Aspirational identities and desire through discourses of productivity in marginal individuals: A case study of three women

This article investigates how marginal individuals construct a productive self in an interview. It reports on a case study of three women — a squatter, a rough sleeper, and an Irish Traveller — who inhabit uncertain and threatened homes. In response to dominant discourses of productivity, in the interviews the speakers ’ talk re ﬂ ects the desire to be perceived as able and knowledgeable individuals. Thus, rejecting their marginal subjectivities, the three women propose pro ﬁ table solutions to society ’ s issues along the very same principles of productivity heralded by dominant society. Framed within a performative notion of identity, the study elaborates on the notion of a non-sexual desire as the trigger of most human actions. The results suggest that marginality is not a ﬁ xed and segregated state of being and the stereotype of individuals like those discussed in the study as passive and out of touch must be challenged.

This article investigates how marginal individuals construct a productive self in an interview. It reports on a case study of three women-a squatter, a rough sleeper, and an Irish Traveller-who inhabit uncertain and threatened homes. In response to dominant discourses of productivity, in the interviews the speakers' talk reflects the desire to be perceived as able and knowledgeable individuals. Thus, rejecting their marginal subjectivities, the three women propose profitable solutions to society's issues along the very same principles of productivity heralded by dominant society. Framed within a performative notion of identity, the study elaborates on the notion of a non-sexual desire as the trigger of most human actions. The results suggest that marginality is not a fixed and segregated state of being and the stereotype of individuals like those discussed in the study as passive and out of touch must be challenged. (Marginality, space, squatter, Irish Traveller, rough sleeper, desire=aspiration, epistemic and agentive self, neo-liberalism)*

I N T R O D U C T I O N
The context of this article is the uncertain Heimat, the space of belonging, and its impact on selfhood. In particular, the study focuses on how, through language, marginal or 'liminal' (Turner 1969=1997) people who inhabit a precarious space in society construct their subjectivity in an interactional context. The article's theoretical framework is that of identity as a discursive and situated construct, and narratives as semiotic practices (De Fina 2020) of variable style and shape, including imperfect and even incomplete small stories emerging in the exchange between the interlocutors (Bamberg 2004;Georgakopoulou 2006). The main focus of the paper is on 'desire' and 'aspiration' as drivers that provide the impetus for individuals' identity work through their talk. The main contention is that a focus on desire makes it possible to reconceptualise the phenomenon of social marginality as a dynamic rather than permanent process that profoundly interacts with dominant society discourses and recognises individuals' identities as being performative and transformative (Butler 1990). The marginal identities constructed by the three interviewees in the study show a continuous tension between two opposing ideologies. One that advocates individuality, productivity, and rationality sustained by a 'Western corporate discourse [as] a voluntary discourse system [that] … is the most representative example of the Utilitarian Discourse system, grounded in Utilitarian ideology' (Garcés-Conejos Blitvich 2010:123). Conversely, the other ideology reflects the alternative living space, where people who occupy it share an aspiration to collectivism and alternatives to productivity, and do not necessarily champion (re)integration into mainstream society. The three women in this study generally endorse an anti-establishment ideology and show a critical relationship with the majoritarian society; however, the study argues that in this process of refutation, the economic and symbolic power of the productive ideology is still present in their talk and strongly impacts their identity construction. By highlighting the agency of the interviewed women, the study intends to challenge the widespread narrative that represents marginal people as for ever outcast, asocial, and totally detached from productive society.
While discussing desire as a driving force, the study proposes an analytical framework that, without decoupling desire from sexuality, understands it as a wider construct. At the same time, the article demonstrates the epistemological suitability of desire and aspiration for identity studies of marginality=liminality, because it encourages a more dynamic conceptualisation of socially diverse individuals and groups.

C O N S T R U C T I N G I D E N T I T I E S O F P E R S O N A L E X P E R I E N C E I N I N T E R A C T I O N
The focus of the present study is identity as a social, contextual, and discursive phenomenon. Following on from Gumperz's (1982) work on language and social identity, the research in this area has shown a constant engagement with multiple aspects of identity both in intimate and interpersonal contexts (e.g. O'Doherty, Taft, McNair, & Hegarty 2015;Koven 2020) and in public and institutional fora (e.g. De Fina 1995;Georgakopoulou 2005;Van De Mieroop 2005;Chang & Holt 2009). Scholars have proposed organic interpretations of the construct identity that allow for an individual's multiple and even simultaneous discursive constructions, while at the same time conceptualising subjectivity as a consistent continuum of realisations that preserves a person's 'transportable' (Zimmerman 1998) uniqueness and wholeness.
Identity is understood as a dialogic (Bakhtin 2008) process, first, in terms of the intersubjectivity germane to the 'contrastive' consciousness of self (Rumsey 2003:172) and, second, in the sense of the individual's negotiation with the available discourses and social roles or alignments (e.g. masculinity, migranthood, motherhood, studenthood) that circulate in society and 'interpellate' us (Hall 1996:5-6). Identity, therefore, which for Bucholtz & Hall (2004b:369) is 2 Language in Society (2023) encoded in language in overt and indirect ways, is understood as 'the outcome of intersubjectively negotiated practices and ideologies'.
Within this framework, the present study takes a social interactional approach (De Fina & Georgakopoulou 2008), according to which what the participants share during an exchange with the interviewer is understood as a semiotic practice-in terms of a meaning-producing 'habitual social activity' (Bucholtz & Hall 2004a:377) and an opportunity for speakers to create meaning jointly (Denzin 2001;De Fina & Perrino 2011). The awareness that both the researcher and the object of study impact on the investigative process (Alvesson & Sköldburg 2020) is neither new (see Cicourel 1964) nor does it relate especially to interviews. However, reflexivity and the rejection of the illusion of a researcher's neutrality acquire particular significance when the construction of identity is by people living on the periphery of society as in this study, due to the imbalance between the parties involved.

D E S I R E A N D A S P I R A T I O N A S K E Y T O I D E N T I T Y C O N S T R U C T I O N
Research on identity development and discourse, especially in the domain of organisational studies, has focused on the notion of 'possible selves' (Markus & Nurius 1986) and 'provisional selves' (Ibarra 1999). Thornborrow & Brown (2009:357), for instance, show how the construct of a better self or 'preferred self-conceptions' can be associated with institutional power and used as a strategy to control and discipline British paratroopers. The aim of the present article is to show how people in a marginal social context construct 'preferred versions of the self' (Thornborrow & Brown 2009:356) by portraying themselves as experts of a sector that is overlooked or ignored by the majority, therefore performing as agentive, self-reflective, critical, and able to engage with the productive process in mainstream society albeit in untraditional ways. Similarly to Thornborrow & Brown (2009:370), in the talk of the three interviewees, the analysis traces evidence of a desirous or aspirational self that is 'a story-type or template in which an individual construes him-or herself as an aspirant who is (i) earnestly desirous of being a particular kind of person and (ii) self-consciously and consistently pursuing this objective'.
The construct 'preferred self' is not in contradiction with that of a 'desirous self' proposed in this study in that it is theoretically anchored to the much broader notion of desire. In modern times, 'the enigma of desire' has been associated with sexuality and psychoanalysis (Lacan 1977;Kristeva 1980;Kulick 2003;Atlas 2016). In this context, desire is the pursuit of something that the subject feels is missing and therefore 'the force that both enables and limits human subjectivity and action' (Cameron & Kulick 2003Kulick 2003:124;Cameron 2005). Kulick advocated a move from sexuality to desire which 'would compel research to decisively shift the ground of inquiry from identity categories to culturally Language in Society (2023) 3 grounded semiotic practices ' (2003:123) and Cameron & Kulick theorised a desire that 'is not always and necessarily sexual ' (2003:113). Cameron & Kulick's (2005) contribution to the theorisation of desire and a definition that goes beyond sexuality is a turning point, as is that of Deleuze & Guattari (1977), for whom desire becomes a social drive that directs actions and behaviours in response to a lack produced by the social infrastructure. Similarly, desire as social construction moulded by the narratives circulating in society is a point taken up by Foucault (1979). These views are not inconsistent with those expressed by identity scholars, for instance, Bucholtz & Hall (2004b), who conceptualise desire as being inextricably associated with subjectivity. Canakis (2015:63) insightfully points out that 'Bucholtz and Hall consider desire in the context of the NEGOTIATION of identity, whereas Cameron and Kulick treat desire as pivotal to identity FORMATION' (emphasis added). Crucially, all of the studies discussed in this section understand lack and desire as positive energies that propel people's identity construction, their choices and actions, and as a social and collective (rather than an individualistic) drive. Moreover, as Spivak (2002in Motha & Lin 2014) maintains, such a social constructivist approach can support an understanding of desire as deliberate and conscious, leading to praxis and accompanying individuals' agency.
A significant contribution to the deterritorialisation of desire from an exclusively sexual sphere is Kiesling's (2011) performative notion of 'ontological desire', which not dissimilarly from Thornborrow & Brown (2009) reflects the aspiration to have or reproduce particular identities. Within an investigation of sexual identity, Kiesling focuses on the repeated attempts by the masculine subject to engage with those discourses and social practices that are associated with and ratify masculinity. Proposing a broader interpretation of desire than psychoanalytic studies encompass, Kiesling frames the construct ontological desire within an interactional perspective, in which such strategies as involvement, alignment and affiliation are core to the identity negotiation.
The present study, therefore, follows a very productive theoretical thread that through Deleuze & Guattari (1977), Cameron & Kulick (2003), and then Kiesling (2011), deterrorialises desire from the exclusively sexual sphere and recognises its presence in all human actions, from sleeping to walking or even death (Kiesling 2011:219). In domains other than sexuality, a similar reconceptualisation of desire is present in the work of such discourse scholars as Markus & Nurius (1986), Ibarra (1999), and Thornborrow & Brown (2009). Without decoupling it from sexuality, therefore, this study exploits the broader notion of desire to understand how, in given interactional contexts, individuals who are aware of their social exclusion, can aspire to performing a preferred better self to engage with dominant discourses productively and creatively.
In postcolonial studies, for example, desire is associated with the act of aspiring to and negotiating a social reality based on democracy and equality as in the African theatre by Amkpa (2004) that investigates the legacy of the empire and the 'dramaturgies' of resistance. Murray, Shepherd, & Hall's (2007:1) Desire lines explores 4 Language in Society (2023) the 'intersection of space, memory and identity' in post-apartheid South African cities that are seen as 'sites of memory and desire'. Recently, the construct of 'desire' and 'aspiration' has also been adopted widely in migration studies to explain the drivers of people's movement (see Papadopoulos & Tsianos 2007;Hindman & Oppenheim 2014;Collins 2018). This is a move away from the presumption of migration as being primarily economic towards a socially constructed view of people's movement 'in relation to the subjectivities of migrants, their states of feeling and the circulation of affect within and across borders' (Carling & Collins 2018:913). Desire and aspiration are related to imagined geographies of the world, in terms of semiotically constructed spaces (Said 2003), and often associated with 'ability', when people manage to convert their longing into action (e.g. Carling & Collins 2018). Being socially constructed, desire is the drive behind the decision to leave one's own country, while at the same time those who migrate become the object of desire for their compatriots (Collins 2018:5;Yang 2018). Migratory desire is shaped by the discourses circulated by governments that encourage or vice versa ban certain actions and behaviours, and by the narratives aimed to discipline and organise the labour forces needed at a particular moment in time (Collins 2018:5).
Within a linguistic domain, Prior (2016) folds desire under emotionality in his analysis of Southeast Asian immigrants in the US and Canada. Framed within the notion of researcher's involvement, the study explores the reaction to 'feeling questions' and 'emotion-implicative questions' often associated with the interviewees' awareness of how they are perceived as non-native speakers. In that case, desire takes the shape of a pursuit for social affiliations and success on such practical and crucial matters as asylum petitions. That desire is an effective key to understand the motivations behind actions is demonstrated by Motha & Lin (2014), who theorise it as the drive behind language learning and the promises of cultural, social, and economic capital associated with it. Like the items society promotes in the pursuit of happiness, for example, marriage or home ownership, language learning is the object of a student's desire (Ahmed 2010). In conclusion, the notion of social desire in the sense of aspiration and longing for a particular state of being and subjectivity frames this article. Moreover, the suggestion is that desire provides a privileged entry into identity construction and negotiation especially in a context of social liminality and exclusion. Desire shows that marginal identities are dynamic and people inhabiting liminal spaces in given circumstances wish to engage in a critical yet constructive dialogue with majoritarian beliefs and narratives, which suggests they are not in a state of permanent passivity and lack of agency. Through an accurate choice of discursive resources and topics, the women in this study therefore construct a bettered or preferred self by producing an 'aspirational identity narrative, [that] is an 'epic' literary form in which an individual casts his (her) self as a hero (heroine) whose life is punctuated by a series of obstacles=tests which have to be overcome in order to succeed' (Thornborrow & Brown 2009:371) Language in Society (2023)

M E T H O D O L O G Y
The interviewees whose language is analysed in this article are three women with very different lives. They were selected out of forty-eight interviews within an ethnolinguistic project on the identities of marginal people (Piazza 2021). The idea behind the original investigation was to identify how individuals who do not have a safe and stable place of residence, like squatters, rough sleepers, and Gypsies, Romani, and Travellers (GRT) who traditionally lead mobile and itinerant lives, portray themselves in an interaction with a sympathetic interviewer from settled majoritarian society. A key hypothesis was that the space deprivation would impact the way these women talk and would emerge in the form of an aspiration to a different status. The goal of the study was therefore to identify those discursive strategies that can be associated with longing and can reveal the women's relationship with dominant society.
In this study, particular attention is paid to how the three women relate to other individuals within and outside their community through a 'discursive process whereby selves are located in conversations as observably and subjectively coherent participants in jointly produced storylines' (Davies & Harré 1990:48). In the interviews selected here, the three women construct very idiosyncratic identities. As in the original study, the interviewees were deliberately chosen from three mobile and vulnerable groups (squatters, individuals experiencing homelessness, and Irish Travellers) to highlight some of the ways in which, despite the emotional consequences resulting from space deprivation, desire emerges as a positive drive that pushes the speakers to construct themselves as agentive and in control. The interviews were collected over a number of years (from 2012 to 2019) in the women's own space: for the squatter, they took place in the building she was occupying with her 'crew'; for the rough sleeping woman, in a day centre she often frequented where she got food and washing facilities; and for the Traveller, in a council run and serviced site. Similar questions were asked of all interviewees especially about the provisional space they were inhabiting and what they wished for their future. The interviews were loosely structured and the participants were indulged in their own narrative trajectories.
As is known, the indexical relationship between language and identity is predominantly a 'sociolinguistically distant one' as no one-to-one mapping of linguistic indicators to particular kinds of subjectivities exists, and 'social identity is rarely grammaticized or otherwise explicitly encoded through language' (Ochs 1993:288). Additionally, as Ochs had demonstrated in an earlier study (1992), certain discursive elements are not straightforwardly characteristic of given people, rather they become associated with some identity categories.
In this light, far from trying to define features of marginal people's language and risking to essentialise them, this study traces as much as possible their identity work and captures the way in which the three speakers on the social margin negotiate their relationship with the interviewer as a person whom they certainly perceive as belonging 6 Language in Society (2023) to mainstream society. To reach this goal, following De Fina (2003:23), in the analysis of the interviews, the lexical, interactional, and pragmatic levels are closely observed.
The pragmatic level is especially important in this study. Admittedly, there is an asymmetrical power relationship between the two parties, if not within the interview where the interviewee retains the information that the researcher is after, certainly in the broader terms of the social relation between the interlocutors. It is therefore crucial to consider issues of face (Brown & Levinson 1987). Due to the lower status that they are aware is assigned to them within mainstream society, the speakers try to safeguard their public self by embellishing their subjectivity and showing the strength of their personal resources, knowledge, and expertise. To give due consideration to this negotiation element, the study draws on Zagzebski's (2011) construct of 'epistemic self-trust', which the American philosopher posits as the basis of the modern rejection of authority and at the root of individual autonomy. The construct seems to be in line with an individual's concept of 'epistemic self' or self-worth. Dyer & Keller-Cohen (2000) apply the epistemic self to explain the discursive strategies, from use of pronouns to self and other evaluation that, in an era of egalitarian and democratic teacher-student relations, tutors employ to show students their expertise and knowledge in an indirect and informal way. Based on lexical identification of key phrases and terms, Van De Mieroop (2007) also conceptualises an epistemic self, associated with feelings and thoughts and typical of professional identity, in opposition to an agentive self, linked to actionrelated self-references that she views as characterising institutional identity. In a very similar vein, Garcés-Conejos Blitvich (2010) examines how corporate identity is constructed in the mission statements of several American corporations and finds that epistemic and agentive self-references are not numerically different and probably relate to the genre's communicative purposes.
The present analysis shows how the expression of epistemic and agentive self is the vehicle for the construction of the women's desirous or aspirational identity in the interview. By positively evaluating their own actions and suggesting how the present matters can be improved, the three participants attempt to reverse the power relations by constructing themselves as experts vis-à-vis the interviewer who thus takes up the identity of a novice. The epistemic self-trust that these women show and the construction of their self as epistemic and=or agentive well suit these individuals who, on the one hand, discursively present themselves in contestation of the establishment, while, on the other, by proposing profitable alternatives, seem to compete with mainstream society albeit on different and untraditional grounds.

T H E I N T E R V I E W S . P L A Y I N G T H E E X P E R T , S H O W I N G C O N T R O L : T H E D E S I R E F O R A P O S I T I V E A N D P R O D U C T I V E S E L F
The following section is divided into three parts devoted to each one of the interviewees-a squatter, a rough sleeper, and an Irish Traveller. The analysis shows Language in Society (2023)   7 how the modalities in their construction of subjectivities are similar both in the aspiration to change the situation they are currently in and through the production of constructive criticism of mainstream society, which puts them in the position of experts. Following Gee (1986) and more recently Fodor (2020), within the interview extracts 'stanzas' indicate a set of lines (as in poetry) about an event or a state of affairs centred on a character, place, time, topic, or perspective. Every time the idea changes, a new stanza shapes up.
We can do better as we think outside the box: Squatting as a way of life The first interviewee's talk is understood against the backdrop of society's negative discourses around squatters seen as 'possessing such supposedly deviant values as being foreign, young, criminal, anti-capitalist, drug-using and so on' (Dee & Debelle 2015:120). In resisting such social narratives, this speaker constructs herself as a competent and knowledgeable 'good squatter' (Dee & Debelle 2015) and shows an aspirational identity as a productive business woman.
The interviewee's lexical appropriateness is apparent as she responds to my statement about the interview anonymity. For easier identification, the salient choices in the excerpts are highlighted in italics. 1 (1) INTer: Interviewer; INTee: Interviewee INTer: (…) everything is absolutely anonymous, you don't need to tell me your name, so I will be very, very respectful and don't worry about it, because I know, that (name) told me that sometimes you may be sort of, not you personally, but people in generally [are suspicious? INTee: [We pretty much, we givewe have given full disclosure to the [inaudible] so the court papers came in my name. So yeah, we called in to introduce ourselves to the Honour, sent him an e-mail. Okay.
The use of 'we' in association with the technical phrase 'full disclosure' 2 encodes the speaker's authoritative tone and her high status in the squatter community (Kacewicz, Pennebaker, Davis, Jeon, & Graesser 2013). This is also confirmed by the overlapping with the end of my sentence, interpretable (or at least so I did as the interlocutor) as a light face-threat (Culpeper 2016).
Similarly, in the following extract the speaker continues to construct herself as an expert, this time by setting up an alternative scenario in which instead of evicting the squatters, the local council charges them rent and makes a profit. The refrain "they weren't really thinking outside the box" (that reappears in stanza (3)) sets up an us-versus-them opposition, through which the speaker casts herself as profoundly knowledgeable of a situation ignored or misunderstood by others. The obtuseness of the others inspires the speaker to design creative responses to the housing 8 Language in Society (2023) shortage, which she argues is a false problem ("there are too many empty properties for there to be a housing shortage"). The phrase "they would have had" encodes both her wish the council acted differently and the regret for losing her space. (2) INTee: (…) it ended up costing the council that year that they would have had £12,000 pounds in rent plus all the council tax from each individual that live there, so that's five people at £80 pounds, isn't it, usually £80, some-thing… I don't know how much it is a person but all the council tax plus that rent. Plus the money to evict those people, plus all the money it took to fix the property up. So they weren't really thinking outside the box and so that inspired me to start writing alternative housing solution proposals and development proposals on how to fix the shortage of housing because there are too many empty properties for there to be a housing shortage, it is just… when you do the math, it doesn't work, it doesn't work at all and I learned a lot from it because I started writing at a kitchen table of the squat in late November, or early December of 2007.
Immediately visible is the speaker's clear-minded critique of the council's intervention that she considers obtuse. This speaker's desire is encoded in the continued polarization (Jeffries 2010) between her positive self and the negative rest of society through her self-construction as the expert able to design original and alternative contributions.
Similarly, in stanza (3), the interviewee shows legal prowess (note the technical term IPO: interim possession order) in her detailed discussion of how court orders can at times be served incorrectly. A list of positive interventions encodes the woman's desire to show her planning ability from "we will save them the cost of any additional court fees" to "we will put a fresh coat of paint on it, we will keep it tidy, we will keep it clean". Such a linguistic realisation of what Van De Mieroop (2007) and Dyer & Keller-Cohen (2000) call agentive self is also supported by such mental processes as "we just need ample notice" 3 that rather than defining a specific need, in this case the length of warning before leaving the premises, indicates the speaker's legal knowledge of temporary residence rights. (3) INTee: We have received court papers for IPO and we got notice of court papers yesterday. We could win if the IPO was served incorrectly or it is not right. We've got legal counsel. INTer: What are you aiming for? To stay or turn it into commercial space yourself or what? INTee: Often times the IPO if it is not served correctly they have to get a stay, in your own possession order, it is an immediate one to get you out quickly, Language in Society (2023) whereas with possession order, it depends on when the bailiffs get around and what kind of arrangement we would like to get like a-just an agreement to be able to stay basically, it is a standing order or something like that that we will turn the building over to them, we will save them the cost of any additional court fees, any security and in the cost of bailiffs.
We will return the keys over to them. When they need the property we just need ample notice, we need 30 days.
Later on in another extract still on the theme of the stanza, the speaker describes her plan to upkeep the occupied premises, while the owner ponders what to do with them.
(4) INTee: we will put a fresh coat of paint on it, we will keep it tidy, we will keep it clean.
What do you want to do with it, do you want it to be commercial space, you want to be residential space? that is the theme that we will go with the decorating of it and they have constant access or if a surveyor needs to come by or a planner, somebody to check the plumbing, electric, and there is someone there constant, they don't have to leave their office. The property manager does not have to leave their desk. We are there to do it. So it actually works out to be mutually beneficial arrangement. We have not reached that yet with this landlord, he immediately went to a solicitor's and I am sure they weren't thinking outside of the box, if they read our offer or letter then maybe (.) I don't know, anyway they just want to sell so we are trying to go back and negotiate. We have a legal observer and someone to give us legal advice that will be present after the IPO, so the IPO is granted at court then they [we] will come and stay here because they have to give us 24 hours' notice to see if it has been because most of them execute improperly and people end up to getting to stay. (1.0) But you need to know the laws and that is when you havemost of them are issued improperly.
At the beginning of the stanza is the interviewee's insistence on the IPO topic and disregard of my admittedly not very relevant question, which can again be seen as a face-threat. Throughout the excerpt, the speaker insists on constructing her persona as a far-sighted expert with greater knowledge of the housing situation than professional social actors like council, judges, and bailiffs. Equally relevant is the persistent use of the plural 'we' associated with the disclosure of the people she has at her disposal ("We have a legal observer and someone to give us legal advice"), the mention of her good knowledge of the legal system ("you need to know the laws, most of them are issued improperly"), and the advertising of the group's capacity to offer services ("The property manager does not have to leave their desk. We are there to do it.") The material processes associated with the squatting 'crew' ("we will put a fresh coat of paint on it, we will keep it tidy, we will keep 10 Language in Society (2023) it clean") suggest the speaker's critical engagement with society's negative discourses around the anti-social disruptiveness of squatters by showing how constructive and profit-making 'good' squatters can be when dealing with urban developers.
The extract below concludes the analysis of the speaker's grandiose selfconstruction. On the basis of key phrases and recurring terms, her agentive self appears linked to her being a successful broker in the past and on those grounds, being able to develop a future vision for squatters as productive social actors in a liberal economy. The performative double-voicing (Baxter 2014) by which she conjures up a scenario of negotiations with the owners and developers constructs her as a self-trusting woman committed to defying majoritarian society's rules. for 14 years in the '80s and '90s. (…) So theirs is the kind you approach with an agreement in advance and say, 'We would like to be your onsite security, be available to show it to potential buyers and estate agents, so that people could come in.' In conclusion, in the talk of this interviewee, as in the others that follow, the construction of agentive self, associated with references to doing and carrying out action, occasionally blended with a self-assured epistemic self, connected with planning, is the discursive strategy the speaker resorts to as she negotiates with the interviewer. With such a strategy, this speaker rejects her marginal identity and therefore demystifies the damning narrative around squatters as people going against the productivity ideology. In its stead, she appears as a sophisticatedly agentive individual who constructs an aspirational self that can contribute productively to liberal society and adhere to its profit-making ethos.

I'm very strong. I'm a very, very strong woman': A rough sleeper's desirous identity
The second interviewee is a woman experiencing homelessness. She was interviewed in a day centre that provides the basics from food and washing facilities to IT services, health care, and respite from the street. Much younger than the first interviewee and with a very different experience, she nonetheless is an aspirant and shares with her the very same longing to construct an agentive self, capable of contributing to society's productivity.  This stanza, which continues below in (7), centred on the woman's biography, already shows a strong similarity with the first interviewee's talk. In particular, the reference to the speaker's professional background and past job in a productive and profit-generating world echoes the first speaker's reminiscing about her corporate past (stanza (5)). The admission of being a rough sleeper, therefore, is softened by the mention of her international productive past and minimised by the qualifier 'little' preceding her 'breakdown'. And it's also going to give me a good step to be a social worker when I have my shelters running. One here one in (place name) built with the black women over there because that will also change lives and change thinking a white women with a black woman doing shelters and a white woman who's been homeless so she can relate. Rather than English people going over that are building shelters when they don't have a clue. Or working at orphanages when they don't have a clue what really goes on in the street.
In this second woman's talk the amount of active material processes indexes the clarity of her plan, centred on building shelters, supporting homeless women, and training as a social worker. Similar to the first speaker, she constructs an agentive self through the choice of material processes producing a result ("I did flags of the world in chalk and I did a YouTube video, I've written to the Queen") occasionally combined with an epistemic self, associated with feelings ("it's been a real struggle"). Likewise, through a me-versus-them polarity, she condemns those in charge of services for marginal individuals ("they don't have a clue what really goes on in the street"). Worthy of note is the shift from 'I' to a 'we' in the expression of her future plan ("So what we want to do is we want to have [inaudible] like an allotment") that again reminds us of the squatter's collective reference to her crew.
In the next stanza, the speaker's desire materialises as an aspiration to have a visible impact on reality and create a better and fairer society. Language in Society (2023) 13 involved and seeing things because the perception is I think for a lot of people in the public drug addicts alcoholics prostitutes blah In this last excerpt, towards the end of the interview, the speaker self-categorises as a person who would be willing to re-join mainstream society although 'on a different level'. This positioning resonates with Sack's (1992) notion of Membership Categorization Device (MCD). In a study of police interrogations, Stokoe (2010:59) uses MCD to show how male suspects accused of violence against women defend themselves by constructing 'different categories of men, (…) claiming membership in one (who do not hit women) by recruiting the notion of the other (who do)'. Similarly, in this and the other extracts, the speaker recruits the notion of 'campaigner for social justice' by consistently referring to particular actions ('category-bound activities') or characteristics ('natural predicates') that are associated with that role, such as planning for women's shelters, appealing to the relevant authoritative figures or lecturing the wide public on justice and equality topic. The inclusion in the category of socially committed yet profitmaking person is predicated on the grounds of her past experience as a professional in fitness training and her knowledge of how to involve the corporate world ("Investec Bank and all these big corporate banking") to realise her project. What, on the contrary, the speaker is refuting is the self-categorization as a rough sleeper; to this purpose the few references to this category are skilfully obscured and minimised.

Aspiring to normality in a Travellers' site
The last speaker in this study is an Irish traveller and lives in a caravan on a councilserviced transit site that secures electricity and washing facilities for up to three months in exchange for a weekly rent. Like for the others, her desire to access basic necessities is encoded in the numerous verbs of strong volition ('want', 'need').
At one point, the main speaker is joined by a second woman who enters the caravan during the interview and joins her in expressing her critical views of the authorities. Along similar lines, INTee2 summarises in very clear conditional clauses what the council "should" do to avoid wasting money on court orders against mobile communities, and instead generate income. The women echo each other in constructing themselves as productivity experts, in a guise that is strongly reminiscent of the first interviewee's critique of the authorities being obtuse and "not thinking outside the box". The hypothetical clauses encode the women's desire for better life standards and their expertise as site residents. Once again, therefore, these marginal speakers indicate financially viable alternatives for mainstream society and in so doing establish themselves as experts with inside knowledge.

C O N C L U D I N G D I S C U S S I O N
This study has reflected on how some individuals who are marginal, outcast, and far from productive and profit-making in the labour market construct their self in an interview. None of them has access to a private and secure home and all occupy provisional and threatened spaces that society frowns upon and discriminates against. These women show an awareness of how mainstream society perceives them and of the damning narratives that abound around them, especially the concept that within a neo-liberal society, exclusion is viewed as an individual's responsibility (Lemke 2001) and reintegration is similarly predicated on the grounds of personal agency. Echoing such discourses, these three women take the interview as an opportunity to engage with the reality from which they are excluded. As they interact with a representative of mainstream society, albeit one who is sympathetic to their cause and who approaches them in their own space, the speakers' talk betrays a longing to meet what they believe the interlocutor's expectations are and a desire to show a preferred self that is in line with the dominant hegemonic narratives. Thus, the three women inscribe themselves in a present reality of relative success and positivity, thus showing their potentials for participation in the labour market; they, therefore, discard their experience of space insecurity and denial by casting it in a far-away (past) frame or by diminishing its impact on their self. At the core of this strategy is the construction of an aspirational, positive, agentive, and epistemic self, capable of proposing viable productive solutions that can be embraced by mainstream society.
In a neo-liberal society, liminal individuals like the three interviewees are generally seen as responsible for their condition, blamed for not contributing to society's productivity, and on this basis are ostracised. However, the discussion in this article has shown how they can appropriate a discourse of constructive criticism of mainstream society that, rather than refuting its ideology, advocates and supports it. By doing this, the three interviewees convincingly argue against the failed Language in Society (2023) productivity of the authorities and in its stead invoke more practical and visionary profit-making alternatives. Zagzebski's (2011) construct of 'epistemic self-trust', instigating the rejection of authority and individual autonomy, seems to apply to the discursive strategies of the three women and their political critique of the establishment. However, what is noteworthy is that it is along the very same principles of productivity heralded by dominant society that their attack is formulated. In expressing such a criticism, the three interviewees' words encode their desire to construct themselves as both cognizant of the way their condition is perceived by hegemonic neoliberal society and able to produce alternatives that, in their view, can improve productivity, hence profit. Therefore, when analysing performative identities, desire as a broad drive to obtain what the subject feels it's missing reflects what we 'wish to create in others in order to gain approval of our identities (which could be said more colloquially as 'we care what other people think')' (Kiesling 2011:235).
To conclude, this case study shows that the relationship between mainstream and marginal sections of society is complex, varied, and fluid rather than static and neatly delineated. Individuals living on the margin are in opposition to but also in a dialogue with hegemonic society and their talk suggests that they feel the desire to, if not conform, at least challenge it on the same grounds on which their exclusion is predicated. It is hoped that this study casts a better and clearer light on social marginality, on the one hand, arguing against a stereotyped discourse of exclusion that sees people like the three interviewees as unable to have agency and highlighting, on the contrary, their sophisticated individual skills for critical analysis. On the other hand, the study also suggests that marginality can be like homelessness, not a permanent but a temporary state that individuals can abandon if their voices are heeded and taken into serious consideration.