

Social psychological research often suggests that social identification will be accompanied by a tendency to self-attribute ingroup characteristics and to use these as a normative guide to action. However, although people in England were more inclined to call themselves English than they had been previously, there was no evidence of them being inclined to hold any more specific images of English character or culture. This was as true of the BNP supporters questioned as it was for the members of the English Nationals panel. Although the BNP supporters argued strongly for English political independence, and for English identity as a basis for social inclusion, none of these respondents was able to provide a substantive account of what English identity or culture actually was.

People of Pakistani ethnic heritage commonly use the category English to refer to white majority culture, and in this context could invest the category with a particular substantive content (Christianity, individualism, particular cultural practices etc.). However, a tendency to treat English as a reference to the cultural other was not monolithic (for example, it was also possible to cast English as a territorial referent) and did not always imply a complete sense of personal alienation from the category: for example, it was common for people to talk about occasions on which they 'acted English'.

The absence of any specific, or contemporary, images of Englishness on the part of the white panel respondents did not appear simply to reflect a lack of available cultural resources. On the contrary, respondents were typically actively resistant to substantive notions of what it is to be English, and demonstrated an investment in keeping the category under-specified. They typically reacted negatively to questioning on this subject, by pointing to the existence of regional, class and ethnic diversity, and to an historic loss of indigenous English culture (either early in history, or later through the influence of Americanisation).

2.5.6 The scope and limits of claims to English identity.

It was notable that although respondents relatively frequently cited devolution as a warrant for calling themselves English, very few respondents reported that devolution had influenced their identity in the sense of making them subjectively more aware of, or concerned about, being English. That is, while their use of the label may have increased, the strength, salience and personal commitment to the identity had not.

When the English national panel respondents claimed an English national identity, this was almost always accompanied by an attempt to specify Ð and usually severely to delimit Ð the personal significance and action-relevance of the identity in question. In particular, respondents of all ages and political backgrounds were inclined to:

• Distinguish their claim to national identity from connotations of ethnocentrism or parochialism
• Indicate the limited personal significance of the identity in question
• Distinguish the act of claiming a national identity from the realm practical action.

In contrast, the BNP members treated claims to English identity as politically meaningful. They expressed an English identity together with displays of national pride, they treated their national identity as of central personal significance, and they treated English national identity as a legitimate basis for political action.

People of Pakistani background could orient to English identity in one of four ways.

• Treat English as an ethno-racial referent, and treat it as other-than-self. First generation respondents, and women in particular, were inclined to treat the Asian versus English distinction as a reference to distinctive, and relatively closed, social networks. Amongst the other respondents, however, this stance did not necessarily entail a sense of exclusion from civil society, however. Rather, it was often used a community cohesion repertoire, in which a mutual recognition of ethnic or racial difference (symbolised through labelling practices) was treated as a necessary precursor to civic engagement and mixing.

• Treat England solely as a territorial referent. In these contexts, respondents could present themselves as identifying more strongly with England (with its connotations of locality) than with the more abstract and generic Britain. In relation to talk about devolution, respondents could adopt implicitly an English national 'we' and argue in favour of the specific territorial interests of England.

• During the specific period when the research was conducted there developed a trend (which may or may not turn out to be generalisable) among young third-generation men to use claims and displays to English identity linked to national football team support as a means by which to differentiate themselves from their second generation parents who self-defined as British.

• There was some indication of people who were more politically active at the local level in areas where there was a visible BNP presence, were challenging the BNP's racist application of the category English to civil society, by self-consciously adopting the category as self-relevant.

2.5.7 English national identity as a basis for social inclusion?

In our original research proposal, we suggested that matters of social inclusion might be related to the ways in which people understand the 'range of those defined as fellow nationals. as people-like-us'. This formulation squared well with the discourse used by the BNP, for whom matters of civil society and citizenship status were commonly understood to be a matter of 'people like us', understood specifically in English national terms. However, these were the only respondents in our study who did evidence this form of understanding.

People of Pakistani background were inclined to adopt an understanding of society as ideally (if not always in practice) involving a multicultural 'community of communities'. On the other hand, they also often adopted a public/private distinction, whereby the domain of the public was English. In either case, social inclusion was seen to be a matter of mixing (and possibly in the latter formulation, 'fitting in') rather than common identity per se.

The white panel members' attempts to ring-fence English national identity claims from the domain of practical action (what we might term an 'identity versus action' perspective) squared with the ways in which these respondents discussed matters relating to social inclusion and civil society (which tended to involve categories of the 'local' and 'community').

When discussing matters relating to social inclusion and civic participation, the panel respondents typically oriented to a normative requirement to display in principle an appreciation Ð or at the very least toleration - of diversity. This was often (ironically) demonstrated through a tendency to differentiate self from those ('other') people who routinely made 'them' versus 'us' distinctions. For these respondents, the very notion of allocating civic rights and duties on the basis of common identity ('people like us') was anathema, and commonly understood to constitute a form of prejudice.

In practice, respondents routinely utilised a whole range of ingroup-outgroup distinctions when making moral or evaluative judgements, establishing norms of proper conduct, adjudicating concerning matters of justice and legitimacy, of needs and deserts in relation to informal social relations and civil society. In order to avoid the normative opprobrium associated with such acts of discrimination, these category distinctions could often be discursively-coded, or their expression restricted to back-stage social encounters. The major exceptions, in which people could explicitly invoke 'us' versus 'them' distinctions, involved situations in which (a) the respondent positioned themselves as a member of an oppressed or disadvantaged minority, (b) the 'others' were positioned as temporary visitors rather than permanent residents in the local area (e.g. 'holidaymakers, 'out-of-towners', 'asylum seekers'), or (relatedly) (c) when distinctions between ingroup and outgroup were represented in territorial terms ('there' versus 'here').

The kinds of ingroup-outgroup distinction used to determine norms of appropriate conduct or to justify particular forms of social activity, virtually never involved a simple English/non-English axis of differentiation. This is, of course, quite consistent with the observation that even respondents who were happy to self-categorize as English were disinclined to accept images of the English as a singular people, or an English identity as grounds upon which to establish common interests, ways of life, taste, moral codes, needs or entitlements.

Rather than being articulated around an English vs. non-English axis, the kinds of constructions of people like us that tended in practice to be used with respect to questions of social activity typically involved distinctions such as urban-rural, local-incomer, or generational groups. The major axes of practical differentiation (under which other forms of category distinction were often, in practice, subordinated) involved some form of social class or status distinction or some reference (usually heavily coded) to racially or ethnically defined categories.

2.5.8 Football and English national identity.

One sphere of activity that is conventionally strongly associated with English national identity is that of football support. Over the time-span of this project there were two major international football tournaments. The evident increase in popular support of, and insignia associated with, the England team support (notably the St George's cross) prompted widespread media and popular speculation that this represented evidence of the anticipated explosion of English nationalism.

In view of the symbolic importance of football in popular images of English identity, we had considered this issue potentially worth pursuing during the first round of interviews. As this topic emerged as a potentially significant issue in its own right, we supplemented the panel interview data with research conducted specifically around major games, most notably a series of interviews conducted on the street in our research locales before during and after the England-Argentina match in the World Cup. Analysis of this material demonstrated that talk about England national football support, far from constituting an exception, in fact conformed closely to the general rules of national identity in England:

• In the interview context, respondents normally mentioned football spontaneously in the context of discussions of national identity. Respondents normally treated England support as reprehensible or (if they were themselves an England supporter) as relatively trivial or unimportant. During the first round interviews, we also asked people about supporting other 'home' sides. Somewhat predictably, respondents tended to say that they would support Wales or Scotland as well as England. Respondents who could claim ancestral connection to other countries in the British Isles often said that they would support these sides in preference to England (once again, Irish ancestry figured strongly).

• Throughout the duration of the project, England support was treated as an accountable matter: that is, people oriented to a need to justify supporting the England team. In warranting their support, people typically sought to delimit its significance for practical activity: supporting England was 'just' a matter of accepting the fact of one's birthplace. England support was almost always distinguished clearly from the domain of the political, and from the domain of 'serious' football support (which was reserved for club support).

• References to English football support were frequently used by young people explicitly to delimit the significance of their claims to national identity, and to exclude matters of national identity from the frame of 'serious talk'. It became increasingly common over the five years that we were conducting the research for young people to claim to see themselves as English 'but just because of the football', and to suggest that they would 'grow out of it'.

• The attempt to render England football support effectively socially meaningless and distanced from the realm of practical activity (i.e. as not symbolic or symptomatic of more general nationalistic concerns) was mirrored in successful attempts to democratise the basis of England support during the time-span of our project. This had an interesting consequence. The commercialisation of England insignia, and the democratisation of the basis of support (to include, notably, women, children and Ð very visibly in the Greater Manchester ex-mill towns - young men of Pakistani ethnic background) was strongly resisted by the men we interviewed from Far Right groups. The reaction of our BNP respondents to these changes was to distance themselves from what they understood to be a commodification, trivialisation, and de-politicisation of English identity. By way of response, they made a point of not sporting any visible England insignia at England matches (instead using Ulster Unionist insignia) and indicated an interest in moving their energies away from the commercially and socially contaminated arena of football, to the domain of England rugby support.

2.5.9 Changing meanings of 'English' vis a vis 'British'?

It is currently common for survey research to attempt to read political attitudes off the categories that people use to describe themselves. Generally, it has been found with samples in England that preference for self-labelling relates very weakly, if at all, to political party support, attitudes towards the constitution etc.. The reason for this became apparent in the course of interview research, in which it became clear that both the terms English and British were capable of multiple interpretations, as geographical, political/administrative or cultural entities. This allowed a good degree of flexibility in national talk, since the respondents could understand and use these terms in different ways in different contexts.

Flexibility of referent adds somewhat to the complexity of interpreting identity claims, since what a person calls themselves (English, British, both or neither) is, by itself, relatively meaningless. For example, it was quite possible to customise an English identity claim to render it compatible with Ð or even cast it as indicative of Ð a whole gamut of political positions. Hence, a claim to English identity could be used as a marker of right wing politics, but could also be used by liberals as a means by which to demonstrate their support for subordinated regions of the UK, or to flag their resistance to the British state. Similarly, a claim to British identity could be cast within both liberal and conservative political perspectives. And, in both cases, a claim to identity could be dissociated altogether from the domain of the political.

In particular, it is worth noting the complex relationship between national identity claims and values relating to 'race' and racism. All of the white respondents who had far right political affiliations claimed an identity as English, and disclaimed a British identity. However, a wide variety of orientations to national identity could be justified as an antiracist stance. White respondents who were concerned to avoid potential charges of racism or xenophobia could variously: (a) Refuse to describe themselves in national terms at all; (b) Claim to prefer the label British on the grounds that it is more inclusive; (c) Claim to prefer the label English as a means by which to flag a recognition of the imperialist connotations of the category British; (d) Claim to prefer not to distinguish between the categories English and British.

Although respondents often suggested that there was little functional purpose in distinguishing between the categories of English and British with respect to many specific domains of action, it was rare for people not to distinguish the two constructs on some level. Only four respondents did not at any stage orient to the terms as pertaining to qualitatively different constructs. All of these respondents currently lived within three miles of where they had been born, had no formal educational qualifications, and were generally resistant to describing their lives or attitudes in abstract rather than concrete terms.

The question of the ways and extent to which people differentiate between the categories English and British can be approached in two ways. First, we can consider how the categories English and British may be mindlessly used: that is, how (and to what extent) do respondents in fact apply these categories to different topics or objects in the course of conversation?

Analysis indicated that even respondents who claimed in principle that they did not know the difference between the categories English and British in practice displayed a surprising degree of consistency in their use of the terms. The general patterns in the use of category referents (which hold true across the respondent group) were as follows:

The term England/English tended to be used in a rather restricted range of contexts:

• In the context of accounts of place and landscape
• In the context of accounts of leisure, football, and popular or high culture
• In relation to history, heritage and the past.

In contrast, the labels Britain/British were more generally applied in talk about political issues and matters relating to contemporary social life and to the future. The respondents from Pakistani backgrounds commonly formulated British identity as a matter of citizenship status, but it was very unusual for white respondents to do so. In addition, respondents from Pakistani backgrounds could associate claims to British identity with connotations of generation, and of cultural and economic capital, associations which were not made by the white respondents.

The situation in which the greatest degree of category confusion arose was in the context of talk about Europe, which could prime a number of different topics simultaneously (politics, football, culture, landscape, history etc.). Consequently, in the context of talk about Europe, respondents who elsewhere in their interview accounts had demonstrated a high degree of consistency in their category label use (and possibly even a high level of reflexive awareness concerning the rules concerning when one label or the other was appropriate) could be seen to slip: for example, talking about the 'English parliament', or 'the British football team'.

The second form of category differentiation that we can consider, involves respondents' in-principle accounts of their use of, and preference for, one category label over the other.

In this case, it was notable that even when respondents were explicitly discussing the different meanings of the terms English and British, these were rarely understood as competing, or even as logical alternatives. Respondents rarely spontaneously oriented to the question of national identity as a matter of choice of one label over the other, and some respondents demonstrated difficulty in understanding why one would need or want to choose once and for all, rather than adopt the label which would be most appropriate to any particular local context of action.

The relationship between categories British and English could be constructed in various ways, but the most common form of representation cast them in some way as nested (e.g. superordinate/subordinate) categories, as exemplified in the common phrase: 'England is a part, and Britain is the whole lot'. This partly explains why an apparent rise in English identity has not been accompanied by a decline of British identity in England.

Another reason why the predicted post-devolution decline in British identity has failed so far to occur is that people in England do not typically understand the term British as pertaining solely to the UK. British is also widely used to refer to a territorial entity (the 'island'), or to a people and (possibly) a culture. This ability to strategically dissociate the category British from the UK is illustrated by two relatively common rhetorical devices: the tracing of the category Britain to Roman origins, and the trope (used in moments of extreme cod-history pedantry) that as an ethnic category, 'British' refers 'only to the Welsh'.

2.5.10 Changing constructions of the UK?

Even when British is understood as a reference to the UK, the particular way in which the British state is understood in England militates against a Scottish parliament being seen to fundamentally undermine the construct.

In England, the UK is not typically understood specifically as 'the' (Scotland-England) union. Rather, it is understood as a union of three (morally equivalent) parts Ð Scotland, Wales and England, plus Northern Ireland (which is not generally treated as having the same status, typically being regarded as being effectively misplaced in the UK and 'really' being part of Ireland). This being the case, people see little reason why a Scottish parliament should be viewed as especially upsetting to established constitutional arrangements. Most respondents did not even view the prospect of Scottish independence as a necessary threat to the existence of a composite British state.

Although people generally regarded the three 'real' component parts of the UK as of equal moral worth, they recognized England's status as dominant member of the set. Sometimes people regarded Scotland and Wales as functionally equivalent both in terms of their current power and status, and in terms of the legitimacy of their claims to self-governance. When respondents did differentiate between Scotland and Wales, there was no consensus concerning the nature of the difference: in some cases, respondents regarded Scotland as the more central to the UK, or having the greater claims for national status, or to have experienced the greater degree of oppression. In other cases, it was Wales that was deemed the more important, authentic or deserving nation. The symbolic role of Northern Ireland was generally to act as a reminder of 'trouble', serving as a warning either of what happens when the English throw their political weight around, or as a warning against the destructive potential of nationalist politics.

2.5.11 Intra-UK national prejudice

Respondents were generally more inclined spontaneously to mention the existence of anti-English antipathy in Wales than in Scotland. Interestingly, accounts of anti-English antipathy in Wales generally glossed this as irrational and unreasonable. However, accounts of anti-English feeling in Scotland often cast this as reasonable and as a justifiable response to the historic wrongs the country had suffered.

There were different patterns of evaluative orientation expressed towards other UK nations. There was a general rule by which any suggestion that England was 'better' than Wales or Scotland was treated as highly normative accountable. If respondents wished to express this type of attitude, they had to present it as a joke or to provide extensive justification to establish its rational basis. Criticism of Scotland was also treated as highly accountable. It was very unusual indeed that a respondent attempted to say anything explicitly negative about the Scots or Scotland (except with respect to the climate). On the few occasions respondents did so, they warranted their right to voice such criticisms (e.g. by flagging their own ability to claim Scottish identity) or heavily justified or mitigated these accounts. In contrast, people's attitudes towards Wales or the Welsh varied much more (from the highly favourable to the very unfavourable), and explicit criticism or belittling of Wales or Welsh people was not treated as nearly so problematic.

By way of contrast, we may note the existence of two national groups with respect to which criticism was treated as normatively expected (to the extent that any positive comments required justification): Northern Ireland (although not necessarily the Northern Irish) and 'Americans'.

2.6 Attitudes to Constitutional Change

2.6.1 Attitudes toward the Scottish Parliament

At the time of the first round of interviews (2000-2001), most respondents demonstrated some prior awareness of the establishment of the Scottish Parliament and/or of the National Assembly for Wales. However, very few people indicated that this was an issue about which they held any particularly strong views. The only respondents who treated this as an issue in which they had any personal or emotional investment were those who had been recruited specifically on these grounds (e.g. people with some connection to English RDAs) and those individuals who understood devolution primarily within an EU (rather than domestic political) context.

Of those respondents who did express an evaluative stance concerning the Scottish Parliament and National Assembly for Wales, the majority expressed positive attitudes. Generally respondents did not distinguish between the constitutional positions of, or justifications for self-governance on the part of, Wales and Scotland (most people said that Wales should have a Parliament too 'if they wanted it').

Individual respondents' attitudes to the Scottish Parliament and National Assembly for Wales remained quite remarkably consistent over the three rounds of interviews. Only one respondent switched sides, in the sense of moving from an originally 'anti' to adopt a 'pro' position. In this instance, the respondent had originally represented a deviant case, since his original 'anti' stance was out of alignment with that of other respondents with similar (conservative) political views.

In general, although white respondents expressed positive attitudes towards the Scottish Parliament and Welsh Assembly (i.e. describing them as a 'good thing'), these attitudes were seldom backed up by any substantive opinions. In fact, detailed analysis indicated that these 'pro' attitude statements were typically presented as part of an argument that was entirely devoid of references to any specific information that could not be deduced from the mere fact that these bodies existed together with (in most cases) some reasonably accurate account of when they were established, and that there had been a referendum.

In the absence of specific information, 'pro' attitudes were generally warranted with reference to generic 'facts', including the 'fact' of distance (the Scottish Parliament being warranted on the basis that it was required since 'Scotland is a long way away'); or the 'fact' of cultural 'difference' (the Scottish Parliament and National Assembly for Wales being warranted on the grounds of 'their' being 'different'). This latter is particularly interesting, since it exemplifies the way in which liberal discourses of recognition can easily slip into reified representations of (purportedly self-defining) outgroups. Very often, the same respondents who argued that the Scottish parliament was 'a good thing', because Scotland is and Scottish people are different, had, earlier, when asked in principle, denied that any important differences existed between England and Scotland. Moreover, when asked whether an English parliament would be a good idea, these same respondents usually denied this with reference to the ethnic and cultural heterogeneity of England (thus effectively, and inadvertently, othering the populations of Scotland and Wales by representing them as essentialised, culturally homogenous, communities in contrast to our heterogeneity).

Underlying a good deal of talk about devolved governance was a set of generic values concerning rights. The first (typically politically conservative) version saw devolution (and even independence if desired) as justifiable in terms of a discourse of universal rights to national self-determination, and regarded devolved governance as part of a progressive historical move away from English colonialism and imperialism. This same set of values could on occasions be invoked to legitimate the idea of an English Parliament.

The second (typically liberal/left) version justified devolved governance to Scotland and Wales in terms of a discourse of 'minority rights' to self-determination. This line of argument (which was used by almost three quarters of the panel) tended to justify the existence of the Scottish Parliament and National Assembly for Wales in procedural rather than substantive terms. That is, respondents argued that it is right that the populations of Wales and Scotland be allowed to 'have what they want', irrespective of whether the outcome is practically beneficial for the countries themselves, and irrespective of any possible consequences for England.

The few respondents who did display any form of opposition to the Scottish Parliament and National Assembly for Wales tended to have a relatively high level of cultural capital, which presumably afforded them a potential claim to entitlement to express negative opinions (they included an MP, a respondent with a PhD in Sociology, two respondents with Oxbridge degrees in history, and three respondents who were in a position to claim (if they wished) an identity as Scottish or Welsh).

Interestingly, in no case did any white respondent, in any round of the panel interviews, oppose the Scottish Parliament and Welsh Assemblies in the way that the moral panic discourses suggested would be the case. That is, they never voiced this opposition as a defence of English political interests, or as a defence of British national identity or as a concern to preserve the British state. Rather, opposition to the existence of the Scottish Parliament and National Assembly for Wales was expressed in the following terms (interestingly, all of these explanations were used by people spanning a range of political affiliations, from the hard left to the right wing of the Conservative party):

• That the fundamental division within the UK is in fact one of class rather than of nation, and that the devolution settlement perpetuates class-based inequality of political representation and opportunity in so far as it legitimates a form of nationalistic false consciousness.
• In terms of minority needs rather than rights: that is, that the consequences of devolved governance are not in fact in the real (economic etc) interests of Scotland or of Wales. Occasionally, respondents suggested that the populations of Wales and Scotland had been led to misperceive their own interests (e.g. by the Labour party, possibly in cahoots with the EU).
• In terms of a generic opposition to all forms of political nationalism (i.e. attempts to align states with nations). In this respect, it is interesting to note that people in England often found it hard to square the principles of devolved governance within the UK with the process of EU integration. Apart from those few people whose own experience of working with RDAs had led them to understand UK devolution as part of a wider process of EU federalisation, most people who supported the EU had at least mixed feelings about devolution, which was seen to represent a backward step to smaller, national-specific units of governance. In this case, the moral exemplar typically used was, predictably, that of Balkanisation.

2.6.2 English political voice?

The dominant minority rights perspective tended to be accompanied by a tendency on the part of the respondent to disclaim rights to political voice on the grounds of their membership of a majority (English) identity group. That is, in this context, national identity entailed a moral obligation to political inaction. In this respect, it was notable that respondents' reluctance to engage actively with the issue of devolution could not be attributed simply to political apathy or lack of knowledge (although both of these were also likely to represent important factors). Respondents who were generally politically knowledgeable and active could often be seen to be orienting to a moral requirement to 'take a back seat' vis a vis the issue of devolution, and to display a recognition of the fact that the existence, or possible consequences, of the Welsh Assembly and the Scottish Parliament were in fact, 'none of my business'. Respondents with soft liberal/left political affiliations often adopted the view that the had no right to oppose the devolution settlement, and that any form of opposition would risk causing trouble (typically invoking the moral exemplum of Northern Ireland).

2.6.3 Resentment?

Our interview data revealed little evidence of resentment directed towards the Scottish Parliament and no evidence of any resentment towards the National Assembly for Wales (possibly because respondents almost entirely lacked any form of substantive knowledge about this) or, indeed, the London Assembly (probably because respondents were generally unaware of its existence). The one issue which respondents were inclined to treat as potentially unfair pertained to the West Lothian Question (that is, MPs from constituencies in Scotland voting at Westminster on matters which had been devolved). However, even respondents who regarded the current system as seriously unfair did not treat this as a criticism of the basic ideal Ð or in general of the procedure Ð of devolved governance.

It is, however, worth noting that respondents' acceptance of the fairness of the current system often involved reference to (a) the fact that the Scottish Parliament, and Scotland-specific policy decisions, were currently being funded 'by the Scots', and (b) the fact that the Scots suffered from some kind of minority status, and that the Scottish parliament was essentially rectifying a democratic deficit.

It was notable that there was a clear inverse relationship between the amount of specific factual information that a respondent displayed concerning the devolution settlement and their attitude towards the Scottish Parliament. Those five panel respondents who could be classed as the most well informed expressed the most negative attitudes of all participants. However, it is not clear what the direction of this relationship may have been. It is quite possible that people who are inclined to hold negative attitudes will be particularly motivated to seek out information on the topic.

2.6.4 Attitudes to proposed English Regional Assemblies.

At the time of our first round interviews, the only panel respondents who indicated any awareness of proposals for regional assemblies in England were those who were active members of political parties, or who were working with RDAs. Over the course of five years, very little changed. Some more respondents indicated some awareness of the issue having been mentioned, but had no substantive information. In 2003, as a prompt to interview discussions, we showed respondents a map of the proposed regional boundaries. This prompted a great deal of discussion. Respondents typically expressed opposition on the grounds that the proposed regions did not encapsulate meaningful communities of either identity or of interest.

• Typically, the proposed regions were contrasted with real communities of identity, exemplified by Counties, towns of (in the case of Pakistani-origin respondents) religion.
• Typically, the proposed regions were contrasted with real communities of interest, which were typically constructed along class lines or according to an urban/rural axis.

Although respondents commonly oriented to a North/South divide this was generally understood to refer to a cultural distinction (generally understood in terms of friendliness). People living in the South East often complained about being disadvantaged by the stereotype of South Eastern English affluence. The major regional distinction invoked as a matter of political interest was between London and 'the rest'. However, the only respondents who demonstrated any awareness of the existence of the London Assembly were those who had themselves previously moved from London.

Respondents often expressed dissatisfaction with the government, and expressed the view that their local area, or 'people like them', were not being properly represented, or were being materially disadvantaged in some respect. However, it was rarely the case that this dissatisfaction was expressed in conjunction with some positive statement concerning how things might be improved. Rather, the dissatisfaction seemed to relate to government, and to bureaucracy, per se: 'they're all as bad as each other'. Consequently, whilst almost all of the respondents we interviewed displayed some measure of dissatisfaction, and often resentment, concerning the democratic process, only a few saw a possible solution in a change of government, EU membership, leaving the EU, English Regional Assemblies or an English Parliament. Generally respondents were, and remained, of the view that however unsatisfactory the current situation might be, any change would only risk making things worse.

2.7 Concluding Comments: English National Identity and Action

• In this report we have been considering some of the findings from our 'English Nationals' panel study as they pertain specifically to the objectives outlined in the 'Migrants and Nationals' research proposal. However, we can also see how the issues covered here relate to the broader concerns of the Programme as a whole concerning the relationship between changing forms of national identity and developing processes of social and political action.

• In England currently, claims to English identity are becoming more common and, it seems, more normatively acceptable. However, this does not seem to have been accompanied by any appreciable trend towards treating English national identity as a useful, or legitimate, basis for individual or collective action.

• The one area in which it has become increasingly acceptable to express English national identity through action is in relation to national football support. However, the domestication and widening of the base of England football support has been accompanied by a clear tendency to mark this activity as trivial, and to distinguish it from the serious business of everyday life.

• General matters of social engagement, belonging, entitlement, the policing of norms of proper conduct etc. are spoken about in connection with a wide range of us/them distinctions, the most notable being status, class, race, ethnicity, generation and local status. Even given the propensity for some axes of discrimination to be discursively coded, it does not appear, from our interview studies with a highly heterogeneous sample of people, that people in England are currently inclined even tacitly to mobilise English/non-English distinctions in this sort of context.

• Of course, there are limits to what can be legitimately concluded on the basis of interview data alone. It is not possible to use interview data to draw conclusions about the extent to which English national identity may be being symbolically invoked in the course of practical activity occurring in the course of people's mundane everyday lives. However, what these interview data do quite reliably indicate is that, in so far as this is happening, instances in which English national identity qua national identity are used to warrant, justify or explain practical activity are likely to be (at the moment at least) occurring against a fairly generalised, and well-established, normative grain.
  

3. The Scotland-born ('Migrant') panel

In this section we summarise some of our key findings from the 'Scottish Migrants' panel (that is, people born in Scotland currently resident in England).

3.1 General background

Our primary research questions with respect to the Scottish Migrant respondents, as taken from our initial research proposal (see Appendix 2), were as follows:

The overall theme of this study is the question of how individuals negotiate entry

and are received into national communities, with particular focus on:

the degree to which different criteria of national community membership are voiced and receive support;

the consequences for action of these constructions for English people in Scotland, Scots living in England [É];

the responses and strategies of those, such as ethnic groups [sic], who are defined as national 'outsiders' as they attempt to negotiate entry into new national communities.

As part of our general research design we considered the possible effects of interviewer nationality (using a combination of Northern English, Southern English, Northern Irish and Scottish interviewers).

3.2 Logistic issues

At the outset we were aware of two logistic difficulties.

First, the research process was complicated by the complete absence of any benchmark data on this subject that could be used for the purposes of planning the sample, or checking the reliability of any findings. Even the Scottish 'Nationals' panel does not really constitute an optimal or necessarily appropriate source of benchmark data, for two reasons. (i) In so far as we allow that historical change may occur, we cannot assume that the attitudes expressed by members of the Scottish Nationals panel reflect the sorts of understandings and vocabulary that would have been current at the time that our migrant respondents were actually living in Scotland; (ii) Apart from those (few) middle class professional respondents originating from the Glasgow area, the composition of the Scottish Nationals panel did not match our sample of Scottish Migrants, many of whom originated from other regions in Scotland, and about half of whom were (originally at least) from working class backgrounds (for example: most of our Glaswegian Migrants came from the large estates, whereas the working-class members of our Glasgow Nationals panel do not).

In the absence of any existing demographic data, we were at the outset in no position to even speculate on the characteristics of the population from which we were drawing (for example, we knew little about patterns of migration from Scotland to England, about length of stay, class or age patterns in cross-border mobility etc.). Our sole original source of reliable information was almost a decade old: the 1991 Census. This indicated that 743,856% (1.58%) of the population of England had been born in Scotland. Overall, Scotland-born people were the single largest migrant category in England (14.7% of non-England born residents), except in the case of the 50-64 age group (in which case the largest migrant category was people born in the Republic of Ireland). The 2001 census indicated a slight rise in the number of Scotland-born individuals currently resident in England over the following decade (794,57, constituting 1.62% of the population). The proportion of Scotland-born people in the populations our two selected research sites fairly closely mirrored the national statistics (Greater Manchester: 1.3%; Rother (East Sussex): 1.4%).

We attempted to compensate for the absence of any relevant sources of qualitative interview data by:

(a) Conducting a few supplementary interviews with people who had moved to England from Wales (n=6), Northern Ireland (n=6) and the Republic or Ireland (n=6) to get a feel for what (if anything) might be specific to the Scottish migrant experience.

(b) Using a technique sometimes advocated by linguists conducting ethnographic fieldwork to check the generalisability of interview accounts: to tacitly use the interview guide in everyday social encounters with people we met who had moved from Scotland. This experience convinced us that the kinds of self-narratives that we collected in the interviews mapped very well onto the ways in which people from Scotland normally talk about themselves and their lives, at least with strangers.

The second general logistic issue related to the study of change over time. The design of this particular study does not easily permit analysis of change over time in so far as it does not allow a disambiguation of the effects of various axes of temporality. Specifically, in so far as a respondent does change their account of themselves during the course of the research, we cannot know whether this is due to: getting older, growing up, having been in England longer or to historical change. Other than conduct a 50-year study that combines longitudinal with cohort data, it is difficult to see any way around these problems, other than to be alert to the possible influence of these various factors upon our data.

As with the Nationals panel, we attempted to check repeat-interviewing effects by conducting new round 1 style interviews with newly recruited respondents throughout the duration of the research. Paradoxically, the most significant findings from these interviews was to flag the possibility of a generation or cohort effect. In 2004 we interviewed four students newly arrived at Manchester University from Scotland. All of these respondents demonstrated far less concern with (and generally less awareness of) the Scottish parliament than had similar respondents originally interviewed in 2000. In addition to the possibilities of sample bias, there are two plausible explanations for this finding. First, that news about the Scottish parliament has begun to be experienced as banal in Scotland. Second, that this reflects the different relationship that these new citizens have to historical process. It was notable that three of these 18-year-old students appeared unaware of the fact that the Scottish parliament was a particularly new historical development. None of them was aware of when it had been established, or had any conscious experience of a before and after kind. Although these data do not tolerate much in the way of reliable interpretation, they may act as a signal that, in future, analysis of historical change might do well to allow for the possibility that this may occur through generational shifts in perspective, rather than through the transformation of individual adults' established attitudes and identities.

3.3 Panel Sample

Our selection of respondents was based primarily on considerations of length of residence in England (ranging from 2 weeks to 70 years at time of first interview). For purposes of conceptual development, it was necessary to access as wide a variety of respondents as possible. We concentrated recruiting people from range of social class backgrounds (from homeless unemployed people, to the managing director of a major high street chain), people who were more or less integrated or active in civic society (from homeless people with mental health problems, to Church of England Ministers and an MP), people with a variety of political orientations, and people who originally came to England for different reasons. (Full details of the sample are available in our interim report). Sixty-seven people were interviewed in round 1. From these, 55 were selected for follow up (round 2) interviews.

AGE AND GENDER OF MIGRANT SAMPLE

PLACE OF ORIGIN OF MIGRANT SAMPLE

A relatively high proportion (29; 44.5%) of the Migrants sample originated from Glasgow and surrounding areas. Overall, however, we have a reasonable balance of respondents from the East and the West of Scotland. However, we have relatively few respondents from the North of Scotland (and only one from any of the islands).

On the basis of analysis of the first two stages of the panel study, we selected 12 respondents (six from Rother and six from Manchester) for the third stage of the research on four theoretically-driven bases: (1) People who had been short term migrants at time 1; (2) Other people who had originally come to England for different reasons, who were approaching what, on the basis of our first two rounds of research, we understood as various critical stages of the migration process (e.g. the ten year crisis point; children leaving home etc).; (3) People who had adopted various orientations to the question of UK constitutional change; (4) (In the light of the Edinburgh team's interest in 'return migrants') respondents who had either returned to Scotland or who continued to adopt the strategy of 'border hopping' (e.g. working during the week in England, but returning to their family in Scotland at the week end).

3.4 Summary of key findings

3.4.1 Concern over matters of national identity

Unlike our English Nationals respondents, the people who participated in the Scots Migrants panel found the topic of national identity in general, and of their own identity as Scots, thoroughly fascinating. We were often congratulated on our choice of research topic, and people expressed surprise that no previous research on this topic existed. To some extent, of course, this could be attributed to the fact that our panel was, of necessity, self-selecting (the only way in which we could identify potential panel members was by snowballing or by eliciting volunteers). However, in order to avoid overly biasing our sample at the recruitment stage, we fudged our particular interest in people from Scotland (advertising our interest in talking to, 'people who have moved to England from other parts of the UK'), and did not flag our interest in the question of identity per se. Interestingly, the six people we interviewed from Wales, whilst orienting to the relevance of their identity as Welsh, were much more sceptical concerning the ultimate point of the research.

With only two exceptions, there was no problem in getting the Scottish respondents to talk about their national identity, or to keep on topic. Quite the contrary. Respondents from Scotland had a tendency to spontaneously run through the entire gamut of our interview schedule without ever having been asked a direct question and with no further intervention from the interviewer.

On the one hand this was clearly a positive thing Ð it was certainly impossible to attribute our data to the effects of leading questions for example. However, there were two negative aspects to this. The first was that Scottish respondents had a tendency to cover everything we wanted to know before we had a chance to turn the tape recorder on. The second was that one reason why the Scottish respondents were so easy to interview was that their narrative accounts of self were typically well rehearsed. On the one hand, this meant that our data clearly had ecological validity (for example, when we interviewed Scottish people together with their English family members, it was clear that their spouses and children had 'heard it all before'). On the other hand, this had the disadvantage of making it hard for us to dig deeper than this glossy, superficial and well-structured narrative of self.

The schematic quality of may of the Scots' self-narratives made the repeat interviews particularly difficult to deal with: in many cases, the round 2 interviews were pretty well word-for-word the same as the round 1 ones. (Although this is, of course, interesting in its own right). The major exceptions were in the cases of new migrants (who were still constructing their narratives) and people undergoing some form of 'life crisis' that prompted a reconstruction of these narratives. Since the question of the ways in which a sense of one's life may intersect with a migrant career is interesting in its own right, for the purposes of the round 3 interviews, we focussed particularly on people who had already indicated that they had reached, or who were (by our estimate) ready to reach, a crisis point.

3.4.2 What is the problem'?

One reason why so many of our Scottish respondents were prepared to engage so enthusiastically with the research was that they could easily guess our research agenda, since it mapped almost perfectly on to their own lay sense of the migration problematic (except that not all respondents spontaneously recognized the potential significance of constitutional change).

Many respondents used a meta or micro-narrative starting point ('once upon a time') that referred to potential problems that they had envisaged when they had first moved to England. Two problems predominated, which mapped well on to our research problematics:

• the problem of social integration (would they be accepted?)

• the problem of identity, and especially of identity loss (would they be absorbed?)

Almost universally, respondents' narratives went on to suggest that these 'na•ve' concerns had proven entirely ill founded. However, this is not to say that their tales of migration always had a 'happy ever after' quality. Quite the contrary. As we shall see, they often involved elements of narrative complication, and sometimes even had a tragic quality.

In the following pages, we shall briefly summarise our findings as they relate first to the Scottish Migrants' accounts of social inclusion and civic participation, and second, to their own national and personal identity.

3.5 Scottish identity as a 'passport' to social inclusion

Respondents normally mentioned one or both of two concerns relating to whether they would fit in or be accepted in England.

á a concern that they might experience harassment or direct discrimination

á a concern that people in England would look down on them

However, respondents overwhelmingly reported that they had in fact experienced no real prejudice or difficulty concerning social acceptance. In addition, respondents commonly noted:

(a) that it took them a while to realise that they were in fact being accepted, since English norms of social reticence meant that it had been hard for them to spot the cues. However, once they realised the way things worked they 'made the effort' and found it easy to 'fit in'.

(b) Longer-term residents often noted how, early in their migrant career, their defensiveness itself, paradoxically, militated against their fitting in as easily or quickly as might otherwise have been the case. This also squares with our observations with respect to the changing attitudes of the shorter term migrants, who characteristically start out by complaining concerning the coldness, arrogance and unfriendliness of their new communities, but 'mellow' after about two years.

From our findings from the English Nationals panel, it seems likely that the ease with which the Scottish Migrants find themselves integrated into civil society in England is likely to be a function of three factors in the host culture.

(a) The fact that people in England do not typically employ an English/non-English axis of differentiation at the level of the local

(b) The generally positive attitudes that people in England have towards Scots in particular (it was notable that the people we spoke to from Wales reported far more experience of petty harassment and rudeness, although no instances of outright discrimination)

(c) The cache attached in England to the ability to claim ('interesting') extra-English national credentials

This is consistent with the accounts of many respondents that they had found (and continued to use) a display of Scottish identity (and especially a Scottish accent) as a social passport which gained them rapid entry into social arenas, and even allowed them to queue jump with respect to social relations. For example, respondents often noted that they found being Scottish speeded up the process of making friends, and helped them integrate into new workplace environments. We interviewed two young women in Manchester who were from visible ethnic minority groups (one of Pakistani ethnic heritage, and one of Hong Kong Chinese background) in the first three weeks of their arrival. Both women reported being quite astonished by their sudden popularity, as the combination of their skin colour and Scottish accents rendered them doubly distinctive and interesting in what was generally an ethnically and racially cosmopolitan area. (Interesting, England-born Pakistani women of the same age living in the same areas, reported experiencing Ð and enjoying Ð the functional personal invisibility which came from living in an area of high ethnic diversity).

3.6 The 'different cultures' of Scotland and England

As noted above, Scottish Migrants often displayed awareness of the very different rules of social intercourse in England and in Scotland.

This image may, in some cases, have been magnified by a tendency to elide the local with the national. As a consequence, respondents typically did not factor in any other potential axes of variability into their accounts of England versus Scotland. For example, people could treat their experiences of life in a large council estate in Glasgow or in a small rural community as exemplary of Scottish community life, and contrast this with their experiences of living in a wealthy middle-class Cheshire suburb ('England').

Many of the Scots Migrants who were interviewed in Manchester maintained their stereotypes of English inhospitality and incivility notwithstanding their own experiences of social inclusion by invoking a North/South distinction, and suggesting that it was easier for Scots to fit in in the North of England. However, people who had moved from Scotland to the South of England did not in practice report any more difficulties in fitting in than did those who had moved to the North. Although we did not include London as a research site in our study (a decision which, in retrospect, is possibly regrettable) on several occasions, respondents who had previously lived in London told us that they thought of themselves as a 'London Scot'.

3.7 Scottish social networking in England

Most of the Scots Migrants oriented to a norm precluding Scottish ex-pat cultural forms, especially membership of Scottish societies, and maintaining a Scottish network of personal friends. This form of cliquish behaviour was stereotypically associated with 'the Irish' way of living in England, from which many respondents were keen to distance themselves.

Paradoxically, there was a clear in-principle/in-practice distinction apparent with respect to the question of Scottish networking in England. Almost three-quarters of our respondents in fact were members of some sort of Scottish organization. Our examination of social networks in round 2 demonstrated that more than half of the long-term female migrants had a best friend who was also a Scot living in England. The majority of respondents at some stage intimated that they actively sought out other Scots in their local area.

To some extent this evidence of Scottish associational networks in England may have been a function of our respondent recruitment process. However, our experience of attempting to contact people who had been born in Scotland also demonstrated this rule. In practice, most of our Scots Migrants were recruited with the help of other Scots. This did not mean that they were always personally known to each other, but rather that people tended to be networked in such a way as to allow them to identify other Scots in the local area by word of mouth.

3.8 Civic participation and acculturation

One particularly striking finding in relation to the Scots Migrant respondents was their extremely high level of civic participation. We had originally attempted to sample people for varying levels of civic activity, but in the event found that even people who might, on a priori grounds, be assumed to occupy socially marginalised positions were, in practice, likely to be heavily and directly involved in community life, running voluntary organisations, organising community activities and so forth.

Again, this seemed to be a matter with respect to which a Scottish identity could function as a social passport. In this case, a habitas of 'doing community', in conjunction with a sense of 'permission' granted by the local population, combined to facilitate and legitimate people from Scotland running things.

Once we began our microanalysis of the interview transcripts, an interesting and (to some extent) counterintuitive finding emerged. Engagement in community activities in England did not relate in any appreciable respect to apparent levels of acculturation as evidenced either by the respondents' own self-report testimony, or by various empirical markers of a distinctively Scottish mindset.

In addition, for many respondents, aspects of a Scottish mentality persisted notwithstanding several decades of fully socially integrated, residence in England. Sometimes these could be effectively combined with aspects of English mentality (often lending an inherently contradictory quality to respondents' accounts). On other occasions, however, aspects of a distinctively Scottish mindset appear not only to have been entirely preserved, but for the respondent to lack any discursive consciousness of this fact.

One example provides a link to the topic of our next section Ð the use of the (self) category British. However respondents' self concepts might change over the course of their time in England, there did not appear any evidence that they picked up a tendency to call themselves British. Respondents adopted a variety of different stances towards British identity. However, the patterns of variability apparent within our migrant sample were entirely consistent with those apparent in the Scottish Nationals sample. Acceptance of the label British as self-descriptive was a function of generation and (conservative) political views. In addition, respondents' understandings of the term British in this context tended to have much more in common with those of people of a similar generation and political affiliation in the Scots Nationals panel than in the English Nationals panel, irrespective of length of residence in England, or opportunity to learn a new political and identity lexicon.

In addition to there being no obvious relationship between level of acculturation and level of engagement in civil society, we also noted that some aspects of a Scottish mindset appeared imperfectly related to formal educational experiences. In particular, it was notable that some accounting tendencies which are relatively common in Scotland, but virtually never seen in England (such as referring to the UK as the British Isles, or understanding the UK as the product of the England-Scotland union etc) could be seen in even in people who had moved from Scotland to England in early infancy, and second-generation individuals who had been born in England but who nevertheless self-identified as Scots.

3.9 Identity threat

As noted above, people who had moved from Scotland to England often oriented to a concern over potential identity threat. Many respondents reported or displayed a concern that they would need to protect their Scottish identity in England. Respondents often expressed concern that they might loose their accent, or become English. This was in turn related to concerns over recognition: that people in Scotland should continue to recognize them as really Scottish and that people in England should be aware of their distinctive identity as Scots.

According to many longer-term migrants' self-reports (which generally square quite well with our observations of people at the relevant stages of their migrant careers) this concern over the need to protect their Scottish identity from external threat was strongest for the first few years of residence. Thereafter, respondents were likely to become more 'mellow' as they came to realise that no-one was in fact likely to challenge their identity claims. Consequently, they slowly became 'less Scottish' (in the sense of being 'less obsessed about it'), although notably not 'more' English or British.

All of our respondents bar three described themselves as Scottish. The three exceptions Ð all women - being an Orkadian, the Hong Kong Chinese British woman, and a woman of mixed (Russian/Latvian) background. When asked on the scales: 'How often do you call yourself Scottish?'84% said 'always' or 'usually'. When asked, 'How proud are you of being Scottish'? 85% said, 'very' or 'extremely'.

However, behind these superficial patterns of self-labelling and reported identity attachment was a considerable measure of diversity. However, this diversity fell apart into a fairly systematic pattern relating to qualitatively different ways in which an individual could inhabit and identity as Scottish. This was related to different narratives of belonging, which assembled the narrative components of (i) self-as-individual (ii) nationality (iii) people and (iv) place in different ways. In the next section, we shall consider how these operated in terms of four common lines of self-narrative