Identity categories and colonial legacies in the British urban space

ABSTRACT

This article advances conceptual debates regarding identity and marginalisation in valuable ways and informs methodological improvements for representation within data collection systems. It discusses findings from a study exploring the inclusion – and exclusion – of identity categories within official data collection exercises. A theme that emerged from the research was how intricacies present within the lived experiences of ethno-linguistic groups can be ignored through the application of broad – and more politically visible – identity categories, such as Muslim. Furthermore, some participants felt that identity categories were discouraged based on colonial legacies and British government priorities for diplomatic relations. One such identity category discussed here is Kashmiri, which despite an ongoing campaign for official recognition, remained absent from the list of ethnic categories used for quantifying populations in the United Kingdom. Furthermore, within the research findings, Somalis and Pashtuns were also highlighted as groups, who despite being recognised as distinct communities within the urban space and thus acknowledged categories of practice in daily interactions within neighbourhoods in the UK, remain ‘unofficial’.

KEYWORDS:

• Diasporas
• self-identification
• politics of recognitions
• census

Introduction

This article discusses findings from a study exploring the perceptions and experiences of British Muslims in completing the National Census form. As such, the inclusion – and exclusion – of identity categories within official data collection exercises were explored. A theme that emerged from the research was how intricacies present within the lived experiences of ethno-linguistic groups within our neighbourhoods can be ignored through the application of broad – and more politically visible – identity categories, such as Muslim. This was viewed as particularly noteworthy given several communities who are amalgamated within a classification, despite competing for resources and recognition, may be masked and even disempowered through being analysed as a single identity category. Furthermore, some participants felt that identity categories were discouraged based on colonial legacies and British government priorities for diplomatic relations. One such identity category discussed here is Kashmiri, which despite an ongoing campaign for official recognition, remained absent from the list of ethnic categories used for quantifying populations in the United Kingdom. Furthermore, within the research findings, Somalis and Pashtuns were also highlighted as groups, who despite being recognised as distinct communities within the urban space and thus acknowledged categories of practice in daily interactions within neighbourhoods in the UK, remain ‘unofficial’.

The discussion presented within this article contributes to important reflection on how we view minority groups within our cities, by showing how lived identity categories differ from official categories, as well as by demonstrating the continuation of the political nature of categorisation and its contestation by groups in the post-colonial landscape. The article therefore adds to existing theory on the use of official categories as political tools both by the state (Bhagat, 2001; Cohn, 1996; Reddy & Gleibs, 2019) and by communities themselves (Appadurai, 2006; Brubaker, 2000, 2002; Modood, 2020).

By building upon earlier work by Cohn (1996), Bhagat (2001), and Khan and Qaiser (2022) it demonstrates how the legacy of the census as a political tool for defining social groups in the Indian subcontinent during the colonial period, and now post-colonial period, for example in Pakistan, continues within the diaspora. As such the article adds to our understanding of how contested ethnic categories can be imported with populations as they migrate, and therefore the demands for official recognition can often be embedded within wider historical and global context that pre-date and supersede the geographical parameters of such campaigns within Britain or the receiving country.

Secondly, the discussion in this article contributes to our understanding of the perceived temporal usefulness of official categories as a means of recognition and access to power, and in doing so furthers theoretical debates on drivers for the salience of collective identity markers (Nagel, 1994). By exploring whether the Muslim category could mask the needs of other political advocacy groups such as Kashmiri, it provides insight on how categories can lose support or be perceived as less relevant in the face of different competing political positions among the same population. Just as group identities are argued to increase and decrease in the salience groups afford them – this article argues that the meaning attached to official categories can change for those who they are meant to represent.

Although the study was not intended to provide a representative overview of how British Muslims view the Census, it does offer an indicative exploration on how collecting large-scale statistics on Muslims is perceived and thus highlights areas for consideration by both academic researchers and government agencies, when engaging with Muslim communities for data gathering exercises. In doing so it advances conceptual debates regarding identity and marginalisation in valuable ways and informs methodological improvements for representation within data collection systems.

Background

Large-scale migration to the United Kingdom took place after the Second World War, as migrants from former colonies were encouraged to take up unskilled labour in Britain to rebuild the country. Although there had been smaller scale migration before this period from the colonies, for example Ansari (2004) and Ahmed (2014) describe seafarers settling in port towns during the 1800s, it was not until the post war period that the presence of non-European origin ethnic diversity witnessed today became a feature of Britain. Migrants from the Indian subcontinent and the Caribbean began to establish visible communities from the 1950s in Britain’s cities and towns (Bharadwaj et al., 2008; Modood, 2020). Subsequent waves of migration from other locations have followed, such as refugees from Somalia, the former Yugoslavia and more recently, Poland, Romania, and Ukraine (Blachnicka-Ciacek & Budginaite-Mackine, 2022; Valentine et al., 2009).

This has now led to the UK becoming one of the most ethnically and linguistically diverse countries in the world, however, the geographical distribution of ethnic minorities and migrant groups is far from even (Vertovec, 2019). One of the key sources of information on the growth and distribution of Britain’s various minority groups is the National Census of Population, conducted every 10 years. Despite the established presence of ethnic minority communities in the UK, it was not until the introduction of ethnic (1991) and religious (2001) categories in the National Census that policy makers and academics could fully understand the ethno-religious features of British towns and cities.

The National Census in particular is an important symbolic apparatus for recognising ethnic groups, and as such it acts as an important source of reference for all other government surveys (Aspinall, 2009). Thus, when a group or community have successfully gained a place on the Census form, it is an indication that they have obtained official recognition as a collective, which in turn requires other agencies to also acknowledge them as such (Modood, 2020). Gaining recognition as an ‘official category’ in the Census, therefore, affords a minority community legitimacy, as a stakeholder group in their own right, for government attention and access to resources.

Although the ethnic and religious categories included within official government surveys are often taken for granted by citizens who are asked to complete forms, there has been considerable academic attention paid to why some minority group categories are included (become official) and why others are not (Aspinall, 2009; Brubaker, 2000, 2002; Modood, 2020). For example, although Somalis make up a sizable population in the United Kingdom, they do not have their own ethnic category on the Census form, yet Chinese, who are smaller in number, do (Hussain, 2007). Furthermore, Kashmiris form the majority share of the British Pakistani population, yet despite campaigns (see for example, British Kashmiri Identity Campaign) to gain their own category they remain amalgamated within this broader classification (Hussain et al., 2021). Thus, within our cities there exist both official and unofficial identity categories, with the former gaining status as legitimate stakeholders for government attention and access to resources (Brubaker, 2002).

This article discusses such ‘taken for granted’ official categories using the example of the Muslim category, which was added to the National Census (together with other faith categories such as Christian, Jewish, Hindu, Sikh, and Buddhist) in 2001 in order to understand what categorisation does in terms of capturing certain experiences while also leaving out others within the British urban context. As such, the paper explores how by granting some identity categories official status may contribute to an ethno-religious classification scheme that allows marginalities to be seen while also collapsing diverse groups into a singular collective. Furthermore, it illustrates through research conducted within the British urban space how unofficial group categories continue to matter within the post-colonial city, even when they are not officially recognised by government apparatus.

The next section discusses why categorisation of ethnic and racial groups was an important strategy undertaken as part of the colonial project and how categorisation continues to be a political endeavour (Bhagat, 2001) in the post-colonial metropolis. In doing so it explores how categories are not equal in terms of the status afforded to them and how the unofficial/official dichotomy plays out at the neighbourhood level within ethnically diverse cities.

The article adds to the existing theoretical discussion by building upon the earlier work of Cohn (1996), Bhagat (2001), and Khan and Qaiser (2022) on how the legacy of the census as a political tool for defining social groups in the Indian subcontinent during the colonial period, and now post-colonial period (e.g. in Pakistan), are also present within the diaspora. As such the article adds to our understanding of how contested ethnic categories can be imported with populations as they migrate, and therefore, how demands for official recognition can often be embedded within wider historical and global context that pre-date and supersede the geographical parameters of such campaigns within Britain or the receiving country.

Categorisation matters

Religion, ethnicity, and national origin are important collective identity markers for many individuals (Brubaker, 2000, 2002; Fukuyama, 2018). Social actors organise themselves around such identities, which provide meaning and value. Social construction theorists also describe how group identity is dialectical and relational and thus self-identification is negotiated through social interaction with others; and group boundaries (real or imagined) become established as social categories (Jaspal & Cinnirella, 2012; Stets & Burke, 2005; Wimmer, 2008). As a result, such categories are often developed and defined by their markers of membership by enhancing descriptions not only of who they are, but who they are not (Eriksen, 2001). In this way, when social categories become official – that is formally recognised within government statistics – not only do they provide an acknowledgement of a group’s existence within society, but also cement boundaries for belonging. The benefits attached to gaining such recognition, Mateos et al. (2009) argue, can be seen through an increase in official ethnic categories between the 1991 and 2011 in the UK Censuses, as a result of groups lobbying for their identities to be included.

Yet Jenkins (2002) highlights the analytical importance of differentiating between self-identification with a group and official social categories. In other words, ethnic identities can be ascribed to actors as in-group members and ascribed to actors by out-groups and official ‘others’. The latter is discussed by authors as not only a function of governance but also of unequal power relations and even dominance of groups over others. Aspinall (2009) argues that official categories are compulsory categories that are forced upon society. Groups are obligated to choose from a limited number of politically defined classifications. In addition he argues that such defined categories carry varying degrees of privilege or stigma which groups may internalise and begin to accept, even when such categories do not accurately represent those they are ascribed to (Manuel, 2006; Nagel, 1994).

Therefore, discussed within literature on social categories is the problematic nature of groups being assigned to categories by others rather than having the right to self-definition and determination for nominal identification. This has been covered by debates within post-colonial studies, most notably in Edwards Said’s work (1978), as well as Cohn (1996), Bhagat (2001), and more recently, by authors Reddy and Gleibs (2019). They discuss how there remains a direct link between the implementation of categories by colonial rulers and the way ethnic and racial populations continued to be thought of in the post-colonial period. In understanding why official categories can hold such weight in terms of how they continue to stratify and ascribe (often negative) meaning about populations they describe, Reddy and Gleibs write,

this formal categorisation process [which] has its roots in British colonial management of diverse populations is relevant because British colonial strategies of ‘define and rule’ were created to determine people’s function in the colonial economy’ (Mamdani, 2012). For example, Indians were mainly recruited to work as ‘coolies’ in plantations, Chinese peasants were segregated in the tin mines and the local Malay peasantry were largely left bound to their rural-based activities (Hua, 1983). This separation meant that the heterogeneity within the diverse populations was collapsed into simplified racial categories for ease of administration. Not only did colonial administrators stratify the population administratively, they also attributed specific characteristics to members of each racial category. (2019, p. 4)

The authors conducted a study examining how individuals viewed the three dominant ethnic categories – Malay, Chinese, and Indian – present within contemporary Singapore, which was colonised under British rule. They were able to link how participants currently viewed these groups and the types of characteristics they ascribed to them with those established during the colonial period.

Yet despite the seemingly formal nature of official categories as means for imposed stratification, and units of analysis (e.g. keeping population records) as part of the colonial project – categories were not fixed in terms of who they included as the following discussion by Stoler (1989, p. 154) describes,

Something as apparently basic as who could legally be deemed a European differed across the colonial context, revealing discrepant and changing criteria by which racial superiority and attendant European privilege were assigned. For example, in the Netherlands Indies during the early twentieth century, the legal category of ‘European’ paralleled only loosely the idea of ethnic European origin. Included within it were Japanese, Jews, Arabs, Armenians, Filipinos, naturalized Javanese, the Sundanese wives of Dutch-born bureaucrats, the recognized children of mixed marriages, and Christian Africans, among others’

Thus, the above quotes demonstrate how ethnic, racial, and class/economic identities were created and assigned to populations by colonisers. Yet far from being static or rigid in terms of their inclusion, diverse populations were aggregated and collapsed into a single category and others were expanded to temporarily include less privileged groups depending on economic or political agendas. Furthermore, categories carry with them ideas about populations generated during the colonial period into the contemporary urban space. The key argument for the purpose of this paper is that social categories were legitimised by ruling elites and as such became ways in which groups were defined and stratified. The legacy of this exists within the British urban space in terms of the decision to grant a social category official status or remains essentially a political one.

Authors such as Brubaker (2000) discuss how the continued use of official or government adopted, recognised, and approved categories, which he describes as ‘categories of analysis’ differ from unofficial ‘categories of practice’. The former are those included on the Census form, as they are employed by social analysts within government agencies to classify populations and understand trends. However, ‘categories of practice’, which, although not directly quantifiable or officially recognised by governments, exist nonetheless, in the lived experience of groups or collectives which are, ‘native or folk or lay categories … of everyday social experience, developed and deployed by ordinary social actors’ (Brubaker, 2000, p. 62). Therefore, a group can self-identify with an ethnic label, which is embedded within historical narratives and even acknowledged as a category of esteem before arriving in Britain, yet due to lacking official recognition within the UK, their identity remains ‘lay’ (Brubaker, 2012).

Within Britain, categories of analysis remain at the forefront of how we stratify communities, including those groups who select the same broad tick box, yet differentiate themselves in keyways to others within the same category. Although unacknowledged by officialdom, categories of practice are present for all to experience within the lived realities of exchange, be it in how they impact decision making on political allegiances (such as baradaris) or life choices such as suitable marital partners (see Ballard, 2004).

Although social categories have been widely criticised as a form of imposed identity, they have also been hailed for providing visibility to groups, increasing their chances of gaining access to resources and tackling inequalities (Burton et al., 2008). The benefits of becoming a social category and thus an officially recognised group are paramount for the acknowledgement of ethnic identities. Nevertheless, this paper thus shows that categories of practice must be accounted for when re-conceptualising the post-colonial city and its relationship to the nation-state.

The Muslim category

Faith-based groups lobbied extensively to gain official statistics on religion. This took the form of counting ‘religious affiliation’ which is very much a sociological term by focusing on group membership, identity, and self-definition, rather than a yard stick to measure levels of spiritual practice. The inclusion of a question on religious affiliation was viewed by practitioners as both an important tool for gathering evidence by policy makers to better cater for communities, as well as recognition of the value placed on religion for the self-identification of citizens in contemporary Britain. Hussain and Sherif (2014, p. 417) write,

The issue raised about distinguishing the ‘religious’ element from ‘cultural’ or ‘ethnic’ identification became increasingly contested by Muslims who challenged the notion that religious beliefs ought to be situated within domestic and personal domains (Weller, 1998). Rather than, with time, coming to de-emphasise cultural and indeed religious markers, the opposite occurred. Greater recognition as what was, in the British context, increasingly identified as a ‘faith community’ correlated with the intensified demand for the right to practise the tenets of their religious duties outside of the home, in response to which more mosques, Islamic schools and cemeteries were established (Peach & Gale, 2003). In addition to growing pressure to accommodate religious practices in public arenas, there was a documented increase in the number of Muslims using their religious identity in addition to, or in place of, other social markers.

In addition, commentators described how by grouping hitherto disparate communities together provided a greater force for domestic lobby, as well as for international political networks, particularly for responding to the surge in hostility towards Muslims post 9/11 (Ballard, 1996; Saint-Blancat, 2002). As such, a unified Muslim category turned a number of smaller ethnic minority communities into the second largest faith group in the country. There was also existing evidence presented by social scientists on how religious affiliation was a key predictor of socio-economic and demographic outcomes and behaviours, including the propensity for marriage, divorce, types of occupations, and women’s engagement in employment, for example (Hussain, 2017; Khattab & Johnston, 2013).

Nevertheless, presenting Muslims as a single community was cautioned by others (Hellyer, 2005). Here, the dangers of an organisation acting as a representative body for all Muslims when lobbying for policy changes could ignore the needs or views of many falling within this category (Ali, 2007; Hussain & Sherif, 2014). Additionally, Hellyer (2005, p. 83) argued against assuming a commonality among Muslims because they share a faith background, he writes,

Muslims [are not] an ethno-cultural group by virtue of being Muslim. Islam is not per se a culture rather it may give rise to cultures, or be a fundamental root of culture, but it is not, nor has it ever been a culture in and of itself. Hence whereas Islam as a faith spread across the world, the culture of the first Muslim community (Arabs of the Hijaz) did not … Thus the concept of an Islamic ‘identity’ per se is problematic.

Nevertheless, the campaign for the inclusion of religious categories in official surveys was successful and identities based on religious affiliation became ‘categories of analysis’. Yet the decision to collect data on religious groups took place before the swift rise in both reported Islamophobia and widespread public scrutiny of Muslims, post 9/11 which was followed by government policies in the UK known as PREVENT (Abbas, 2020; Hafez, 2018; Law et al., 2019). Thus, what had been seen as a historically defining moment for British Muslims, in gaining official recognition as a community, came to be perceived as a potential means for surveillance and scrutiny (Hussain et al., 2021, et al).

Against this backdrop, there remain communities who lobby for official recognition through categories which are believed to better represent their needs (Ali, 2007; Kalra et al., 2019). Some activists argue that by presenting their communities as ‘Muslim’ has cast a shadow over important nuances and led to a neglect in catering for specific community needs. Two such groups are Kashmiris and Somalis – both of whom within a British context are almost entirely Muslim in terms of faith affiliation, yet despite forming substantial shares of the British Muslim population, and indeed the United Kingdom’s ethnic minority populations, neither have an ethnic category within official data collection exercises (Hussain et al., 2021).

Migration from Azad Jammu and Kashmir (AJK or Azad Kashmir for short) to Britain commenced during the colonial period, however it was not until the 1950s that larger waves of migration from the disputed territory began to take place. Today it is estimated the diaspora form a third of all British Muslims, comprising one of the largest non-European origin communities in the UK. Yet very little is known about the socio-economic, political, and ethno-linguistic nuances outside of the community itself, despite it forming a large segment of Britain’s non-white population. Based on various ‘guesstimations’ there are more than one million Azad Kashmiris (also simply referred to as Kashmiris) in Britain, among whom a growing number of scholars, activists, and writers continue to emerge. However, their work remains at the periphery when discussing minorities in the UK. This is largely a result of coming from an internationally disputed territory (Jammu and Kashmir) and being amalgamated with Pakistanis, by British officialdom, upon settlement in the UK.

Nevertheless, as grassroots identity movements continue to demonstrate, a significant number of those from the community (in the homeland and diaspora) believe this masks both internal nuances, but also has a detrimental impact on the lived experiences of the community it terms of representation and policy provision. Furthermore, there is evidence of discrimination towards Azad Kashmiris by other South Asian heritage groups within a British context (Kalra et al., 2019) this often involves negative stereotyping and degradation of language and culture, as well as access to political platforms as ‘Kashmiris’.

Therefore, what is unique about this community is that, although it forms a large share of the ethnic minority population in Britain, it in many ways remains hidden. This is despite the fact that there was migration to the UK from Jammu and Kashmir before Pakistan had become a nation-state, and there are more people within the British Pakistani category from Azad Kashmir than Pakistan itself. Furthermore, many pioneers from South Asia to participate in political activism and trade unions, in a British context, were from Azad Kashmir. Yet issues of identity and recognition as a result of originating from an internationally disputed territory continue to play out, both decades on and thousands of miles away.

Unlike Kashmiris who arrived overwhelmingly as economic migrants, the largest share of the Somali population in the UK arrived as forced migrants due to communal conflict, famines, and natural disasters from the 1990s (Harris, 2004; Valentine et al., 2009). There were Somalis in Britain prior to this however, and earlier generations were seafarers, forming self-contained communities since the nineteenth century (Ahmed, 2014). This earliest settlement reflects Britain’s colonial ties with Somalia, recruiting Somalis to fight and serve in the royal navy before and during the First World War; and to work in the docklands areas such as Bristol, Cardiff, and South Shields. The second phase of Somali migration to Britain came at the end of the 1950s and although the demand for seamen had declined, an economic boom meant that there were opportunities for employment in industry. Communities grew in Sheffield and Manchester as whole families, not just single men as previously the case, settled in these areas. The third phase of Somali settlement was not due to labour shortage but due to political instability in Somalia (Valentine et al., 2009). As a result, the most significant numbers of Somalis arrived in Britain as asylum seekers. In general, it is believed that 90% of Somalis arrived in the UK from the 1990s onwards (Hussain, 2007).

Overall the calculations for Somalis in Britain without the presence of official data are unsatisfactory. Despite the length of residence in the UK their presence remains largely overlooked by mainstream society. Khan and Jones (2002) explain how most attempts to classify Somalis fail to take into account their distinct position as the interface of two very different cultures. Subsumed within the category Black African, Somalis do not share a language, diet, dress, or religious practice with their near neighbours. As Muslims, Somalis worship at mosques alongside others from Asian and Arab countries yet do not share other aspects of their culture. Harris (2004) and Hussain (2017) argue that British Somalis suffer from multiple forms of discrimination, firstly as asylum seekers, who are known to face hostility within public forums, secondly through racism as a black community, and thirdly, as a result of their dress and religious practice, Islamophobia, as visibly identifiable Muslims. Yet the lack of sensitivity in monitoring categories has frequently resulted in the Somali community’s needs being overlooked.

The next section discusses findings from a study with Muslims in England on their engagement with the census form. The research employed a grounded theory approach to understand processes taking place as perceived by the actors involved (Glaser, 1992; Glaser & Strauss, 1967). The findings were expounded using existing theoretical approaches for exploring the existence or absence of official group categories, and on how the census can act as a political tool for group recognition.

Methodology

As described in the sections above, the Census is an important tool for understanding Muslim demographics and social trends, however, as studies on perceptions and participation with other minority groups in the UK demonstrate (Cartwright, 2021), there is evidence of lower engagement with the Census among some groups. As a result, a research study conducted with members of the British Muslim population was undertaken to explore perceptions of the Census and how the data is used; as well as to understand participant’s experiences of engaging with the Census exercise itself.

Participants were recruited through a number of platforms. Flyers providing information on the study were distributed through national and regional organisational networks such as British Muslim Heritage Centre, the Muslim Youth Foundation, the Positive Youth Forum, Community Cohesion West Midlands, the Coventry Census Community Engagement team; as well as through Somali networks via mosques and WhatsApp groups; and Kashmiri Facebook pages, WhatsApp groups and the Kashmir Youth Foundation.

As Census 2021 was conducted online it is likely that younger members of Muslim households would have completed the form. The mixture of in person and online focus groups was partly a necessity of COVID-19 restrictions, during April and May 2021. In-person focus groups, particularly with young people (aged 18 and above) were allowed as a result of guidelines which permitted people to gather at sports and leisure facilities. These primarily took place in the West Midlands. The Zoom focus groups on the other hand allowed for participants from across England and Wales to attend the discussions. Locations from which participants attended were London, Luton, Bradford, Birmingham, Manchester, Oldham, Coventry, and Rochdale.

Focus groups were chosen as a method to explore what can be complex lived experience. Gibson (2008) explains how group interviews can provide a more familiar setting and allow individuals to discuss topics in using language they would employ during everyday interaction with their peers. Furthermore, focus groups involve discussions, which allow multiple views and experiences on a topic to be explored (Robson et al., 2011). Focus groups were therefore believed to allow (i) more in-depth exploration of perceptions of the Census and (ii) experiences of completing the census form.

As the research was conducted during the period in which Census 2021 was underway, in order to capture experiences of participants as close as possible to completing the form, the timeframe in which to carry out the focus groups was constrained. As a result, nine focus groups were conducted. As such, the study was not intended to provide a representative overview of how British Muslims view the Census, however it does offer an indicative exploration on how collecting large-scale statistics on Muslims is perceived and thus highlights areas for consideration by both academic researchers and government agencies, when engaging with Muslim communities for data gathering exercises. Future studies exploring official statistics, without such time constraints would benefit from undertaking larger numbers of focus groups to allow participants from a wider demographic and geographic range to further explore the diversity of perspectives among British Muslims.

Of the nine focus groups, four took place in person. The ages of participants were categorised into three broad brackets 18 to 29; 30 to 40; and 41 and older. The majority of participants fell between 18 and 30 reflecting the demographic profile of Muslims in Britain, who are skewed towards the younger age brackets. The sample included a range of ethnic groups, with approximately two thirds of participants from South Asian backgrounds, however there were also participants who identified as having North African, Arab, and Somali heritage. The sample included full time students, retired, unemployed, and employed in varied professions, including mosque imams and a barrister.

The research was approved by the author’s university ethics committee. All participants provided either written or verbal consent to having the discussion audio recorded, and to treat the views and experiences shared during the focus groups as confidential. All focus group audio recordings were professionally transcribed in written verbatim form. The author then checked transcriptions against the recordings, and the transcribed data were read multiple times to establish familiarity before commencing the analysis, with initial notes taken during this familiarisation process. In order to organise the data, a coding process was initiated by breaking down the data into manageable coherent segments. Each focus group was comprehensively assessed to identify significant sections that were subsequently highlighted for coding purposes. Codes were then aggregated into categories, with codes sharing similar conceptual ideas being grouped together. Potential themes were then generated based on patterns identified within the categories, thus by merging categories broader themes began to take shape.

This was followed by a process of reviewing the preliminary themes through discussions with the focus group facilitators. This step was instrumental in ensuring that the identified themes were both meaningfully represented and in alignment with the original data. The goal was to both provide consensus on identified themes, as well as prevent any potential themes from being overlooked. This final refinement stage involved defining and naming the themes in a manner in which they were accurately conveyed and without compromising their core concepts. The analysis therefore allowed meaning to be derived from the focus group interviews through identifying themes that influenced how participants viewed the census as well as factors for hesitation when completing census forms.

The following discussion which arises from the focus group analysis therefore provides a valuable insight for exploring how members of ethnic minority groups view ethnic and religious categories on such forms. In doing so the next section discusses how group categories can be imported with populations as they migrate, and how the demands for official recognition can often be embedded within wider historical and global context that pre-date and supersede the geographical parameters of such campaigns within Britain or the receiving country.

The unofficial hidden by the official or minorities within minorities?

Although the ethnic and religious categories included within official government surveys are often taken for granted by citizens who are asked to complete forms, there has been considerable academic attention paid to why some minority group categories are included and others are not.

Participants who took part in focus groups were asked to discuss how they felt about communities such as the aforementioned Kashmiris and Somalis, who despite forming sizable portions of the Muslim population, do not have their own ethnic categories in the Census. Discussions on this particular topic provided valuable insights into how some participants believed by placing emphasis on ethnic identities, above a religious one, could lead to divisions among British Muslims. In this view, such differentiation could potentially weaken campaigns for the practical needs of all Muslims.

Nevertheless, others challenged such views. A female participant described a discussion she had with her neighbour on whether Kashmiris should ask for language recognition, and how recognising different cultures, languages, and identities did not detract from the belief that Muslims are ‘all one body’. In this view, respondents discussed how it was of course possible to choose to tick the Muslim box in addition to an ethnic box on the Census form. However, the ability to simultaneously identify with both an ethnic and religious group did not change the fact that numerically significant Muslim communities had expressed dissatisfaction at the fact that their identities are not represented within such forms (Ali, 2007; Hussain et al., 2021). This was exacerbated by the fact that Muslim populations experience poverty and discrimination differently based on ethnicity and other identity markers (Khattab & Johnston, 2013), not least due to the existence of inter-community prejudice based on heritage (Kalra et al., 2019).

There has been opposition in the face of attempts to differentiate groups from existing official categories as the case with The British Kashmiri Identity Campaign, which was established to lobby agencies for a separate Kashmiri identity. Despite such efforts, the 2021 Census did not include a Kashmiri ethnic category but provided an option to select Pakistani or self-identify as Kashmiri using the ‘Other’ category option.

The Kashmiri community had a very visible and pro-active campaign for the Census, which likely encouraged more people to engage as a result. Several respondents were aware of the campaign and had commented on the need for a separate category. One described her experience in encouraging members of her community to do the same:

We did get a bit of funny looks, when there was certain people that were in the supermarket that weren’t from Kashmir and weren’t speaking that [Pahari] language and they were Urdu or Punjabi speaking. It was almost like, I think some people find it quite offensive that we are doing this [asking for Kashmiri recognition] and we are trying to segregate. But whereas what we were trying to say is that it’s just us getting recognised, that’s all it is.

A point of discussion to arise from the focus groups worth elaborating on here, is the ongoing legacy of the colonial project and how it continues to play out both officially – in terms of awarding communities legitimacy based on the identity marker they choose; and within grassroots communities through ongoing politics of the homeland impacting the lives of diaspora. On 5 August 2019, India unilaterally revoked the semi-autonomous status of Jammu and Kashmir (hereon referred to simply as Kashmir). This led to large protests by members of the Kashmiri diaspora in Britain’s major cities against the Indian government; as well as calls upon the British government to intervene diplomatically, given it was party to the Kashmir dispute in the first place. One of the direct consequences of de-colonialisation was the carving up of lands that has left populations fragmented and as internationally disputed territories – as the case with Kashmir. Malik (2020, n.p.) writes, ‘According to Cohen (2003), the tensions in the [Kashmir] conflict have been nurtured by a pervasive psychological distrust between the two countries [India and Pakistan], the root causes of which can be traced back to British colonisation’. This distrust was discussed within focus groups, however here it was in the form of both British Pakistanis and British Indians being perceived as unsupportive of the Kashmiri census campaign.

Interestingly, within focus group discussions there were arguments presented as to why it is more legitimate for some groups to make demands for an ethnic category over others; as well as motivations on the part of agencies for not officially recognising certain groups, as the following quotes describe:

Africa is a continent with so many cultures, but Somalis are expected to select ‘African’ as though that can tell you anything. I definitely think that Somalis do have specific cultural, social, political, historical needs that otherwise would be [overlooked], that data would not exist and I think it’s important that this is recorded.

Another respondent added:

Especially [noticeable] with Somalis. The Kashmiri thing, I do feel like maybe it was a political reason not to include it, whereas Somali, I don’t feel like there’s much politics behind it, it’s more, it’s overlooking something or overlooking a, a growing number of people, Somalis, living in the UK.

Here there is a differentiation expressed between a potentially deliberate omission – Kashmiri, and an oversight – Somali, however even this should be framed within the wider colonial lens. The previous quote discusses the single ‘Black African’ category in the census, pointing out that this is problematic as Africa is a whole continent with diverse ethno-linguistic and cultural groups. Yet collapsing all Black Africans into a single category for official data collection could be argued to echo portrayals of peoples and geographies developed during the colonial period. There is a substantive literature on how Africa was deliberately ‘downsized’ in the popular imagination in order to reduce its perceived significance, and a feature of this was to minimise its size in cartographic representations (Mutua, 1995), as well as reducing it from a diverse continent, within the broader worldview, to a monolithic space (Abrahamsen, 2003; Ndlovu-Gatsheni, 2018). It is therefore worth considering how such collapsed categories of this kind within contemporary Britain may reinforce such notions of peoples and places.

Similar to the above respondent, another participant felt that the omission of a Kashmiri category could be a political decision. This was framed as being an outcome of political lobby from India and British Indians, and the government deciding on whether they should promote particular lobbies:

I think it’s like a sanction, to say that wait, wait a minute, we stand with such a group but not with such a group. So to keep one happy and not the other.

The thing is of course the Indians wouldn’t be happy and Pakistani would lose a massive chunk of their numbers, because I think it’s like 70% of [British] ‘Pakistanis’ are actually Kashmiri! So it’s definitely politically tricky on both sides if Kashmiris get their tick box and separate out.

For many within the diaspora, as raised within the interviews, there is a perception that international politics continues to influence state approaches to grassroots administration in British cities (Heath et al., 2013). When Kashmir became forcibly split between the newly formed India and Pakistan in 1947, the United Nations Security Council (UNSC) stated that Kashmiris should be supported to undertake a referendum so that they can decide on their sovereignty. In this view, accepting Kashmiri among British government categories would provide the diaspora with ‘symbolic legitimate force’ (Brubaker, 2000) as a collective, officially recognised as an ethnic group in their own right.

This has consequences both for their right to lobby the British government – but also for India – who insist that Kashmir is an integral part of the Indian Union (and thus Kashmiris are Indians) rejecting any move that would add legitimacy to their struggle as an independent population. Becoming a legal ethnic group in one of the world’s most powerful nations would afford Kashmiris – by proxy – a very real ethnic identity. Whether or not this is indeed the motivating factor behind the British Kashmiri lobby being rebuffed for three Censuses in a row, it is worth reminding ourselves how the colonial project continues to have very real consequences for how communities in contemporary Britain experience their collective identity and perceive marginalisation.

In addition to Somalis and Kashmiris there were other groups who were discussed as not having their own official ethnic category during the research including Kurds and Pashtuns. During a focus group in Birmingham in which there were several ethnic Pashtun youth present the conversation turned to how they, here referred to as Pathans, were identified on a macro level as Pakistani, however they expressed experiences of the contrary:

P1: I feel basically when you look on and you are filling the form, just there is for example it just says Pakistani, British Pakistani, it doesn’t specifically say what we are Pathan and stuff so I feel like we are being pushed as a minority-

P2: Even further.

P1: Even further and I feel like it is a concern because are we being looked at or are we just pushed into a corner like what it has already been defined as, as Pakistani.

P3: I feel like we are just basically just a tick of a simple box, we are not given that [Pathan] identity in a certain sense.

In the UK, ethnic Pashtuns originate from Afghanistan and Pakistan, where in the latter they form the majority of the population of two of the country’s provinces – Baluchistan and the Khyber Pakhtunkhwa. Although Pashtun’s are overwhelmingly Muslim (there are Sikhs and Hindu minorities) they are considered within Pakistan, and indeed the Pakistani diaspora, to form a distinct ethno-linguistic group. Pashtuns are believed to follow a traditional code of conduct known as Pukhtunwali which continues to influence their social norms, although Khan (2016) discusses how increasingly Pashtun communities have been moving away from such traditions. Nevertheless, a distinction remains as a category of practice within the British urban space.

Like Kashmiris, Pashtuns, primarily from the Attock area, were recruited by the British to work in the naval forces during the colonial period, and therefore had initial exposure to Britain, before the 1950s wave of migration. Yet the relationship between the British and Pashtuns was known to be fraught and during colonial rule, harsh draconian policies were implemented within locations in which ethnic Pashtun were numerically dominant due to stereotypes about the ‘exceptional character of the frontier’ and its ‘warlike’ people (Condos, 2015, p. 479). Such colonial policies were inherited by Pakistan (see for example Callen et al., 2015 on the Frontier Crime Regulation) and only recently abolished due to their harsh and discriminatory nature. Therefore, the legacy of the colonial stigma imposed upon this ethnic group remained long after their rule ended within Pakistan and influenced how the state and the general public viewed Pashtuns prior to their migration to the UK (Rais, 2021).

The discussions above on the categories of practice described by Brubaker (2002) are best illustrated through scenes within our neighbourhoods, such as Small Heath in Birmingham, where the focus group took place. On Ladypool Road, which is home to a large community from Azad Kashmir we frequently hear Pahari being spoken by both the shopkeepers and their customers. Yet walking into a fashion boutique, the tailor employed sitting in the back of the store swiftly shifts from Pahari to Punjabi as a bride-to-be visits from out of town and discusses the alterations of her outfit. There is a saying that whenever you find more than one Pathan they will only speak in Pashto, the sound of which can be heard under the Pashtun pop music playing in the takeaway that sells Kabuli Pillau by the kilo. Despite all three of the above-described groups falling within the Pakistani category of analysis, we see the differentiation play out within the daily lives of communities within the streets of our cities. It was stated earlier how a group can self-identify with an identity label – e.g. Pashtun or Kashmiri, which may well have been known and well established, even before their arrival on British shores, yet due to lacking official recognition it remains ‘lay’. Despite this, in the British context, such identity-labels remain apparent and even those who would tick the same box for a broader category – e.g. Muslim and Pakistani, distinguish between themselves for what are perceived as important matters. Is this problematic? After all Brubaker himself suggests not all groups who self-identify with an ethnic identity need to have a corresponding category of analysis.

Members of the British Kashmiri community who actively lobbied for official recognition as a category in their own right would argue that it is deeply problematic based on findings that suggest they suffer from some of the highest levels of disadvantage and discrimination from both outside the Muslim community and within it (Ali, 2007; Kalra et al., 2019). If the mere existence of a collective identity (as a category of practice) is not enough to demand official recognition, how can we ensure equal representation, as a non-official stakeholder group, for government attention and access to resources?

As Muslims are the most ethnically diverse minority faith group in Britain it is important to undertake additional research on the impact of a lack of ethnic categories, given some sub-communities within the Muslim umbrella demonstrate higher levels of discrimination (Khattab & Johnston, 2013) and socio-economic deprivation (Hussain, 2017). Evidence exists of ethnic hierarchies within Muslim population outcomes, in that racism is not experienced in the same way by all groups, and this too is a consequence of colonial ideas that were exported across the Muslim world. Feagin (2014, p. 8) argues, ‘European colonialism and imperialism … reached much of the globe and created a global racial order, which has had severe consequences for the world’s peoples for centuries’. There is also a growing literature on colourism within the Muslim community that deserves more attention in understanding the interplay of categories of practice within our neighbourhoods (see for example Black Muslim Forum, 2019). Therefore, gaining accurate information for targeted service provision and support, as well as understanding how the intersectionality of race, culture, and religion impacts communities lived experience is crucial.

This section discussed how the imagined collapsing of some lands, and carving up of others and their populations as consequences of the British colonial project exist within our cities. The continued impact of this is seen through communities who protest for the right to self-determination for their ancestral land of origin; and their right for self-definition through being recognised as an official ethnic group in the UK. The discussion within this section therefore argues how categories of practice remain and ought to be acknowledged when re-conceptualising the post-colonial city and its relationship to the nation-state.

Conclusion

Although the ethnic and religious categories included within official government surveys are often taken for granted by citizens who are asked to complete forms, there has been considerable academic attention paid to why some minority group categories are included and others are not. This article contributes to scholarship on this topic. As such, it explores how by granting some identity categories official status may contribute to an ethno-religious classification scheme that allows marginalities to be seen while also collapsing diverse groups into a singular collective.

A key argument presented within the discussion of this article is that social categories were legitimised by ruling elites and as such became ways in which groups were defined and stratified (Khan & Qaiser, 2022). The legacy of this exists within the British urban space in terms of the decision to grant a social category official status and therefore remains essentially a political act. The article provides an illustration of this through research conducted within the British urban space, which discusses how unofficial group categories continue to operate within the post-colonial city, even when they are not officially recognised by the government apparatus.

The inclusion of categories on religious affiliation in the Census provided legitimacy for religious minorities, who hitherto had been present within public policy debates based on race and then ethnic group status, with Muslim organisations actively pursuing the government to collect statistics on religion (Hussain 2007; Modood, 2020). However, some communities claim that this led to the de-emphasising of other identity markers, and even rendered the benefits of officially acknowledging them as no longer necessary (Ali, 2007). In this view, it is crucial that policy makers ensure that intra-Muslim dynamics are also given due consideration, and that lived experiences are not mistakenly deemed less relevant because they do not neatly map against the categories that ‘count’.

Although the article discusses findings from a small-scale study the discussion contributes to important reflection on how we view ethnic groups within our cities, by showing how lived identity categories differ from official ethnic categories, as well as by demonstrating the continuation of the political nature of categorisation and its contestation by groups in the post-colonial landscape. As such the article adds to discussions on how we as a society can acknowledge and respect unofficial collective identities, so that they do not become marginalised among their neighbours belonging to official groups and thus ‘legitimate’ stakeholders for government attention and access to resources. In doing so, it adds to existing literature on the use of official categories as political tools both by the state and by communities themselves.

Building upon earlier work by Cohn (1996), Bhagat (2001), and Khan and Qaiser (2022), this article adds to the theory on how the legacy of the census as a political tool for defining social groups in the Indian subcontinent during the colonial period, and now post-colonial period (for example in Pakistan), continues within the diaspora. As such it contributes to our understanding of how contested ethnic categories can be imported with populations as they migrate, and therefore demands for official recognition can often be embedded within wider historical and global context that pre-date and supersede the geographical parameters of such campaigns within Britain or the receiving country.

Secondly, the discussion in this article contributes to our understanding of the perceived temporal usefulness of official categories as a means of recognition and access to power, and in doing so furthers theoretical debates on drivers for the salience of collective identity markers (Nagel, 1994). By exploring whether the Muslim category could mask the needs of other political advocacy groups such as Kashmiri, it provides insight on how categories can lose support or be perceived as less relevant in the face of different competing political positions among the same population. Just as group identities are argued to increase and decrease in the salience groups afford them. This article therefore demonstrates how there needs to be both theoretical and methodological considerations for the meaning attached to official categories and how they can both differ and change for those who they are meant to represent.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by UK Research and Innovation.