Abstract
The purpose of this research is to characterise the internal conversations of individual young people in transition from state care, with a goal of supporting practitioners’ approaches to ‘pathway planning’, and contributing to theories of agency in emerging adulthood under conditions of long-term adversity. We used Margaret Archer’s theory of agency, in which internal conversations are regarded as playing a central role in mediating between structure and agency. Archer describes three different modes of (actively agential) internal conversation: communicative reflexivity, autonomous reflexivity and meta-reflexivity; and one mode of passive agency: fractured reflexivity – associated with adverse contextual constraints. In a qualitative design, using Archer’s open-ended interview framework we met nine participants, aged 19–24, varying considerably in outcome, recruited at a specialised service for care leavers in London UK. Analysis of the interview data included an extended phase of individual case analysis. Two participants experienced emerging active agency (however, both experienced elements of fractured reflexivity). Three participants experienced fractured reflexivity, with limited internal conversation, and low planning agency (however, each experienced green shoots of focused internal dialogue). Four participants experienced a survival-oriented mode of internal conversations, a category not found in Archer’s previous work. All four of these participants showed some elements of fractured reflexivity, but narrated rich internal conversations, including accounts of adverse experiences in which self-reliance was seen as having been essential. The majority of participants (across modes) found the idea of planning (beyond day-to-day) anathema.
Introduction
Martha Nussbaum (2011) defines practical reason as: “Being able to form a conception of the good and to engage in critical reflection about the planning of one’s life” (p. 34). But how precisely do individuals plan, imagine, make decisions and reflexively organise their lives in contextually adverse circumstances? ‘Planning agency’ has been recognised as a key dimension of agency in sociology (Emirbayer and Mische, 1998: 983–993), psychology (Bandura, 2006; Larson and Angus, 2011) and philosophy (Bratman, 2013). But which theoretical frameworks acknowledge substantively both the potential capabilities of individuals to plan future-oriented goals and the real-life nature of contextual adversity, especially the corrosive contexts of repeated and ongoing social disadvantage (Nussbaum, 2011)?
In this paper we report a study of young adults who are in transition from care to an independent and interdependent adult life, a population known to experience compounded contextual discontinuities and disadvantages, and who are required to plan their lives with the help of the state in pathway planning (Stein, 2012).
A key underlying theme of this paper is that individual reflexive planning during the transition from state care is both crucial (how else is the young person to make progress in his or her life?), and is under real-life current and historic threat from corrosive and undermining experiences.
Internal conversations: Agency and reflexivity
In this study we use Margaret Archer’s realist general theory of agency / structure, in which practical reason and reflexivity are characterised as internal conversation. In an approach that resonates with Nussbaum’s work on capabilities, internal conversations are seen as having a central role in individual agential functioning (Archer, 2003, 2007, 2012). For Archer, human agency is made possible by our first-person capacity for reflexivity: “the regular exercise of the mental ability, shared by all normal people, to consider themselves in relation to their (social) contexts and vice versa. Such deliberations are important since they form the basis upon which people determine their future courses of action – always fallibly and always under their own descriptions” (Archer, 2007: 4). Internal conversations (Emirbayer and Mische, 1998: 974) make it possible for us to identify and define concerns that are our own (concerns are defined as: those areas of one’s life that matter most, or are personally most important) – these form an individual pattern, with ultimate concerns forming the core aspects of our identities, or emerging identities. Planning is regarded as a key function of internal conversations, as it enables us, over time, to modify ourselves by reflecting upon what we care about most, and how our various concerns and commitments might or might not dovetail (i.e. fit together in a satisfying and sustainable way). And crucially: “we make our lives, at least in part, by deliberating upon the structural and cultural contexts in which we find ourselves, often involuntarily” (Archer, 2003: 52).
In empirical studies, using qualitative interviews with a UK community sample, and with a UK university student sample, Archer has described patterns of internal conversation associated with (a) active agency – clear and efficacious governance of self in the world and (b) passive agency, in which people find themselves ‘disorientated’ and unable to plan coherently (Archer, 2007, 2012).
According to Archer (2003, 2007, 2012), active agency is characterised by three different modes of internal conversation: communicative reflexivity (‘those whose internal conversations require completion and confirmation by others before resulting in courses of action’), autonomous reflexivity (‘those who sustain self-contained internal conversations, leading directly to action’) and meta-reflexivity (‘those who are critically reflexive about their own internal conversations and critical about effective action in society’).
In respect of passive agency, Archer argues that adverse contextual constraints may lead individuals to find themselves unable to adopt an actively agential stance, unable to use internal conversation clearly to delineate concerns, and unable to define, plan and execute life projects, i.e. to experience fractured reflexivity. Fractured reflexivity, the ‘passively agential’ mode, is defined as ‘those whose internal conversations intensify their distress and disorientation rather than leading to purposeful courses of action’ (Archer, 2007: 93). This may be a temporary state (Archer, 2012), or may perhaps be maintained over time by cascades of negative interplay processes between real-world adversities and agential passivity.
The relevance of Margaret Archer’s work for care leavers
Young adults making the transition from being ‘looked after’ (in out of home care) to adult independent (and interdependent) living provide one strong example of individuals dealing with multiple contextual stress (Stein, 2012). Long-term educational disadvantage, lack of job opportunities and restricted housing options compound the adversity already experienced by young people whose family relationships have often been characterised by maltreatment and/or parental loss, followed by frequent foster and residential care placement breakdowns or changes (Geenan and Powers, 2007; Jackson and Cameron, 2012; Jones et al., 2011; Stein, 2006; Vinnerljung and Sallnas, 2008; Wade and Dixon, 2006). Mental health problems and substance misuse are frequent outcomes for some young adults leaving care (Akister et al., 2010; Dixon, 2008).
The lives of children and young people in care cannot of course be summarised by statistical data. There is widespread recognition that experiential aspects of being in care can include recurrent lack of emotional availability of caregivers; chronic anxiety and low self-worth; feelings of powerlessness (in relationships, and in decisions about moves and transitions); and feelings of isolation (Schofield and Beek, 2009). While some of these experiences may have been balanced by positive and strengthening processes (during, for instance, a stable foster-care or residential care experience), recurrent negative experiences are likely to have taken their toll on developing reflexivity. It would therefore not be surprising if the experiential magnitude of transition might trigger existential anxiety and self-doubt.
Margaret Archer’s theoretical framework seems particularly relevant to the focus of this study for several reasons. First, most of the support services provided to young people who are leaving care are based on the gradual development of reflexivity, planning and communication. Archer’s categories of reflexivity, including the notion of fractured reflexivity, provide one theoretical (and potential practical) starting point for understanding individual practical reason and planning under conditions of severe contextual adversity (Archer, 2012, chapter 7, 2003, chapter 9). The idea of internal conversation may be relevant for practitioners looking for different ways to engage young people in the complex process of planning and imagining their futures (Stein, 2012). Second, the notion of internal conversation has a rich theoretical background, specifically in Archer’s theory (Archer, 2003), but also in a wider literature relevant to our understanding of early development of private speech and social interaction (Morin, 2005; Vygotsky, 1962; Winsler et al., 2009). For instance, in a recent study of typically developing 4–7 year-olds, use of private speech during cognitive tasks was associated with autobiographical memory measures – narrative recall, narrative cohesion and the specificity with which events were recalled (Al-Namlah et al., 2012). Greater understanding of developmental aspects of internal conversations and autobiographical memory will likely enhance opportunities for early intervention in areas such as life-story work with young people who are in / have been in care.
Although Archer’s work seems particularly relevant, and theoretically very rich, we were nevertheless sceptical of Archer’s discrete formulation of reflexivity (Archer formulates modes of reflexivity in individuals as categorical (i.e. present or absent), and typical of the individual (e.g. ‘a communicative reflexive’) (Archer, 2003)). We were especially sceptical when this is applied to young adults who, whatever their social background, are still discovering their stance towards the world (Arnett, 2004). We therefore ensured that our methodology allowed us to use both deductive and inductive approaches to the analysis.
In summary, Archer’s theory would appear to be relevant to our interest in agency in young people leaving care, and to help us formulate the research question as one of characterising the internal conversations of individual young people in transition from care to an adult independent and interdependent life. Using Archer’s model we had several related research questions in mind: For those young people experiencing active and planful agency, would particular modes of reflexivity (e.g. communicative, autonomous, meta-reflexive) be evident? For those experiencing passive agency, would the specific pattern of fractured reflexivity described by Archer apply, or might other forms of fractured reflexivity be described by participants? Would individual young people in our study experience one clear form of reflexivity? We were also interested in whether young people might describe modes of reflexivity and planning agency which did not clearly fit Archer’s model.
Qualitative enquiry
The formulation of our research question, emphasising the first-person perspective, drew us to qualitative methods, using individual interviews.
For our analysis method there were several factors to bear in mind. First, in line with the research questions, we wished to choose an approach emphasising very detailed analysis of individual interview transcripts (before developing across-case themes). Second, our epistemological assumptions were, with Archer, critical realist rather than purely social constructionist. While we were deeply interested in the constructions participants brought to their circumstances, we were committed to the idea of structural and contextual events in a ‘real world’, including events associated with structural inequality. Third, we wished to balance Archer’s theoretical approach to internal conversations (i.e. with ‘given’ specific categories and processes of internal conversation) with an inductive approach, respecting the specific individual experiences of young people who had been in care, and allowing the possibility of types of internal conversation not described by Archer.
Taking each of these factors into account, we chose an analysis method which actively combined interpretative phenomenological analysis (IPA) and Miles and Huberman's Interactive Model (Miles and Huberman, 1994; Robinson and Smith, 2010; Smith et al., 2009). Both these approaches are regarded as appropriate for critical realist epistemologies (Robinson and Smith, 2010); Miles and Huberman’s approach is particularly useful when balancing deductive and inductive analyses; and IPA is now regarded as an extremely well developed methodology for the detailed analysis of individual data (Smith, 2011). Furthermore, use of the specific combination of analysis methods had been carefully described in the context of a study of young adults, a focus close to the present study (Robinson and Smith, 2010).
Method
Research setting and sample selection
The research was conducted at a public service resource centre for socially excluded young people in London, UK. Young people who had left care were free to attend. Courses were available, and mentorship from staff and from peers. All participants had used these services.
Participant characteristics.
Heterogeneity was maximised to include a range of outcomes following transition from care. Our total sample purposively included young people who could be broadly described as ‘moving on’, or ‘survivor’, or ‘victim’ (Stein, 2006). However, no attempt was made formally to measure these variables – rather discussions were held with staff at the Centre, explicitly seeking participants who would cover the range of Stein’s categories. Staff then suggested individual participants, and introduced each participant to one of the researchers (IH). In each case IH discussed the research and research procedures informally with potential participants before entry into the study.
The participants were diverse on many contextual variables: ethnicity, number of family placements, accommodation, contact with foster/biological family, employment, education, involvement with criminal justice system and sexual orientation.
The following exclusion criteria were applied: any current involvement in mental health services, or involvement in the last 12 months; participants considered at risk of harm to self or others; participants needing a translator; asylum-seekers; those with a suspected IQ below 70; and those currently experiencing an episode of psychosis.
Interviews
Individual interviews, which were conducted during 2010 and 2011, took place in a room at the centre, which was familiar to all participants. Participants had already indicated willingness to engage in the interviews, and all procedures were explained at the beginning of the first interview session, at which written informed consent was also sought. There were two interviews, which corresponded to the two parts of the Archer 2003 qualitative interview framework (see below). All participants completed both interviews. These interviews were open-ended, and focused on helping the young person communicate her or his reflexive thoughts about internal conversations, and about real-life situations that had occurred.
Interview 1
Participants were sensitively introduced to the idea of the internal conversation and then asked to elaborate on whether they found themselves using an internal conversation. Participants were asked to describe a range of uses for their internal conversation. In addition there were 10 specific mental activities (see Archer, 2003: 161) which were used as prompts for further discussion – planning, rehearsing, mulling-over, deciding, imagining, prioritising, re-living, clarifying, imaginary conversations and budgeting (money, time, effort). Prompts were used in an open-ended and conversational way during interviews – we found them extremely useful in cueing the participant to describe a range of different internal conversations. Afterwards, participants were invited to offer any other themes which featured in their own internal conversation (Archer, 2003: 161–162).
Interview 2
The second interview focused on two main areas: First, participants were asked about their current concerns (i.e. which areas of their life mattered most to them, or were personally most important). We ensured that participants had an opportunity to discuss how long a concern had been evident, whether the concerns dovetailed, whether or not time was spent in thinking about actions that might follow from concerns, and whether or not it was felt that anything in their backgrounds had been helpful or obstructive in relation to the realising of concerns. The second area for discussion was to look forward and outline ‘life-projects’. (Archer, 2003: 161–162)
The interviews lasted approximately 1 h each and were recorded. Participants were also reimbursed for their travel and given a £10 voucher per interview for a local shopping centre, to thank them for their participation.
Ethics
The research was approved by the University of Essex’s Health and Human Sciences Care Research Ethics Committee, and the UK Social Research Ethics Committee (SCREC). We took several steps to ensure that participation might be a positive and safe experience for this potentially vulnerable population. First, the first four participants were asked for detailed feedback about the interview process. Following feedback, minor adjustments were necessary to the interview process. Second, regular meetings between the primary researcher and the research site practitioner team ensured good understanding of the research by the team. Third, a safety net of support was put in place, in the event of any young person reporting distress with any autobiographical material that they disclosed.
Analysis
All interviews were transcribed verbatim. In the reporting of data, all names of participants are anonymised, and some biographical details have been omitted to maintain confidentiality. Our analysis method, which prioritises very detailed analysis of individual interview transcripts, systematically combined IPA and Miles and Huberman's Interactive Model, for reasons outlined above (Miles and Huberman, 1994; Robinson and Smith, 2010; Smith et al., 2009). The combined analysis method (Robinson and Smith, 2010) has two stages: (a) individual interview transcript notation and theme creation; (b) theme integration and clustering of individuals. In the first stage, marginal notes were made on individual transcripts, with first responses and possible themes. Themes were then systematically developed for each individual. Clusters of themes (superordinate themes) were then constructed for each individual participant, together with narrative individual case summaries. In the second stage, where superordinate themes appeared to be similar across individuals, preliminary clustering of individuals was then considered, with continuous iteration between within-case and across-case analysis. Intensive further analytical work with individual interviews, and with participant clustering, led us to a formulation which had only a partial correspondence with Archer’s categories of internal conversation.
The researcher who conducted the majority of interviews (IH) led on the initial analysis of data; the other researcher (PA) reviewed all transcripts and initial theme lists; detailed discussions were then held between the researchers, reviewing and then agreeing the emerging themes and clustering of cases.
Results
Analysis of individual interviews, working iteratively between inductive and theory-driven themes, led us to think of participants as falling into three categories: those who predominantly experienced (a) emerging active agency, or (b) survival-oriented reflexivity, or (c) passive agency / fractured reflexivity. The first and third categories are very broadly in line with existing categories described by Archer (2003), but the second category emerged inductively from our analyses and did not ‘fit’ with Archer’s categories. Indeed the second category (survival-oriented reflexivity) has elements of both active and passive agency. We give greater space below to those individuals describing survival-oriented reflexivity because (i) participants described it as rooted in their experience of adversity, which is a prime focus of this study, and (ii) it is interestingly different from Archer’s existing reflexivity categories.
Emerging active agency
Two participants (Tyreece and Zavie) experienced emerging active agency. Both had rich and inventive internal conversations, very clear concerns, and were fully engaged in day-to-day projects (including work and/or education) consistent with their beliefs and values. However, the sense of active agency was partial for these two individuals: Tyreece experienced highly productive internal conversation, but also fractured reflexivity; Zavie found himself engaging in productive internal conversation at work, and less so out of work, and expressed deep ambivalence about planning. Neither showed a fully-fledged mode of reflexivity (i.e. communicative, autonomous, meta-reflexivity) and therefore we labelled the category as emerging active agency.
Tyreece was deeply thoughtful about internal conversations, describing spontaneously his own personal style: “I use visual, I visualize in my head – that’s my way of learning and my way of thinking”. And, while listening to others’ ideas: “I’d be thinking further along, way further along to the point where, OK, what can his ideas benefit – let’s put his ideas into play.” “if it’s a more serious thing – that could be life or death or could be something that would affect someone else”, and “it’s just about finding the road which you’re going to take”. For example, at work “I realise it most because I’ve got a certain amount of people that I have to look after and I have a certain responsibility – and my negligence can cause harm”. However, he acknowledged that “sometimes I don’t listen to myself and I – it gets me into trouble”.
Neither participant experienced ‘pure’ active agency: Tyreece picked out threads of fractured reflexivity in which he referred to some situations in which he ruminated and experienced distress: “I overanalyse too much”; and Zavie was ambivalent about planning: “I don’t really plan my future – I’m someone who believes that you don’t know what’s going to happen – so I’m someone who lives for each day as it comes.”
Survival-oriented reflexivity
Four participants (Danny, Don, Charelle, and Nailah) experienced what we termed survival-oriented reflexivity. Internal conversations were partly driven by a focus on profound self-reliance and/or day-to-day survival, justified reflexively by detailed and specific autobiographical accounts of adverse family, foster-care (and sometimes institutional) experiences in which strict self-reliance was seen as having been essential. This stance was seen by participants as having provided self-reliance-oriented safety (rather than socially supported safety) in an unsafe world.
The only person I could trust is myself and my head
For each of these four participants, their sense of their own internal conversations originated with adversity. The following quotes do not refer to internal conversations, but rather to the reflexive reasons given by these participants for their personal approach to ‘making their ways through the world’ (Archer, 2007). Danny: “I ain’t gonna lie – I had trouble when I was a child. No-one would talk to me; my parents didn’t pay any attention – stuff like that. And I never really had friends – I just had people who wanted to hang around with me to be a clown, be someone to laugh at for a couple of hours. So I developed a state of mind where the only person I could trust is myself and my head. And my head tells me what I want to do”. And: “I feel like I’m my own mum. It’s kind of scary”. “(Being in foster care) was helpful for me because I know what I have to do when I’m alone – um – I have to just get on with things, I have to just make sure to eat and know when to do – but it also wasn’t helpful because I didn’t get to experience things in a more wider scale, I was um set apart from everybody else, which in turn made me a bit of a loner - -.” “when you grow up in care – you don’t let nobody get close. That’s the way it is. You’ve got to protect yourself ‘cos nobody else is there for you. That’s how it is. Like with your parents like that’s your backbone – they hold you up when you’re falling and whatever. We never had that. It’s only us keeping ourselves up. So when it comes to people – it’s like – hold on – I’m not going to let no-one push me, ‘cos I’ve only got myself to pick myself up.” “I do things on my own too much. I think ‘cos how I was raised and brought up like, left alone to defend myself and do everything myself I just have this natural habit where I just get up go and do things myself. I know I will only ask for help when I’m in really, really deep trouble”.
Internal conversations
Don is a knowledgeable internal conversationalist. He outlined in considerable detail in the first interview his “process of thought”, which was complex and flexible. He discussed it reflexively: its origins and its current usages, and the metaphors and models it was based on, including football. The positions of different members of a football team acted as a metaphor for thinking differently and responsively as situations change and “they have to rethink their process of thought”.
Danny’s discussions about internal conversation were detailed, although fractured and low in instrumental orientation. The richness was evoked at the beginning of the first interview, when he explained: “it kind of abbreviates on what I am myself”; “A man is what your mind tells you what you know is going to be right for yourself”; “I love the fact that peoples’ minds can work in different sets of styles and ways”.
Charelle’s thoughts about internal conversations were, throughout both interviews, deeply ambivalent. On the one hand, she reactively followed her instincts and feelings about many situations, frequently “daydreamed”, was easily “distracted”, and tended to “go with the flow” rather than plan. On the other hand, she valued the help internal conversations could have for concentration, and for reviewing very complex and important situations, such as “interviews, job centre, you can’t just go there” – “everything you do you’ve got to think”. In line with her survival-oriented stance and horizon, Charelle’s primary concern was “myself”, focusing on budgeting and paying the rent, and mentally preparing for her college course. “I’m not really thinking ahead of that at the moment and I don’t want to as I always tell myself: “Take one step at a time ‘cos I always think the more far ahead you think about something the less likely it is to happen”. On the one hand “I have a habit of thinking too much”, and “it leads me kind of to feel anxious like anxiety and um – especially anxiety – you get paranoid”. On the other hand, “I think it’s good to analyse as I said, because it makes you more aware of things”. “Like I’m very determined and proactive so – like if I want to do something, I need to sort something out, I think how I’m going to do it, how would I go about it and I’d just get up and go and do it after”. “I think more on the worry. I give it about 65 to 70% worry and the rest out of 100 helpful (laughs)”.
I don’t believe in planning
Although the ‘survival-oriented’ participants experienced rich internal conversations, planning was regarded with scepticism (by all four). Charelle: “I don’t like planning ahead, I like to take my days as they come”. Don: “At first I’m ambitious, but then something goes wrong and then I kind of write things down to make sure that doesn’t happen again”. Danny: “If there’s just one day, I’ll plan for it.” “I don’t believe in planning further ahead. Because you never know – you never know.” Nailah also felt that her plans rarely worked out.
Passive agency / fractured reflexivity
Three participants (Brittany, Joe, and Corrina) experienced passive agency / fractured reflexivity: ‘thin’ internal conversation, an internal dialogue preoccupied with negative affect and providing virtually no instrumental guidance for them, and a lack of engagement in substantive projects. They were ‘getting by’ on a daily basis. However, ‘green shoots’ of potential sources of active agency in our participants’ thinking was evident in all three cases, linked especially to contemplation of family and/or friends.
For Brittany internal conversations were thin and unhelpful, and were linked to dealing with “annoying stuff”. They made “annoying stuff” “even more annoying”. Planning, for Brittany, was: “pointless” – “I don’t plan for tomorrow, I plan for today”. “I don’t really think long-term” – “I don’t really think about (the future) apart from when I come to this centre ‘cos I’m made to. Other than that, my head’s blank; I don’t think about it and take each day as it comes”. “I can’t plan, I have to do it the day before”. On rehearsing: “I don’t. I never do. I don’t. I just say it”. On thinking about the future: “No, not really ‘cos it never goes to plan anyway”.
Corrina, when asked about concerns, what was most important in her life, said “my family - - - er - - - my friends”, and “I wanna go to college” (the narrative then focused mainly on family, rather than college plans). She explained that when her grandfather (to whom she had been very close) had died three years ago, she and her father had re-established contact (they had been estranged throughout her childhood and adolescence) and “now we see each other all the time” – “at first it was hard but now it’s OK”. Whereas much of Corrina’s internal conversations seemed wafer-thin, she talked clearly about the profound decision-making involved in this new contact (initially via a telephone call from him): “yeah, like part of me was thinking “well should I go and meet him?, part of me was thinking “should I meet him in a public place?”, and part of me was like “should I meet you in your house?” so it was really confusing”, and “Do I really want to know him after all this time?”.
Discussion
In this study we have been interested in what aspects of practical reason and reflexivity (conceptualised as internal conversations; Archer, 2003) might be experienced by individual young people with histories of severe contextual stress – in transition from care. In this section we will focus on six areas: (a) reflections on Archer’s theory; (b) the analysis of participants’ reflexivity; (c) young peoples’ perspectives on making plans; (d) survival-oriented reflexivity; (e) study limitations; (f) practice relevance.
Reflections on Archer’s theory
Archer’s theory guided our analysis, but we were struck by a number of inductive findings that, if confirmed by further studies, may lead to revision of Archer’s model. First, in what might be thought of as co-existing forms of practical reason: the participants who predominantly experienced ‘active agency’ also experienced fractured reflexivity; the participants who predominantly experienced fractured reflexivity also found areas of their lives in which they had experienced what seemed to us like ‘green shoots’ of thoughtful, active, and rich internal conversations; and those with survival-oriented reflexivity experienced rich internal conversation, but also specific forms of ‘low-horizon’ reflexivity focused on ‘survival’ (i.e. from day to day) and ‘safety’. In summary, we did not find discrete or categoric forms of reflexivity in this study, i.e. perhaps not as stable as suggested by Archer’s(2003)framework.
Second, the two participants experiencing clear active agency did not seem to specialise fully in particular ‘modes’, such as communicative, autonomous, or meta-reflexive, as described by Archer (2003, 2007). However, both participants were strongly ‘engaged’ in society, had ‘a conception of the good’ (Nussbaum, 2011), were committed to social values and had strong bonds with others (Donati, 2013) – could we regard these emergent forms of active reflexivity as partially meta-reflexive, but in the specific context of autobiographies of social exclusion?
Third, Archer regards fractured reflexivity as a temporary state: “powers of reflexivity have been suspended” (2003: 298); and this state “can be overcome given the appropriate relational conditions” (2012: 290). However, some of our participants may have experienced elements of fractured reflexivity for some considerable time. For some individuals this stance may have been maintained by unresolved mental health symptoms, and/or repeated structural constraints and losses, with compounded effects on concerns and plans (Brown, 2002). Our findings on fractured reflexivity and survival-oriented reflexivity offer empirical data in a context of long-standing adversity, i.e. different from some of the contextual assumptions built into Archer’s primary theory.
The analysis of participants’ reflexivity
The open-ended interview used by Archer in her empirical studies of internal conversation (Archer, 2003: 161–162), and by us in the present study, allowed us to develop detailed discussions with individual participants about their internal dialogues. The extended interview we used in this study (two interviews, rather than Archer’s one interview) did perhaps allow most of the young people to ‘find’ areas of their lives in which they experienced rich internal dialogue. We would emphasise the possible methodological importance of this extended opportunity for discussion of complex and sometimes troubling personal concerns, and of course the possibility that patterns of reflexivity may emerge differently with the benefit of additional interviews.
We asked participants to reflect, during the interviews, on their own recalled internal conversations – this raises a number of methodological and theoretical questions. First, the interviewer asks participants to be reflexive about being reflexive. We have assumed, as perhaps Archer does, that the more experience a person has of being reflexive, then the more able she or he will be to communicate reflexively about internal conversations in a facilitative interview situation. This assumption perhaps needs testing ethnographically, giving participants more opportunities to be actively reflexive, i.e. outside an interview situation. Second, a concern arose during the study (and was picked up by an anonymous reviewer) that a certain level of reflexivity is presupposed by the demand for recall of internal conversations, and therefore surely a participant’s accounts could not validly be characterised as ‘purely’ fractured reflexivity. Archer addresses this by her definition of fractured reflexivity, which focuses on lack of (reported) ‘concerns’, failure to plan, and apparent lack of efficacious internal conversations (Archer, 2003). However, people experiencing fractured reflexivity are indeed likely to be at a relative disadvantage in an interview situation (and in a planning dialogue with practitioners?), compared with peers experiencing active agency. Our use of two interviews (rather than Archer’s one interview) may have been helpful in eliciting the ‘green shoots’ of reported internal conversations about some topics in an otherwise fractured reflexivity scenario.
Planning
Many of the young people were deeply unenthusiastic about planning their lives – a topic that is a key part of the research interview. Scepticism about planning was evident for those experiencing fractured reflexivity, survival-oriented reflexivity, and for one of the (otherwise) actively agential young people’s thinking. This was despite having had access to a responsive local service, with continuity of individual professional help, including regular life planning opportunities.
Disruption of previous plans was frequently discussed by participants, ranging from specific childhood trauma experiences (e.g. creative and systematic work by the child being physically destroyed by a parent), to sudden changes of foster placements, to recent disappointing and fruitless attempts to engage with work or education.
We would urge further research into young peoples’ subjective interpretations of ‘planning’. In the meantime, this finding may have relevance to practice-based work with young people, which, for instance in the UK, seeks to engage young people leaving care through ‘pathway planning’ (Dixon and Robey, 2014; Stein, 2012). This is discussed further below.
Survival-oriented reflexivity
Four participants experienced a form of reflexivity which did not fit the reflexivity categories developed by Archer, and emerged inductively from our analyses. Internal conversations were rich, thoughtful and reflective. However, the stance was fractured, in Archer’s sense of the person not having confidence to imagine, plan and dovetail future-oriented projects. Internal conversations were described as focusing on day-to-day survival and/or profound self-reliance. This stance was spontaneously justified by detailed accounts of family, foster-care (and sometimes institutional) experiences in which strict self-reliance (i.e. for day-to-day survival) appeared to be the only appropriate strategy.
Our findings on survival-oriented reflexivity provide an independent confirmation of US data reported by Kools (1999) and by Samuels and Pryce (2008). They describe “survivalist self-reliance” in some young adults aging out of foster care. We were not aware of their study when we conducted our analyses. The emphasis in the US findings of young people feeling responsible for their own development and safety / security fits completely with our London data. The construct is also evident in Stein’s ‘survivor’ group (Stein, 2006: 277).
The distinctive and new message from our own data on individuals with this (survival-oriented) stance is that the active use of internal conversation, at least in our sample, seems to be rich and thoughtful. Very detailed accounts of specific autobiographical episodes in the past, linked to a narrative about how the person conducts him or herself from day to day, and for the future, characterised the accounts of these young people.
Although fascinatingly meta-reflexive in an autobiographical sense, the narratives of these individuals did not meet Archer’s meta-reflexivity criteria of being critical of effective action in society (Archer, 2003), preoccupied by moral issues (Archer, 2007: 127–132), and critical of market and state (Archer, 2012). However, the strong autobiographical justifications for ‘low-horizon’ stances may be regarded as based on repeated real-life experience of structural constraints, and in this sense carrying a thoughtful and reflexive summary of developmental experience.
Limitations
Sampling was for heterogeneity, but within the inevitable biases of a relatively well-resourced public service. We undoubtedly missed hard-to-reach young people, and we might have missed some who were so successful that they did not need a professional service. It would be useful to repeat the study with an especially hard-to-reach group.
Our sample was small, but the epistemological grounding and analysis method focused on in-depth work with individual recordings and transcripts, drawing out the particular voices of each participant (Archer, 2003; Smith, 2011; Smith et al., 2009). While this may be a limitation in respect of ‘breadth’, our aims were to maximise the potential for understanding individuals in their own contexts, and making full use of interconnections within individual transcripts (Packer, 2011, chapter 3).
We relied almost wholly on participants’ accounts. The study would have been very significantly improved if information had also been gathered from informal and formal social network members and from case-notes. Some key-workers had known clients for some years, but we were not able to use information from these sources.
The data were not longitudinal. For young people who are beginning to translate agential concerns into projects, under conditions of social exclusion, data on the intricacies of interplay between individuals and linked sequences of situations over a specific period would be informative, and would go beyond internal conversations to include real-world interactions.
Practice relevance
Bearing in mind the limitations, we outline some suggested directions of practice relevance, each of which would require further empirical research, and would require resourcing.
Internal conversations are regarded in this research as mediating between structure and agency (Archer, 2000, 2003, 2007, 2012), and are therefore a potential focus for collaborative work with young people who have been subjected to corrosive disadvantage, yet who have an adult life ahead of them requiring thoughtful planning (Nussbaum, 2011).
Our first point begins with the majority of participants in this study for whom life-planning is anathema. We were struck by the emotional force of many participants’ views on planning (given with justifications). Our data suggest that professionals and network members bear in mind the deep scepticism of some young people about the implicit assumptions underlying discussions that presuppose that planning is a useful joint activity. The phrase ‘lost in translation’ comes to mind. Both the young people who experienced fractured reflexivity, and those who had a survival-oriented stance, did not wish to look ahead too far. This point has relevance for individual discussions in health contexts (e.g. medical care, including prescription and non-prescription drugs), housing and homelessness contexts (e.g. discussing future plans), and education contexts (e.g. discussing initial plans, organising assignments, discussing detailed goals), and of course raises questions about how pathway planning (Dixon and Robey, 2014; Stein, 2012) is conducted. Pathway planning may require much deeper recognition of both the reflexive ‘starting position’ taken by individual young people, and the intricate detail of support that may be necessary during specific engagements by the young person with ‘planning and doing’ particular projects in an environment that will inevitably be more than challenging.
Our second point begins with the research interview framework. As clinicians we were struck by its rich possibilities. Perhaps, in suitably modified form, the framework might provide a ‘thread’ of discussion with young people during the process of pathway planning. We are not suggesting a formal interview protocol or ‘tool’, which would run the risk of replicating bureaucratic procedures (Stein, 2012). Discussions about internal conversations might provide an opportunity to map, in two-way discussion, the young person’s approach to practical day-to-day living, and crucially, his or her social network (Blakeslee, 2011). If a nascent mode of reflexivity becomes evident, this would provide the practitioner with a potential ‘space’ to help the individual client advance their internal conversations and plans – for instance a communicatively-oriented young person might be helped to build on this ‘networking’ approach while navigating structural barriers. The notion of internal conversation, including imagination, reliving and clarifying, opens up possibilities for conversations that do not routinely focus on ‘formal’ planning. ‘Green shoots’ of potential sources of active agency in our participants’ thinking was evident for all individuals. Further research on ‘green shoots’ within the experience of individuals struggling with fractured reflexivity should be of benefit to those providing interventions.
Conclusion
Martha Nussbaum’s (2011) definition of practical reason, with which we began this paper, emphasises not only the planning of one’s life but also ‘critical reflection about the planning of one’s life’ (italics added). Participants in the study reported here provided detailed accounts of their own thinking about internal conversations, including planning, in contexts of severe biographical adversity. It is hoped that the study can inform policy and practice in collaborative work on pathway planning with young people leaving state care.
Footnotes
Acknowledgements
We are grateful to all participants and staff members at the research site, to Margaret Archer for her encouragement and feedback at all stages of the research, and to Oliver Robinson, Susan McPherson, Miriam Richardson and Zachari Duncalf. We also thank anonymous reviewers and Karen Broadhurst for thoughtful feedback which led to considerable improvement of the paper.