‘Just eating and sleeping’: asylum seekers’ constructions of belonging within a restrictive policy environment

ABSTRACT The ‘refugee crisis’ in Europe has drawn attention to the reasons why people risk desperate journeys to seek safety. However, less research has focussed on what happens to those on the move once they have reached their destination country. In recent years the UK government's ‘hostile environment’ policy for asylum seekers has taken precedence over attempts to integrate refugees, creating a system in which destitution, dispersal and detention have all become pervasive features. This paper takes a discursive psychological approach to the analysis of interviews with asylum seekers in Wales, UK. It argues that participants draw on economic repertoires of effortfulness to construct accounts in which belonging is dependent upon being able to contribute to the economic and civic life of the host society. It further highlights how participants construct accounts in which restriction from the asylum system is positioned as the reason for not belonging and that time spent as an asylum seeker is policy-imposed liminality. The findings suggest that allowing asylum seekers to work would be a key step forward in integration policy and contribute to generating a greater sense of belonging.


Introduction
In recent years the Mediterranean 'refugee crisis' has drawn attention to the increasing movement of people from countries in the Middle East and Africa, fleeing war and persecution, who have arrived in Europe. At the height of the so-called 'crisis' in 2015 over 1.2 million applications for asylum were received in EU countries (Eurostat, 2016). As such, the 'refugee crisis' has led to an increase in research by academics in this area. Whilst this has principally been concerned with media coverage of the crisis (e.g. Goodman, Sirriyeh, & McMahon, 2017) or on the drivers of migration (e.g. Crawley, Duvell, Jones, McMahon, & Sigona, 2018), some research has also focussed on what happens to those on the move once they have arrived in their destination country (e.g. Bennett, 2018). This paper offers a novel contribution to the literature on the ways in which belonging is discursively constructed by refugees and asylum seekers in the UK.
In this paper, I use the term 'asylum seeker' to refer to those who have 'crossed an international border in search of protection, but whose claim for refugee status has not yet been decided' (Castles, De Haas, & Miller, 2014, p. 222). I use the term 'refugee' for those who have been recognised by a national government as meeting the requirements of the 1951 United Nations Convention relating to the Status of Refugees. However, it should also be noted that there is often conceptual slippage in relation to such categories particularly through media usage whilst migration scholars have also problematised this distinction, arguing that it creates opposing binaries and may not represent real-world migration patterns where an individual may change status or belong to more than one category (Crawley & Skleparis, 2018). In a policy context, this distinction is important, particularly in the UK case, because those recognised as refugees by the UK government receive similar social and legal rights to those of UK residents, whereas asylum seekers do not. Indeed, the UK government adopt a similar approach in relation to 'integration' policy where it is only those with the ability to plan for a long-term future in the UK (i.e. those awarded refugee status) who are encouraged to 'integrate'. Such an approach fails to recognise that 'integration' experiences may occur whilst waiting for a decision on an asylum application but also draws attention to the ways in which 'integration' is a contested and 'chaotic' concept (Robinson, 1998, p. 118).
This paper begins by describing the current 'hostile environment' policy (Bloch & Schuster, 2005) for asylum seekers to situate this paper in context. In this paper, I take a discursive psychological approach (Potter & Wetherell, 1987) to the analysis of interviews with asylum seekers in the UK. Thus, this paper will review recent research from this approach and how it has been applied to the analysis of talk by refugees and asylum seekers. Following a discussion of the methods used in this study, I present a discursive psychological analysis of interview extracts which demonstrate the ways in which participants constructed the UK asylum system as restrictive. I also discuss the implications of this for longer term feelings of belonging and integration in the UK.
1.1. The 'hostile environment' for asylum seekers in the UK When the Immigration and Asylum Act was introduced in 1999 it marked the first significant changes to UK asylum and immigration policy since the early 1980s. It began a process of increasingly hostile measures targeting asylum seekers that became a focus of New Labour policy. Indeed, reducing asylum applications was the favoured response of the then prime minister Tony Blair to the growing concern amongst the British public regarding the increasing number of asylum seekers coming to the UK in the late 1990s and early twenty-first century (Somerville, 2007). This was also a period in which the issue of asylum became increasingly politicised and mediatised, particularly in relation to the Sangatte refugee camp, which opened in France in 1999, and public perception in the South East that they were experiencing an 'unfair burden' from asylum seekers (Bennett, 2018). The 1999 Immigration and Asylum Act, which created a 'hostile environment' for asylum seekers, included the triple threat of enforced destitution, dispersal away from London and the South East and the threat of indefinite detention (Bloch & Schuster, 2005). This 'hostile environment' therefore aimed to deter asylum seekers from coming to the UK but also to encourage refused asylum seekers (those not granted refugee status or another form of protection) to leave the UK.
One of the key elements of the 1999 Act was the creation of the National Asylum Support Service (NASS). The role of NASS was to co-ordinate the accommodation and financial support for asylum seekers, and thus marked a move away from access to general welfare benefits and housing, to a separate system of welfare for asylum seekers only. As part of this reform, one of the key measures required asylum seekers, who needed accommodation and subsistence support, to agree to compulsory 'dispersal' to cluster areas outside London and the South East. A wide range of research has focussed on the impacts of asylum dispersal in the UK, particularly on the decision to disperse asylum seekers to areas of deprivation and social exclusion (Zetter & Pearl, 2000). In addition, it has been found that many of the dispersal areas lack expertise and Refugee Community Organisations to provide support to vulnerable asylum-seeking populations (Zetter, Griffiths, & Sigona, 2005). However, a more prominent feature of the literature in this area has been the finding that the housing provided to asylum seekers since the creation of NASS has often been of poor quality as a result of difficult to let housing stock being offered (Garvie, 2001Phillips, 2006. The 1999 Act also saw restrictions on financial support for asylum seekers introduced, which have contributed to a system of enforced destitution for asylum seekers (Allsopp, Sigona, & Phillimore, 2014). Initially, financial support was provided as a single payment, at approximately 70% of income support, which was given in the form of vouchers, rather than cash. These vouchers, scrapped in 2001 following complaints by charities supporting asylum seekers, were only accepted at certain retailers, placing a restriction on, not only where, but also what they were able to buy. However, asylum seekers currently receive only £37.75 per week, significantly less than the rate of 70% of income support that was received by asylum claimants in 1999. This is the result of successive governments freezing increases in the support provided until a small rise in 2018. Asylum seekers whose applications are subsequently refused, and who are unable to return to their home country, are able to claim support of £35.59 per week under Section 4 of the 1999 Act. However, this is provided on a payment card that functions in the same way as the previous voucher system, with no cash support being given.
Destitution was further enforced under Section 55 of the 2002 Nationality, Immigration and Asylum Act, which excluded asylum seekers from support if they had not made their claim 'as soon as reasonably practical' after arriving in the UK (Allsopp et al., 2014). Significantly, this Act led to the removal of the right to work for asylum seekers and thus further increased destitution. However, in cases where an asylum seeker had been waiting over 12 months for a decision on their claim, they could apply to work, but only if they could undertake a job on the shortage occupation list. At the time of writing, this list included highly skilled engineering and science roles, but also classically trained ballet dancers (Gov.uk, 2018), meaning that for the majority of asylum seekers, who have waited for more than a year for a decision on their claim, destitution is enforced through reliance on NASS support.
In recent years, restrictions on asylum seekers access to further and higher education have also been implemented. Asylum seekers are able to attend UK universities but are classed as overseas students for fee purposes and, when receiving only £37.75 per week, are unlikely to be able to afford over £10,000 per year in fees. Access to further education courses has also been restricted in England (though not in Wales and Scotland) with a requirement for asylum seekers to have been waiting for a decision on their claim for more than six months before attending a further education institution. Such restrictions in accessing education further add to the 'hostile environment' (Bloch & Schuster, 2005) of dispersal, destitution and detention that have been a feature of the UK asylum system for many years, and which form the context of this paper.
At the same time, the Westminster government has also attempted to 'integrate' refugees only and not asylum seekers, based on the assumption that integration cannot begin in its fullest sense until refugee status is granted and a long-term future in the UK can be planned for. Such an approach tends to assume that the time spent waiting for an asylum decision does not impact upon a forced migrant's sense of belonging or their later integration prospects. Hynes (2011, p. 94) talks of this being a 'policy-imposed liminality', in which asylum seekers are left in limbo as a marginalised outsider who 'has recently left 'there', but who is not yet allowed to be fully 'here' (Lewis, 2007, p. 103). However, few research studies have focused on accounts of experiences of liminality.

Discursive psychological research on asylum seekers and refugees
This paper takes a discursive psychological approach, which treats language as a form of social action, to analyse the ways in which asylum seekers living in Wales spoke about living within such a 'hostile environment'. Discursive psychologists, such as Wetherell and Potter (1992), became critical of cognitive social psychology approaches, arguing that they could not account for the variation they found in their interview participants' talk. Therefore, discursive psychology is concerned with people's practices; in particular what they are doing with their discourses in terms of argument, communication and interaction in specific, situated, settings. Edwards and Potter (1992, p. 158) summarise this epistemological shift from empiricism to constructionism in stating that 'the study of situated discourse redefines and relocates the relation between language and understandings, and it does this by placing language as representation (whether of cognition or of reality) in a position subordinate to language as action'. In the context of the current paper, participants' talk about restrictions they faced as a result of the UK asylum system can be seen as 'discursive accomplishments' (Goodman, Burke, Liebling, & Zasada, 2015) and, as such, the discursive psychological approach is particularly relevant to understanding this talk.
Discursive Psychology has been applied to asylum seeking in three principal ways; media and political constructions of asylum seekers (e.g. Augoustinos & Every, 2007;Leudar, Hayes, Nekvapil, & Baker, 2008;Parker, 2015), lay discourses about asylum seeking (e.g. Goodman & Burke, 2010;Goodman & Speer, 2007) and more recently to the discourses of asylum seekers themselves (Goodman, Burke, Liebling, & Zasada, 2014;Kirkwood, McKinlay, & McVittie, 2013) which I review in this section. Whilst much of the research in the area of discursive psychology has focused on how 'others' are discursively constructed (e.g. Goodman & Burke, 2010), Kirkwood, Liu, and Weatherall (2005) have argued that it is equally important to study how out-groups themselves construct their accounts of their own experiences. The present paper seeks to contribute towards this aim. Leudar et al. (2008) conducted a three-part study involving a discursive analysis of newspaper reports which formed the basis for subsequent qualitative interviews with both local residents of Manchester and refugees. In their discursive analysis, they indicate a number of 'hostility themes' (Leudar et al., 2008, p. 187) present in print media constructions of refugees: that refugees are an economic drain, unable to care for their children and are potentially criminals or carriers of disease. Following analysis of interviews with refugees in the same area Leudar et al. (2008) conclude that their participants had internalised these 'hostility themes'. This meant that they had constructed their own identities around those present in the media and those expressed towards them by the local residents they had met. Leudar et al. (2008) found that refugees commonly positioned themselves as being in a worse position than in their home country but that their country of origin was unliveable for them. Kirkwood (2012) focuses his research on how refugees and asylum seekers in Glasgow negotiate dilemmas in talk about their experiences and views of claiming asylum and living in the UK. The concept of the 'ideological dilemma' was proposed by Billig et al. (1988) who felt that studies of human thinking within cognitive social psychology concentrate too closely on the individual; in how information is processed and inferences drawn. Billig et al. (1988) stress the importance of distinguishing between 'lived ideology' (society's way of life) and 'intellectual ideology' (systems of philosophical, religious or political thought) and suggest that dilemmas may arise between the two. In order to achieve a range of social actions such as justifying their position in the UK, Kirkwood (2012) found that interviewees had to manage dilemmas sensitively. He draws particular attention to the ways in which interviewees would often state that they had no difficulties in the UK yet describe in some detail particular problems they faced at later stages in the interview. He highlights talk about discrimination functioning in a similar way, with respondents playing down incidents of discrimination in order to justify their current position in the UK and to highlight the reality of the persecution they faced in their country of origin. He suggests that a 'delicate' strategy had to be employed, with speakers using strategies such as attributing the violence to the ignorance of the attacker, so as not to appear ' … overly sensitive to racism' or be seen to be making complaints (Kirkwood et al., 2013, p. 758). In both cases, then, attention is paid to presenting themselves as reasonable and of not wanting to criticise the host country in which they seek sanctuary.
Similarly, Goodman et al. (2014) report that although their participants said that they were unhappy in the UK and criticised the UK asylum system, this represented a dilemma for them that risked them appearing ungrateful and undermining the reason they were claiming asylum in the UK. Goodman et al. therefore report a downplaying of not being happy as a means of resolving this dilemma. Criticisms of the asylum system were constructed in a number of ways including presenting it as unfair; treating different groups unequally; and as being designed to ensure that refugees are denied asylum in the UK. These constructions are all noteworthy because they are all directed at the asylum system itself rather than the British public. This is similar to Goodman and Speer's (2007) finding that opponents of the asylum system construct their accounts in such a way as to blame the system, rather than the individual asylum seekers, as a strategy for not appearing to be racist.
One of the 'hostility themes' identified in Leudar et al.'s (2008) study of print media discourses was the construction of asylum seekers and refugees being unable to provide sufficient care for their children. Clare, Goodman, Liebling, and Laing (2014) investigate this further by conducting semi-structured interviews with African asylum seeking women. Through a discursive analysis of their interviews they identify two predominant repertoires which their participants drew upon when talking about their emotions; 'rejecting pity' and 'being strong'. Clare et al. suggest that use of these repertoires function to promote positive social identities for the women as responsible parents. Similarly, Goodman et al. (2015) show that when asylum seeking participants talk about returning to their home country, this is accomplished by drawing on notions of safety and managing their identity through positioning themselves as being in need of support and as 'genuine refugees'. Here again, it would appear that there is evidence of a relationship between how the mass media and others talk about asylum seekers and how asylum seekers themselves construct their own identities in talk. Indeed it may be that the interviewees in these studies were managing 'ideological dilemmas' and issues of self-presentation. The purpose of the current paper is to extend existing literature in discursive psychology and asylum seeking through an analysis of asylum seekers' talk about belonging within a restrictive policy environment.

Methods
The data analysed in this paper come from a wider study into refugee integration in Wales. For this, 19 individual semi-structured interviews were conducted with refugees and asylum seekers who were living in Wales at the time of the interview. The use of interviews in discursive psychological research has long been debated, with several researchers preferring to use naturally occurring or data 'produced entirely independently of the actions of the researcher' (Potter, 1996, p. 148). However, others, such as Speer (2002, p. 521) have argued that this dichotomy is not clear and that this preference for naturally occurring data has become a 'catch all term with fuzzy boundaries'. I argue that the use of interviews is appropriate as a research method because the 'voice' of forced migrants themselves has often been missing from discursive studies of forced migration in the UK, with a top-down approach of analysis of political, media and majority group discourses dominating the research field in recent years (Perkins, 2016). Indeed, Goodman et al. (2014) argue further that because forced migrants' voices are absent in public discourses relating to migration, analysis of naturally occurring data would only serve to reproduce their voiceless status.
Participants were recruited from three refugee and asylum seeker support organisations in Cardiff and Swansea. In total, 11 participants were male and 8 female, ranging in age from 19 to 58, with an average age of 34. Participants had been living in the UK for between 1 month and 12 years at the time of interview, with an average time in the UK of 40 months. Four of the participants were asylum seekers who had made an initial application for protection to the UK Government; seven were refused asylum seekers who were appealing the decision at the time of interview; seven participants had been recognised as refugees and granted five years leave to remain in the UK; and, one participant had been granted British Citizenship. Participants were from 13 different countries of origin: Sudan, Iraq, Pakistan, Syria, Kenya, Iran, Ethiopia, Eritrea, Nigeria, Sierra Leone, Egypt, Uganda and Chechnya.
Each of the interviews were conducted in English except for one that was conducted partly with the aid of a translator. Interviews were audio-recorded and transcribed using a simplified version of the conventions outlined by Jefferson (2004see Appendix). Interviews lasted for between 18 and 62 min, with an average length of 32 min. A semi-structured interview guide was devised based on the ten domains identified in Ager and Strang's Indicators of Integration Framework (2004). This influential framework was developed to inform government refugee integration strategies following a comprehensive literature review and fieldwork with refugees in two locations (Glasgow and London). Therefore the interview guide included questions about: education, housing, health, employment, social bonds, social bridges, social links, safety and stability, language and citizenship (Ager & Strang, 2004). Whilst refugees are the focus of the framework, Ager and Strang (2004, p. 8) note that it is also applicable to the integration of other groups, such as 'asylum seekers and economic migrants'. Indeed, Bennett (2018) similarly used the ten domains of the framework to design the interview schedule for his research with refugees and European Union migrants.
After a thorough reading and re-reading of the transcripts, each of the interviews were initially coded using NVivo in order to identify sections in which complaints about restriction were raised by either the participants or the researcher. This resulted in extracts from 15 of the interviews being selected for further detailed analysis. Four of the participants therefore did not report restrictions to their daily life within the course of the interviews. Analysis of the 15 selected extracts revealed that four participants constructed accounts in which their lives were completely restricted by asylum policy in the UK. Each of the four extracts presented in the sections which follow came from interviews with participants who, at the time of interview, were asylum seekers awaiting initial decisions on their asylum case or appealing a refusal of asylum. As this is discursive psychological research the focus of analysis was on action, rather than cognition, and in particular I follow the repertoires approach of Potter and Wetherell (1987). Wetherell and Potter (1992, p. 90) use the term 'interpretative repertoires' to mean 'broadly discernible clusters of terms, descriptions and figures of speech often assembled around metaphors or vivid images'. As such, analysis of the interview data involved identifying the social and linguistic resources which participants drew on in the course of the interviews when discussing restrictions of the UK asylum system and the functions that these had in the interaction.

Analysis
In this analysis, I look at the ways in which four participants who were still waiting for a decision on their asylum claim constructed accounts of belonging in the UK and Wales. Here, I follow Jones and Krzyżanowski's (2008, p. 39) approach of identifying how participants position themselves in relation to collectives of both their 'original' community and the host society. I argue that participants criticise the UK asylum system, raise ontological questions about what it means to be a forced migrant in Wales and position themselves as not belonging because of the UK's asylum policies. However, at the same time I show that participants manage their stake and interest when making these complaints so that they do not come across as bitter or appear ungrateful for the protection they have received in the UK.

'Just eating and sleeping'
Extract 1: Munir we have the right I think to go to::: college (.) 21 to college (.) to learn some English In Extract 1, above, Munir is discussing his experience of being an asylum seeker in Wales following the interviewer's (Sam) question in lines 1-3. In lines 4-8, Munir uses a series of extreme case formulations (ECF) (Pomerantz, 1986) to begin to construct an account in which the UK asylum system restricts him from doing 'anything'. However, he does not make reference to the asylum system itself and, rather, uses the liminal phrase 'just waiting'. Weltman (2003, p. 369) suggests that use of the word 'just' may allow two different orders of concern to be active simultaneously 'the personal/interpersonal and the ideological-historical'. Here, Munir is using 'just' to construct his own sense of self whilst also drawing on broader ideological notions of the UK's asylum system as being restrictive. 'Just' is used again at the start of line 8, but this time as a repair to his previous criticism. Here he is beginning to construct a specific account of the importance of work (which he is not allowed to do), rather than to de-value the role that voluntary work plays. Indeed, in lines 8 and 9, Munir positions himself as 'happy' for having the opportunity to be able to volunteer, allowing him to do criticism without coming across as bitter or invested. Using such emotion discourse (Edwards, 1999) to do positive assessments may function to counter arguments that he is either ungrateful or critical. He describes this as 'feeling that you're working', implying that it is not actually working. In lines 12 and 13 he switches to using the first person plural 'we', for 'we' as asylum seekers, and uses two ECFs to construct a sense of self as being 'nothing' or 'not exist(ing)' due to being economically inactive. This works to construct an account in which a sense of belonging (and 'existing') are contingent upon being able to work and thus being economically productive, something which asylum seekers are currently restricted from doing.
Munir uses two further ECFs in lines 15 and 16, 'just eating and sleeping' and 'that's all' to construct a liminal account and suggest that he does not feel a sense of belonging because of the restrictions that the asylum system places on him and an implied absence of other meaningful activities. However, his use of a repair in line 16, to show what he is able to do, functions as a complaint mitigator so that his previous utterances appear less critical.
In this extract, Munir draws on an interpretative repertoire (Potter & Wetherell, 1987), of the economic value of work, as a common-sense way of interpreting the situation he is in. Gibson (2009) has similarly shown how such an 'effortfulness' repertoire was found in comments posted by users of the BBC website in comments about welfare benefits. He states that the ' repertoire was also central to the management of the accountability of the individual poster, and to holding the government or welfare system to account for failing to hold individual 'lazy' people to account' (Gibson, 2009, p. 405). Similarly, Bennett (2018, p. 39) draws on Aihwa Ong's (2005) work on neo-liberalism and citizenship, arguing that 'those who are unable to fully participate in the market are reduced to the level of second-class citizens and that those who 'can't climb the ladder or self-govern' are marginalised as deviant or risky'. He suggests that because asylum seekers are reliant on government assistance they are more likely to have to justify success in terms of a 'can-do attitude' (Bennett, 2018, p. 39). Thus, in this extract, by using economic and effortfulness repertoires Munir is both criticising the asylum system and denying that he is 'lazy'.
The economic repertoire drawn on by Munir is also used in Extract 2 from an interview with Aziz. Like Munir, Aziz, at the time of interview, was an asylum seeker awaiting a decision on the outcome of his asylum claim and had been living in Wales for only one month.
Extract 2: Aziz In addition to use of an economic repertoire, Aziz (l.15-16) also draws on the discourse used by Munir to describe life in the UK as 'just eating and sleeping', reflecting the government imposed economic inactivity that he faces due to current asylum policy and revealing a lack of meaning without economic productivity. In line 4, Aziz begins by constructing an account that is based on responsibilities, here as a father, but later on, as a member of the local community. In lines, 6-10, Aziz appears to be negotiating an ideological dilemma (Billig et al., 1988) between his status as an asylum seeker, which affords him minimal rights, and the responsibilities he has towards his family and the wider community. This dilemma plays out further in lines 14-16 as Aziz acknowledges that he has the right as an asylum seeker to undertake voluntary work but also positions himself as having a moral responsibility to do this, because of the support that he receives from the government. Here, Aziz constructs an account that shows that he lacks individual agency to undertake the activities that he feels he has the moral responsibility to do. However, in lines 20-26 he develops this further as he moves on to construct an account, which can be read as a more general criticism of asylum policy, and one in which all asylum seekers lack agency. This begins in line 21 with the use of 'someone', which contrasts to the first person account given until this point, and which continues with a third person pronoun ('he') and the first person plural pronoun ('us'). The use of terms such as 'efficiency', 'effectiveness' and 'resource' further allude to use of an economic repertoire as Aziz justifies his criticism of the policy to not allow asylum seekers to work. However, he also draws on an 'ability' repertoire to further show the restrictions faced by asylum seekers, who he argues have skills that could be used by the government who are providing them with subsistence and accommodation support. In lines 24-26 he justifies this view on two grounds. Firstly, in terms of giving back to the community, he suggests that current policy leads asylum seekers to 'just hiding somewhere' indicating a separation from the host population as a result of being unable to work and contribute to economic and civil society. Secondly, he uses the economic repertoire, to further suggest that current policy may lead asylum seekers to engage in 'kind of criminal issues' (l.26). At the end of the extract he returns to justify his position as being based on responsibilities to the local community by using the imperative 'I have to do something' (l.34) and the modal verb 'should' (l.39) to strengthen this position which also works to construct an account of liminality.
Through drawing on the economic repertoire in this way, Aziz is able to justify his view that asylum seekers should be able to work, and, more importantly, that such economic activity fosters a greater sense of belonging within the community. Here then, there is a dilemma being played out between rights and responsibilities and Aziz constructs voluntary work as both a right and a responsibility for him, which functions to protect himself from criticisms of being 'lazy'.

'I can't do nothing'
Extract 3, from an interview with Bhaija, a refused asylum seeker who had been living in Wales for one year at the time of interview, begins with Sam asking Bhaija about how being an asylum seeker makes her feel.
Extract 3: Bhaija In this extract, we see an ideological dilemma (Billig et al., 1988) created between safety and restriction. In line 2, Bhaija begins her turn using the discourse marker 'just', with the ECF 'a helpless situation', which starts to construct her account of facing restriction. Indeed, it is the situation which she describes, rather than her feelings, which was the initial question asked by Sam. She then draws upon a safety repertoire in lines 3-5, using a number of terms such as 'seeking peace' (l.3) and 'secure place ' (l.4-5) that are associated with the category 'safety' and function as an implied rhetorical contrast to 'home'. Throughout her first turn, Bhaija uses the second person pronoun 'you' to position herself as talking about all asylum seekers which contrasts with 'they' and 'their' which she uses to talk about the UK Home Office. In lines 6 and 7, she uses the term 'their rules' on three occasions, drawing on the restriction repertoire. This also points to a lack of agency which she feels she has in her current situation and is highlighted through use of the phrase 'implementing their rules on you' (l.7).
Restriction is further shown through the use of a 3-part list in lines 8-10 as Bhaija gives examples of the requirements placed on asylum seekers through current asylum policy. Here she also uses the modal verb 'have to' in order to express the obligation that asylum seekers may feel they have to comply with in order to receive the support they require. However, in lines 12 and 13 she switches to drawing upon the safety repertoire with use of the phrases 'seek peace' and 'secure life' which further consolidates the dilemma she faces as an asylum seeker between needing to be in a safe environment and the restrictions which asylum policy also places on this. Indeed, this dilemma is stated most clearly by Bhaija in lines 21-24 when she says 'not an independent life it's just a life where you are alive'. The phrase 'just a life where you are alive', as seen in earlier extracts suggests a sense of nothingness and liminality, that Bhaija does not have a sense of belonging in Wales and that the restrictions of current asylum policy may be responsible. However, it is also interesting to note how she draws on the ideas of rights and responsibilities in the same way as Munir and Aziz did in the previous extracts, again suggesting that belonging for Bhaija is contingent upon having rights that allow her to live her day-to-day life in a way that is safe and free of restrictions.
Extract 4: Ghirmay A similar construction is evident in Extract 4, above, from an interview with Ghirmay, a refused asylum seeker who had been living in Wales for three years at the time of interview. In line 9 he uses the phrase 'I am like a moving dead' which can be read similarly to Bhaija's use of 'just a life where you are alive' to emphasise that an ontological condition of not belonging and liminality comes from restrictions of the asylum system and an associated lack of rights. His account consists of a series of ECFs in lines 5, 7, 10 and 11 that are combined with lists of rights that he feels he does not have, such as the right to work or the right to study. In each case, he uses the modal verb 'can't' to strengthen his construction of restriction, which he compares (lines 4-8) with those who do have a visa or right to remain in the UK. In this sense, he constructs an account where to have no visa is to be a non-person, supporting Hynes (2011) view that this is policy-imposed liminality.

Discussion and conclusions
The examples presented in this paper have demonstrated how, for these participants, belonging and what it is to 'be a person', is intertwined with immigration status and thus contingent upon having the rights to engage in civil society and the economic life of the country. I have demonstrated how a number of ideological dilemmas are at stake for participants in the way they justify these claims. This may be a dilemma between restriction and a sense of having a responsibility to integrate, or, between feeling safe and feeling restricted. However, the constructions discussed here also represent a dilemma between 'being' and 'not being' a person and being in a position of liminality. This suggests that the absence of rights is critical to this sense of 'not belonging' or 'not being'. Brown and Stenner (2009, p. 199) have distinguished between two impossible extremes or 'dimensions of vitality: unrepeatable chaos and redundant order'. For Brown and Stenner (2009, p. 200) such a redundant order is defined as 'the total static space of complete redundancy where nothing but the monotonous repetition of the same brute reality is possible. No life is possible in this frozen, stratified, non-moving space of redundancy'. In many ways, the participants constructions of belonging, shown in the extracts in this section, support this theory of being in, or near, such a 'redundant order'.
The participants' talk in these extracts constructs belonging as being more than about a geographical location, and that, rather, it is rights and citizenship which underpin feelings of belonging in the UK. This supports Ager and Strang's (2004) Indicators of Integration framework, which also saw rights and citizenship as the foundation of integration experiences. In Extracts 1 and 2, both Munir and Aziz drew on an economic repertoire to construct accounts in which belonging was contingent on them having the right to contribute to the economic life of the country (i.e. through work). However, there was also a dilemma at stake for them as they also constructed accounts which emphasised their personal responsibility to be economically active (and not 'lazy'), both to support their families and to acknowledge the welfare support given to them as asylum seekers. This also reflects the dilemma faced by the UK government between the deterrence of asylum seekers and integration of refugees. However, such a dilemma fails to acknowledge the intimate linking of the categories 'refugee' and 'asylum seeker' (in the UK policy context), where those granted refugee status may have spent several years as an 'asylum seeker' whilst waiting for their cases to be decided. As such, the use of an economic repertoire drawn on by participants in this study points to the need to re-consider the restrictions placed on the ability of asylum seekers to work. Such an approach could help foster a greater sense of belonging for asylum seekers in the UK but would require the UK government to adopt an approach to integration similar to that of the Welsh and Scottish governments who see integration as a process which begins on day one of arrival in the host country, rather than the day refugee status is awarded.
A number of discursive studies Kirkwood, Goodman, McVittie, & McKinlay, 2015;Parker, 2018) suggest that asylum seekers avoid making direct criticisms of the host country so as not to appear ungrateful for the support they have been given in the host country. Indeed, there is evidence of each of the participants' talk functioning in a similar way here, as they present themselves as grateful for the support they have received and focus on their responsibilities and ability to contribute to the economy or society rather than rely purely on state help. Here though there are a number of direct criticisms made of the asylum support system in an account, which is constructed without appearing ungrateful to the host country. In each of the extracts, participants additionally draw on a repertoire of restriction to position themselves as unable to meet their responsibilities to contribute economically to the host country. This draws attention to the multi-faceted practices involved in integration but also acts as criticism of current asylum policy which does not allow most asylum seekers to work. Munir and Aziz both draw on the idea that they are 'just eating and sleeping' at the expense of the British state, however, this does further rhetorical work to suggest that the current asylum system restricts their ability to feel a sense of belonging because they are not permitted to take part in the day-to-day economic life of the country. Such a position is exemplified by Shotter (1993, pp. 162-163) who states that: to live in a community which one senses as being one's ownas both 'mine' and 'yours', as 'ours', rather than 'theirs'one must be more than just an accountable reproducer of it. One must in a real sense also play a part in its creative sustaining of itself as a 'living tradition'. One must feel able to fashion one's own 'position' within the 'argument' or 'arguments' to do with both constituting and reconstituting the tradition.
The extracts from Bhaija and Ghirmay, construct this restriction in a much more extreme and 'total' manner; 'just a life where you are alive but you have no rights'. However, again this rhetorical work means that criticism is directed towards the asylum system itself, rather than the host country as a whole. The extracts also support the view that the asylumseeking participants may be subject to a 'policy-imposed' liminality (Hynes, 2011, p. 94) that can only be ended when refugee status (with its associated rights) is awarded.
This paper contributes towards a growing research area that has applied discursive psychology to the analysis of asylum seekers' talk (e.g. Goodman et al., 2014;Kirkwood et al., 2015). It highlights the importance of paying close attention to the talk of asylum seekers and analysing the function of such talk. Here, analysis of this talk revealed that belonging, for asylum-seeking participants, was constructed around making a contribution to the economic and civic life of the country by drawing on economic repertoires of effortfulness. At the same time, I demonstrated that repertoires of restriction were also drawn on by participants to explain why belonging was difficult. Such repertoires functioned to criticise the UK asylum system directly and thus avoided making direct complaints about the host country or its people. These findings are of importance as they suggest that opportunities for asylum seekers to work may be one key way of ensuring that they feel integrated and have a sense of belonging in their new communities. Note 1. The names of all participants in this paper are pseudonyms in order to protect the identity of participants.

Disclosure statement
No potential conflict of interest was reported by the author.