Raciolinguistic border-making and the elasticity of assessment and believeability in the UK citizenship process

Introduction

Perhaps two of the most symbolic and significant parts of the British naturalisation process include citizenship testing (Fortier, 2016; Khan, 2019) and citizenship ceremonies (Byrne, 2014; Fortier, 2013). The aim of this paper is to broaden the view of this process further and uncover some of the hidden forms of assessment within the citizenship process and even after the ceremony. I do this by analysing a range of assessment sites which are plotted along the path to citizenship through a raciolinguistic perspective (Rosa and Flores, 2015, 2017); that is to say, a focus on how notions of language and race are constructed. Jacques Derrida’s concept of the Shibboleth (Derrida, 2005) based on a biblical passage from the book of Judges in the Old Testament as a metaphor for language testing (McNamara, 2005, 2012; McNamara and Roever, 2006; Shohamy, 2006).  This reveals how the citizenship process is littered with border making and assessments in which there is an inherent instability in how interpretation of the assessed can create the threat for potential discrimination and exclusion from ultimately gaining citizenship (Derrida, 2005). Throughout the process there is a constant struggle by the applicant to be ‘believed’ (Derrida, 1998) by the state whether through language proficiency, the application process, and the symbolic act of a willingness to belong and to wait the requisite time (Cohen, 2013).

I differentiate between the LUK test (referred to as the Life in the UK test) and the process which refers to the wider application process for which the LUK test is a requirement. The focus of this paper is how and where these sites emerge and what they infer about the national community that migrants are joining. Thus, there are demands for believeability which are manifested away from the formal citizenship test which itself is an act of proving worthiness to belong through the technical knowledge of a language and the symbolic act of demonstrating a willingness to act in accordance to the state’s prescribed conditions (Khan, 2019). Both language and citizenship contain a ‘promise’ and desire to be accepted by the Other, in this case the State (Derrida, 1998; Fortier, 2016, 2018). As such, the ‘believeability’ (Derrida, 1998) through assessment only serves to re-affirm the multiple ways in which the differences between non-native and natives, foreign and British people are reified.

I will demonstrate how territorially externalised, pre-emptive border making extends the national linguistic and legal boundary outside of Britain into other countries through language testing and legislative requirements (Khan, 2019; Khan and McNamara, 2017). I will also elucidate on how the citizenship ceremony is a form of examination (Fortier, 2013; Khan, 2019; Khan and Blackledge, 2015). However, the assessments may not stop at the point of this celebratory moment. There may also be a passport interview in which applicants are called into a local office for an interview with an official in which they must confirm their identity and answer several questions. It is therefore perfectly conceivable that the individual is assessed multiple times, if we also include the citizenship test and application process. Of particular significance are the times and spaces of these assessments: before and away from the UK, as well as after the formal test and in the ceremony or passport office. Thus, of specific significance is the way in which standardized language testing and more everyday language assessment can collude with the ‘force of law’ (Derrida, 1990) to act as borders. This paper also draws on notions of boundary making (see also Winter in this issue) while focusing on being believed in the eyes of the state.

The inside and outside of the borders are dissolved and instead judgment on behalf of the nation is experienced in all manner of places (Andrijsaevic, 2009; Parker et al., 2009). Beyond more ‘classic’ concepts of borders as territorial lines, these borders can be weak or strong and diffuse across society with the potential for defining us and them and thus creating highly racialised classifications (Anderson, 2013; Andrijasevic, 2009). The emergent nature of borders within the naturalisation process reinforces the ‘exclusive/inclusive logic of citizenship’ (Bassel et al., 2018; Monforte et al., 2019; Tyler and Marciniak, 2013).

This paper advances citizenship research by focusing on assessment beyond the LUK test but within the citizenship process. In the first section of this paper, I will first set out the background to the current language testing regime for citizenship in the UK before outlining raciolinguistic perspectives (Rosa and Flores, 2015, 2017) and will then show how Derrida’s metaphor of the Shibboleth can be used to analyse the types of borders created by the language requirements in the LUK test. The paper will analyse bordering processes prior to and after the citizenship test, showing the diffusion of borders across society and beyond the state’s territory. This leads to the conclusion that behind these repeated assessments is a central exigency to be believed. The key contribution of this paper is to focus on the various types of assessments that exist in the citizenship process that are often masked by the LUK test. Conceptually, this paper demonstrates how language testing and assessment can function as border-making both in highly standardized forms such as language testing and more arbitrary forms of judgment (such as the citizenship ceremony).

Citizenship language requirements

The inception of national citizenship in the UK for migrants can be traced back to British Nationality Act in 1981. The Act represents a significant separation from former colonies by re-marking the lines of belonging with national rather than Commonwealth citizenship. It changed the requirements from ius soli to ius sanguinis which meant being able to trace back a lineage within the British Isles; as a result, this was discriminatory towards immigrants from non-white Commonwealth countries (Tyler, 2010). Salman Rushdie (1982) referred to this legislation as ‘expressly designed to deprive black ¹ and Asian Britons of their citizenship rights.’ Nadine El-Enany (2020: 4) explained that the Act ‘raised for the first time the spectre of a post-imperial, territorially defined and circumscribed Britain. It severed a notionally white, geographically distinct Britain from the remainder of its colonies and Commonwealth.’ It was essentially the moment in which Britain defined itself as a nation rather than an empire. In this nation, not only was there a distinction between ‘white’ Britain and its former colonies but also an explicit link to English language proficiency. A language requirement was written into the law but was subsequently rarely used (Crick Commission, 2003). However, it would be the foundation for a subsequent standardized test, namely LUK test which was introduced in 2005.

Discourse and attitudes towards languages tell us much about how speakers and communities of those are considered (Pennycook, 2007). Blommaert (2009: 265) argues that ‘Society imposes hierarchies and value scales on language, and the looking-glass of linguistic practice often provides a magnified image of the workings of powers and the deep structures of inequality in society’. Attitudes and policy measures towards language and its speakers often reflect wider and deeper societal concerns (Blackledge, 2005). In recent years, imposing a dominant language assessment on immigrants through language testing for settlement and citizenship has also become a reflection of the assimilationist turn in policies that reject multiculturalism (see Blackledge, 2005; Extra et al., 2009). Criticisms of migrants speaking their own languages are often encoded discourses lamenting what is perceived as their unwillingness to integrate, and can represent the justificatory grounds for more stringent immigration controls through language testing regimes (Blackledge, 2005, 2006, 2009).

The LUK test was introduced as a policy response to civil disturbances in 2001 (Blackledge, 2005). The introduction of a language requirement and a focus on national citizenship marks a shift to discourse aligned to the integration of migrants while also protecting national security (Khan, 2017). Blackledge (2005) provides a robust and forensic analysis of the discourse which justified these language requirements. Blackledge demonstrates how notions of ‘integration’ around the 2001 were essentially formulations of assimilation – particularly as the riots involved British-Asians (mainly British-Pakistani). Previous forms of multiculturalism as a ‘community of communities’ (Parekh, 2000) had supposedly created individuals who lived ‘parallel lives’ (Cantle, 2002) prior to the riots. Whereas the the question of the languages of migrants was largely left untouched, this would become more problematic post-riots as commonality was the focus in creating ‘community cohesion’ on a national and local level (see Blackledge, 2005). English was a key component of this focus on integration, or more specifically assimilation (Blackledge, 2005, 2009; Cameron, 2013). This is in line with research into other European contexts where language for integration has carried an assimilative tone, particularly through citizenship language testing (see Extra et al., 2009; van Avermaet, 2009).

Raciolinguistics

The political use of language proficiency as problematic among migrants opens up the space to circulate other forms of racism that position migrant as unwilling or unable to integrate

(Blackledge, 2005, 2006). When positioning migrants as deficient due to their lack of English, they are also being positioned as deficient in comparison with British natives which inevitably draws on a racial hierarchy (Blackledge, 2005; Rosa and Flores, 2015, 2017). This deficit is emphasised by language requirements which can also function as a tool to discriminate by making migration requirements more onerous for some than others based on where they were educated (Cooke, 2009). Fortier (2018) likens the promotion of English under the guise of citizenship to 19^th century imperial expansion in which English was exported while maintaining prestige over other languages. While English was spread as the borders of Commonwealth expanded, English becomes a way to exclude as those borders contract to defend the nation. By promoting a monolingual national community, the Government not only maps out borders in defining (non)citizens (Heller and McElhinny, 2017) but also exceptionalises the languages of migrants and migrants themselves as inherently problematic both in terms of social cohesion and security (Blackledge, 2005; Heller and McElhinney, 2017; Khan, 2017). Deborah Cameron (2013) notes how the discursive construction of ‘problematic’ Muslims from the riots threatening the security of the UK is symbolized by dichotomizing a ‘good’ national language against ‘bad’ languages. English becomes the ‘secure language’ (Cameron, 2013; see also Khan, 2017) that constructs an idealized community and provides a linguistic and racial standard against which migrants are oriented (Rosa and Flores, 2015, 2017).

Here a raciolinguistic perspective (Rosa and Flores, 2015, 2017) becomes instructive. According to this perspective, recognising the colonial and imperial roots of contemporary inequalities linked to languages and their speakers. This perspective informs why and how certain racialized languages and their speakers may be recognised as deficient, or in this case as not aligned with the national community. Conversely, languages such as English are perceived as the standard against which speakers of other languages may be assessed. For this dynamic to exist, there is what Rosa and Flores (2015, 2017) term the ‘white listening subject’ who determines what linguistic practices and languages are congruent with the national community. This refers to a more ideological and even institutional positioning of how languages are hierarchized and how the existing linguistic practices of racialized speakers may be a) exceptionalised and in the case of this paper, b) require some form of realignment with the ideals of a national community. These ideals may be enforced through forms of testing and assessment which implicitly threaten exclusion for non-compliance (McNamara, 2012). Each assessment creates a standard against which individuals are measured and should they not meet the standard ‘normalizing’ practices may be required (see Foucault, 1977). In the case of the citizenship process, there are several assessments such as the citizenship test and the application process (Khan, 2019). The question then is what these assessments tell us about the nature of the interaction between the individual and those acting on behalf of the state. If the nature of British citizenship language requirements is highly racialised (Blackledge, 2005, 2006, 2009), a raciolinguistic perspective allows us to identify the nature of the listening subject and the forms of assessment taking place.

The Shibboleth and borders

The Shibboleth has come to signify a metaphor for language testing and language assessment more broadly (McNamara, 2012; McNamara and Roever, 2006; Shohamy, 2006; Spolsky, 2005). The ‘Shibboleth’ refers to a passage in the Old Testament, Judges 12:6. Following an inter-ethnic war between two tribes (the Gileads and Ephraimites), the Ephraimites sought safe passage home. However, unable to distinguish physically between members of both communities, the Gileads asked individuals to pronounce the word ‘Shibboleth’ as they knew of the Ephraimite tendency to say ‘sibboleth.’ The biblical passage is below

They [the Gileadites] would tell him say, ‘Shibboleth.’ If he was from Ephraim, he would say “Sibboleth” because people cannot pronounce the word correctly. They they would take him and kill him at the shallow crossings of the Jordan. In all, 42,000 Ephraimites were killed at the time.

This reordering of political and social relations between the state and the individual can be described as a form of ‘domopolitics’ (Walters, 2004). This is described as ‘a reconfiguring of the relations between citizenship, state and territory. At its heart is a fateful conjunction of home, land and security. It rationalizes a series of security measures in the name of a particular conception of home’ (Walter, 2004: 241). These measures effectively created bordering techniques to govern the ‘home’ to the point of extending beyond territorial boundaries (Walter, 2004). The clear message to migrants in this home was that English is the dominant language and the language of the State despite the multilingualism of its inhabitants. Walter also argues that domopolitics ‘is an attempt to contain citizenship, to uphold a certain statist conception of citizenship in the face of social forces that are tracing out other cultural and political possibilities’ (Walter, 2004: 256).

Borders require surveillance for which Derrida describes the Shibboleth as a ‘watchtower’ (Derrida, 2005). Thus, assessments may become mechanisms for the ‘social sorting’ (Lyon, 2003) which surveillance makes necessary. Surveillance of border crossings may involve individuals who work on behalf of the state in judging the crucial difference ‘shi’ or ‘si’ in the Shibboleth. Derrida states ‘at the invisible border between shi and si, [the Ephraimites] betrayed themselves to the sentinel at the risk of their life … between shi and si; they marked themselves with their inability to re-mark a mark thus coded’ (Derrida, 2005: 22). Following the confrontation with shi and si, the individual becomes ‘coded’ with a linguistic sign which is interpreted by others. ‘e Derrida (1998: 23) states ‘I do not know what will happen to me or what awaits me at the end of the sentence, neither who nor what awaits whom or what, I am within this promise or this threat.’ Given that judgment lies in the hand of the other, the assessed is at the behest of the other from whom they desire recognition. Fortier (2016, 2018) captures the effect of this desire and the anxiety that it instils in applicants within the ‘psychic power’ of policy in everyday life (McNamara and Roever, 2006).

Methodology

This paper methodologically and theoretically engages with the ‘practice turn’ in studies on borders which focuses on interactions with the administrative and representative bodies and individuals on behalf of the state (Côté-Boucher et al., 2014). I draw on 158 in-depth interviews with migrants preparing for the test or having taken the test in two highly diverse cities in England (Leicester and London) as part of a four-year project. ² Each interview lasted between thirty to ninety minutes. While each interview followed the same interview schedule, there was sufficient flexibility to probe areas of interest and to allow participants the opportunity to expand on their answers. One unanticipated area of interest which emerged was the post-citizenship ceremony passport interviews which I will later reference.

The interviews painted a picture of the citizenship journey. Migrants at different stages of the citizenship process were interviewed who, for example, were considering entering the test process; had gone through the process and had passed the test; had gone through the process and failed the test; were in the preparation process through colleges, migrant advocacy organisations, private providers; had just taken the test and the ceremony and passport interview that follow.

The spread of data incorporated different social backgrounds, nationalities, and personal characteristics: participants from 39 nationalities were interviewed and there was a total of 158 (63 men and 95 women), ranging from less than a year-over 20 years in the UK (the average was 9.8 years), and a variety of legal statuses (e.g. UK citizens, EU citizens, Indefinite Leave to Remain, Applying for Indefinite Leave to remain). Participants were primarily accessed through migrant advocacy and community organisations, colleges providing language training and snowball sampling. The interviews were then coded using NVivo. Three researchers cross-coded which permitted robust reliability in the coding frames.

Prior to arrival: Pre-emptive borders through spouse reunification

In one case, a participant had to take a test prior to arrival in the UK in India in order to be reunited with his spouse. One of the major difficulties of language testing proficiency prior to arrival is that by locating the exams in the countries of residence prior to arrival in the UK, those in larger cities are in a more advantageous position as they have easier access to the examinations; thus there is an immediate bias towards those in urban areas. For example, in the case of India there are just 27 approved test centres providing valid exams for immigration to the UK (Home Office, 2016). Of these 27 test centres, 11 are in Bangalore, Chennai, Mumbai and New Delhi (Home Office, 2016). Thus, Derrida’s point on the place of the border is crucial as being in a position to actually sit an exam in an approved centre may be an immense task in itself. To raise the stakes further, failure would mean doubling the time and money involved in a re-sit. This is outlined in an interview below:

if somebody fails the exam they have to wait for another three to four months. So people who might not be that financially well off might have to travel there and pay for the overnight stays, something like that. Travelling. And if they are in employment, losing time, whatever. So British Government is making it even more difficult for people. You know the, what was happening, those people who failed and who got the, who do re-enter for the exam in three months (Leicester interview 39, Male, from India).

The need for strong borders and secure language testing has been increasingly underlined in public discourses which depend on the surveillant aspect of border making and language testing (Harding et al., 2019). In 2019, 35,000 student visas were revoked by the Home Office for test takers of the TOEIC exam due to evidence that there had been cheating in some TOEIC providers (The Guardian, 2019). This led to criticisms about the heavy-handed approach by the Home Office and the effects that this had on those test takers leading to comparisons with the Windrush scandal (The Guardian, 2019). This was due to a BBC documentary which exposed fraudulent practices (BBC, 2019). The exam provider of the TOEIC exam, ETS (Educational Testing Service), was struck off the list of SELT (Secure English Testing) providers (Harding et al., 2019) and the ongoing tender for SELT exams was reviewed. The result was a list of approved providers all affiliated with a SELT Consortium (Home Office, 2016). Harding et al. (2019) chart the history of SELT centres which are closely linked to immigration as the only officially recognized locations for testing and therefore serve to regulate migrant mobility. This is significant as not all test centres are treated the same depending on whether they are approved or not and suspicion emerges around the sites of assessments.

While the distances and locations present one set of problems, there is another: passing the test. Perhaps a more apt way of viewing this is in terms of the potential cost of failure. In this case, the cost of failing is not only a duplication of the costs and distance travelled but also in terms of time. A characteristic of border practices is their temporal dimension (Parker et al., 2009). Cohen (2013) argues that time represents ‘assimilation, civic knowledge, social connection, loyalty and many more citizenly traits and experiences’, and therefore acts as a kind of social value, much like the symbolic value of learning a language, to be recognized in becoming a citizen. Parker et al. (2009: 730) state, ‘Temporally…borders are…ever more pre-emptive, risk-assessed and designed to be as mobile as the subjects and the objects in transit they seek to control.’ By extending a linguistic border through pre-emptive, off-shore testing prior to arrival, migrants must demonstrate an additional willingness and deservingness .

The pre-emptive nature of off-shore language testing not only extends the number of stages in the process but also increases the length these stages will take. For those seeking to be with their spouses, this time becomes a precious commodity. In fact, Human Rights Watch (HRW, 2008) has previously argued that such tests are not only discriminatory as they tend to apply to non-Western families but also may in fact impede integration due to the delays in uniting the family unit.

The same participant continued through his interpreter ‘his concern is that the fact that the British Government has checked his status thoroughly, given him permission to come to this country, then why does he still need to do this citizen test? What is, what is the, what are they trying to achieve? What is the aim of this test?’ Here, a key theme emerges that will be revisited later of checking and re-checking language proficiency, specifically for those from outside of the EU. The participant was about to take an IELTS (International English Language Testing System) exam, commonly used for international students, to enter the UK and demonstrate his language proficiency and then another test, the LUK test, to also demonstrate language proficiency and civic knowledge.

The fusion of these disparate tests within a single pathway to citizenship creates an awkward assemblage within the naturalization process. IELTS was created in 1989 while the language requirement for family reunification was introduced in 2010 (May 2010). Thus, IELTS was never designed with the purpose of spouse reunification in mind by its original developers. This is, therefore, a form of test misuse’ (Fulcher, 2015; Fulcher and Davidson, 2009). The fact that there is such post-hoc use of existing tests demonstrates the potential misuse of tests for political aims (Fulcher, 2015; Fulcher and Davidson, 2009). This raises ethical questions about transplanting general education exams beyond the developers’ intended use as a form of border control especially given the hostility and xenophobic discourse which are currently so prevalent. In fairness, there is now an exam specifically for spouses but this was not the case for this participant.

Earlier in the paper, I highlighted Derrida’s (2005) link between four key dimensions: language, place, border and entrance to a community. In this case, language testing takes places in India as a linguistic and legal externalized British border that must be ‘crossed’ in order to arrive in the UK and to be part of a national community. At what point, does the Government ‘believe’ that immigrants are ‘to be believed’ about their language proficiency? At the early point in this respondent’s time in the UK, there is a sense that this border passage is never complete though he has physically arrived. Language, place, border and entrance to community – and the theme of reassessment – appear at other stages of the citizenship process.

In terms of raciolinguistic perspective, there are two key points. Firstly, this stage applies only to those from outside of the EU and non-English dominant countries. Thus automatically those from countries such as the US, Australia, New Zealand, Canada and European countries already have a less onerous entry. Those most likely to be affected are from Asia, South America and Africa where there are far more racialized populations. Secondly, each trial or assessment outlines an ‘ideal’ (Bakhtin, 1981). In other words, there is an ideal as a standard to measure others. In this respect, the ‘white listening subject’ (Rosa and Flores, 2015, 2017) creates the ideal of an English speaking spouse.

After the life in the UK test: Celebrating citizenship…and beyond…

The LUK citizenship test has been described as a ‘rite of passage’ (Byrne, 2014; Khan, 2019) and represents a symbolic engagement with the state as well as evidence of educational and linguistic proficiency. Once successful, the results along with other attendant documentation, are sent to the Home Office. They then undergo another assessment as a final decision is made. Following this, new citizens are invited to the ceremony. This is the symbolic and official moment they will become British, or so it seems (Byrne, 2014; Fortier, 2016; Khan, 2019).

The following example demonstrates the happiness of becoming a citizen. In this case, the individual is a male from Latin America ³ who had become a citizen and had lived in the UK for nine years (London interview 46). He recounted his experiences of becoming a citizen: ‘It was a little bit exciting and I was really happy that…finally, you know, after all the years of struggling.’ He would later say: ‘For me the minute that I was given that certificate was so important’ and concluded ‘It was so valuable, you know, that it just feels something that you have achieved in your life, you know. That all you can say is thank God for it, you know.’

By all accounts this is a satisfactory conclusion. However, not everyone experienced the citizenship ceremony in the same way. The following account is a Latin American woman who had lived in the UK for 12 years (London Interview 55). She initially stated that she was ‘nervous’ about the ceremony. This sense of nerves is surprising given that the respondent had cleared all the hurdles to become a citizen. When probed as to why, she explained:

Because we had, we did the oath and it didn’t come out well. What was easier was the declaration short [laughs] I am Christian and [unclear] jokes, the ceremony was shorter and I did the declaration like I don’t know how long, I didn’t study to prepare but [unclear] I don’t have many problems learning so it didn’t have a bad intention we passed along the front and they asked for my oath and [unclear] and in English because it wasn’t my native tongue, I was qualifying at last. This was scary to say the declaration (London interview 55).

After the ceremony celebration, citizens can apply for a British passport. This is material recognition of one’s arrival point at the ‘end’ of the naturalization process. However, for some there still remains yet another hurdle: the passport interview. As explained by a participant, ⁴ she did not receive a passport straightaway; instead she was invited for an interview in what she describes as ‘in the Home Office.’ The implicit threat of withholding the passport remains as the border has not been fully negotiated. Therefore the respondent had to attend an interview.

The respondent acknowledges that this is ‘another exam.’ She is happy to comply but has a clear understanding of the context of the evaluative interaction with her interviewer. It is also significant that the respondent views the ‘exam’ within a wider security apparatus and points to understanding of prevalent political discourses and policies which condition interactions between the individual and the state. The same respondent also played a key part in the application of her son. This experience of accompanying him to his interview was a difficult one as she explains in the interview:

The official, the son and the mother are all made aware that the context of the interaction is set within exam conditions. By understanding these conditions, we see how the official is more than an interviewer and is in fact an examiner. As with the spouse visa example, the mother has no choice but to comply. She mentions ‘I understand it’s logical, there are many strange things.’ This interview is not technically part of the process as it applies to all applying for a first British passport; however, it is significant that it is perceived as being so. Thus, the interview is experienced as a border even if it is not designed to be so. This interaction above is converted into an examination since the mother is not allowed to enter the interaction, much in the same way that an oral exam would not include someone not being assessed. The extra challenge in making oneself understood is central to the Shibboleth since the principal objective of communication is predicated on being understood and recognised by the Other (McNamara, 2012).

The raciolinguistic perspective tells us much about the ‘white listening subject’ (Rosa and Flores, 2015, 2017). Firstly, British native citizens never actually make an oath or allegiance, it is instead assumed. Thus, there is a suspicion about migrant until proven otherwise. However, more than this there remains suspicion even when providing the performance during the ceremony. McNamara (2012) notes the ‘vulnerability’ of the individual during assessment as the outcome is no longer in their hands. In my related study (2019), a participant explained how some people would mimic movements with their lips as they were unable to recite during the ceremony. The question is here is round the measures required to be believed. Secondly, the passport interview reminds us that even after the celebration the ‘white listening subject’ may intervene. It is worth remembering that Home Office correspondence may be sent to the postal address, thus the citizen may already have proven where they live. The application process may confirm their identity while the LUK test may provide evidence of their language proficiency.

Discussion: Stepping into the law and Believeability

The establishment of testing regimes is essential to border making practices through language. These regimes possess a raciolinguistic dimension due to the forms of assessment which ensure that not only is language proficiency judged . The LUK test was created as an embodiment of legislation aimed at making citizenship more onerous (Blackledge, 2006; Khan, 2019) but for this paper the focus has been on the forms of assessment that take place before and after the LUK test. These forms of border making are distributed unequally by virtue of not applying to all in some cases. This policing also appears through various bodies whether SELT providers, ceremony officials or passport interviews.

Backed by ‘the Force of Law’ (Derrida, 1990), borders can be externalised (as with the spouse route) as pre-emptive frontiers to ensure that a particular type of linguistic profile which is implicitly racial can be privileged while others are devalued (Heller and McElhinny, 2017). This permits keeping some immigrants, particularly from non-EU and non-English speaking countries, as far away as possible for as long as possible by regulating the time and distance to be negotiated. These types of borders mean that the exam process is extended - both before and after the test itself – and the path to citizenship becomes highly attritional while also requiring extra surveillance of their language proficiency and linguistic practices.

Borders can also be integrated within the daily routines and tasks of officials working on behalf of the state as with the ceremonies and passport interviews. Quite often the imposition of a dominant language is perpetuated through mundane, administrative routines (Derrida, 1998) or even as someone whose language proficiency is questioned by ‘native’ speakers. This form of crude assessment is known as ‘indigenous assessment’ (McNamara, 2005; McNamara and Roever, 2006) in which rather than a standardized test, an individual, usually a native speaker, can make instantaneous assessments on a non-native speaker. This may also be transferred in more innocuous ways such as asking where people may be from based on their accents even if they are now British citizens.

It is worth pointing out that through assessment regimes, as is evident in this paper, the individual is essentially at a starting point that is ‘outside of the law.’ Derrida outlines this point of departure from vulnerability

it [Shibboleth/language] becomes what one must know how to recognize and above all to mark if one is to make the step, to step across the border of a place or the threshold of a poem, to see oneself granted the right of asylum or the legitimate habitation of a language. So as no longer to be outside of the law (Derrida, 2005: 26 - original emphasis).

To make the ‘step,’ the individual must make their language proficiency visible to the state, whether through their application, an exam or in their interactions with an official. This form of dispersed assessment and governance (Fortier, 2016, 2018) instils anxiety in the process of ‘earning’ citizenship (Fortier, 2016). This is all a far cry from the objectivity of standardized exams as impartial instruments (Shohamy, 2001, 2006). Although the respondents in this paper managed to navigate their way through their various assessments successfully, the spectre of potential loss and failure always remained present.

Through language and citizenship, the individual always desires recognition from the Other (Derrida, 1998: McNamara, 2012, 2017). In order to belong and be a citizen, the individual negotiates an imposition that promises equality and ‘safe passage’ (Derrida, 2005) to rights; the Other must accept the individual. Thus, the individual is acutely vulnerable due to the inherent fragility in how what is said is interpreted by others. The interpretation of our intended meanings may work against us when it is misconstrued or judged to be insufficient (McNamara, 2012). This article has shown how these promises and desires to be recognised exist throughout the citizenship process from before arrival to after gaining the passport. Not everyone experiences these equally, which demonstrates how arbitrary and potentially discriminative these assessments can be but the ‘white listening subject’ remains an evaluative presence with the power to exclude or create the conditions for vulnerability for those being judged.

In order to enforce the promise, the individual must be believed. This is explained by Derrida as follows: ‘To what can, at any rate, only be believed; to what appeals only to belief and hence to the given word, since it lies beyond the limits of proof, indication, certified acknowledgement [le constat], and knowledge’ (Derrida, 1998: 20). Each interaction of border making practices based on language provides evidence that the individual is to be believed in order to be a citizen. This can be satisfied at the moment of the jjudgment but can be also reconstituted later even to the point where the elasticity of the citizenship process is stretched to include other countries or beyond the ceremony as we have seen.

Believeability cradles the entire citizenship process. On the one hand, there is the linguistic believebility of level of language proficiency that is set arbitrarily (McNamara and Roever, 2006), which can be raised or extended to other countries. On the other hand, believeability takes a symbolic dimension in that there is a constant struggle by the individual to be believed about their identity and the values associated with a willingness to undergo the citizenship process. The examination and application are not enough. Individuals must come face to face with the ‘white listening subject’ on behalf of the state to be believed and prove their capacity and willingness to be judged in order to belong.

Conclusion

This research has shown that the length of the citizenship process depends on what is believed (and therefore not believed) via extended borders and multiple border making practices. Some may never find themselves to be vulnerable between shi and si in uttering Shibboleth while for others, saying the Shibboleth once is never enough nor is uttering it again enough in order to be believed. The ‘white listening subject’ is capable of determining how many assessments are required and when, if at all in some cases. With each assessment, the individual becomes ‘undecidable’ (McNamara, 2012) and find themselves making the ‘step’ between shi and si which leaves them susceptible to harm; for some more than others due to their linguistic proficiency.

There is also the potential for test misuse. Post-hoc forms of employing tests built for one purpose and then used for another, provide ethical concerns and open up criticism around test misuse for other purposes (Bruzos et al., 2018; Fulcher, 2015) which also highlights their political function (McNamara and Roever, 2006; McNamara and Ryan, 2011). That pre-emptive border making through testing is required for non-EU, non-English speaking spouses demonstrates its highly racialized dimension. It is not only the profile of the global population that is privileged/disadvantaged, but also the access to simply be able to sit an exam. The immense time and effort required to access exams means that testing provides an externalised, pre-emptive and racialized border. Of course, all of this is prior to arrival in the UK when the citizenship process continues.

This effort to be believed must be understood in a wider socio-political context. The Windrush scandal, which has seen settled residents sent back to the Caribbean, the incessant portrayal of British Muslims as problematic to British society as well as the open post-Brexit hostility towards migrants have exacerbated the tensions as to who is truly believed to be a legitimate part of the national community and who is not. The constant rechecking of new citizens in becoming a part of the national community raises questions as to the extent to which individuals are believed to belong. This paper demonstrates one dimension of how ‘believeability’ infuses assessments in the citizenship process and this is one part of a wider picture.