Ageing in Place Over Time: The Making and Unmaking of Home

‘Ageing in place’ is a key component of UK policy, aimed at supporting older people to remain living in their own homes and communities for as long as possible. Although wide-ranging, the scholarly literature in this field has not sufficiently examined the interconnections between ageing in place and the changing experience of ‘home’ over time. This article addresses this gap in a novel way by bringing together qualitative secondary analysis of longitudinal data with critical literature on ‘home’ and Mason’s cutting-edge concept of ‘affinities’ to understand the multi-dimensionality of home in relation to ageing in place. The article makes significant methodological, empirical, and theoretical contributions to the field of scholarship on home, by demonstrating how homes are made and unmade over time. Discussions of home emerged organically in the longitudinal data that focused on people’s travel and transport use, allowing our qualitative secondary analysis approach to look anew at how experiences of home are dynamically shaped by people’s potent connections inside and outside the dwelling. Presenting an empirical analysis of four case studies, the article suggests that future discussions in the field of ageing in place should pay closer attention to the factors that shape experiences of the un/making of home over time, such as how deteriorating physical and mental health can shape how people experience their dwelling and neighbourhood as well as their relationships across these settings.


Introduction
Ageing in place policies support older people to remain living in their own homes for as long as possible (Grimmer et al., 2015), encouraging community support and informal care within the existing neighbourhood in an attempt to lower the costs of health and social care (World Health Organization (WHO), 2015). Having gained momentum since the WHO launched its ' Age-Friendly Cities Project' in 2007(WHO, 2007, such policies have become widespread in many countries, including the UK (WHO, 2015). The WHO introduced the Global Network of Age-Friendly Cities and Communities in 2010 to encourage the implementation of policy programmes in member cities, of which there are more than 900 across the world.
While 'ageing in place' is often equated with 'ageing at home', because the home is seen as the ideal place to live in later life (Boyle et al., 2015;Pani-Harreman et al., 2020;Stones and Gullifer, 2016), ageing in place research acknowledges that 'place' refers to both the home and the wider neighbourhood (Dobner et al., 2016;Sixsmith and Sixsmith, 2008;Wiles et al., 2012). The ageing in place model is built on the assumption that the longer a person has lived in an area, the stronger their sense of home and connection to place (Cutchin, 2003;Gilleard et al., 2007). Such assumptions have been supported by Rowles' (1978Rowles' ( , 1983 argument that a sense of 'insideness' with place is acquired and strengthened over time through spatial routines and habits. In contrast, institutional settings such as care homes are commonly regarded as 'unhomely' (Stones and Gullifer, 2016) partly due to the short time that most residents have spent living in them (Lovatt, 2018). Building on Rowles (1978Rowles ( , 1983, Stones and Gullifer (2016) suggest that for older adults who have lived in the same house or neighbourhood for most of their lives, the experience of 'home' is found in the connection between the dwelling and the wider social community (p. 453).
Some research in this field has called for a more critical understanding of the meaning of home (Jarvis and Mountain, 2021), highlighting a range of factors that affect experiences of the home, such as the loss of a partner, friends moving away, worsening mobility, or inadequate housing (Hillcoat-Nallétamby and Ogg, 2014;Means, 2007;Penney, 2013;Wiles et al., 2012). We contend that conceptualisations of ageing in place do not look at the dynamics of the relationship between home and neighbourhood in sufficient detail and thereby imply a static reading of home. To fill this gap, this article asks, 'How does the experience of "home" change over time in the context of ageing in place?' Drawing on literature discussed below, we conceptualise home as not spatially bounded but involving the interrelationship between the dwelling and its surrounding material, social, and relational environment. We furthermore demonstrate that a home is not fixed but is actively 'made' and can become 'unmade' (Baxter and Brickell, 2014;Sou and Webber, 2019;Visser, 2019).
Our empirical material comprises qualitative secondary analysis (QSA) of longitudinal data from an ESRC study of people's mobility and everyday lives, conducted over a 3-year period (Miles et al., 2018). Our analysis of four case studies brings together longitudinal QSA and Mason's (2018) affinities approach, arriving at two central arguments. First, conceptualising the home as porous (Blunt and Dowling, 2006), we use Mason's theorisation of affinities to examine the potent connections within and beyond the dwelling that contribute to people's sense of home. We thus extend existing work on ageing in place by arguing that 'home' and 'neighbourhood' are best understood not as material territories that contain people's lives, but more like focal points where affinities entangle. Second, we build on challenges to the idea that a sense of home strengthens over time (Rowles, 1978(Rowles, , 1983. We advance the argument that not only can home be experienced as both a 'haven' and a 'prison' but can simultaneously be made and unmade (Baxter and Brickell, 2014;Jarvis and Mountain, 2021;Mason, 2018). This allows us to shed light on the simultaneity of enjoyable and toxic experiences of home in the context of ageing in place.
The article is structured as follows. We first outline the theoretical framework of the article, building on critical approaches to home and Mason's (2018) theorisation of affinities as a way of conceptualising home as porous and dynamic. The methodological approach used in the article follows. We present our findings in two sections, discussing four cases to highlight the implications of continued or shrinking connections beyond the house on the changing experience of home over time. The discussion and conclusion section discusses the methodological, empirical, and theoretical contributions made in the article, arguing that a longitudinal QSA of affinities offers a novel contribution to understanding the complex and sometimes contradictory nature of the home in the context of ageing in place.

Making and unmaking home
While the WHO's (2015) guidelines (p. 136) suggest that the preference of many older people is to remain in their own dwellings, they also acknowledge the need to ensure that people age in a place that is 'right' for them. Thus, while the bricks and mortar of the home may be suitable structurally, a sense of home, in the form of ontological security, safety, and comfort may be lacking (Means, 2007;Wiles, 2005;Wiles et al., 2012). Definitions of home within the literature are wide-ranging and disputed (Blunt and Dowling, 2006), with Meers (2021) defining it as an 'essentially contested concept'. Much critical sociological literature on the home, particularly from a feminist perspective, challenges the notion of home as a 'haven', pointing to the ways in which factors such as domestic violence can seriously undermine any of the more positive qualities of the home, such as agency, security, and comfort (Barratt and Green, 2017;Baxter and Brickell, 2014;Blunt and Dowling, 2006;Mallett, 2004). Thus, although remaining in one's home may have some positive benefits such as continued familiarity or remaining close to existing social ties, the experience of home may also change in negative ways in later life. For example, the death of a spouse could lead to the home becoming a place of increased isolation (Jarvis and Mountain, 2021;Wiles et al., 2012). Drawing on the critical literature on the home, we conceptualise the home as more than a fixed physical backdrop. Feeling at home is a dynamic experience that requires work and involves many factors: Home is not a static unchanging concept, but a process that requires work . . . simply being inside one's dwelling does not automatically make it feel like home. Home is the complex interplay of space, relationships, the body, and time. (Visser, 2019: 7) The literature on home also shows that just as home can be made, so too can it be unmade. The phrase 'unmaking of home' was coined by Baxter and Brickell (2014), who defined it as 'the precarious process by which material and/or imaginary components of home are unintentionally or deliberately, temporarily or permanently, divested, damaged or even destroyed' (p. 134). They stress that this unmaking can occur in the banal and every day, not only in extraordinary circumstances such as the destruction of a home (Porteous and Smith, 2001). Baxter and Brickell (2014) were responding to the overwhelming focus on homemaking and argued that by being attuned to the ways in which homes are unmade, also in the context of mundane practices, scholars could better appreciate the 'dynamic and varied' nature of domestic lives (p. 135).
We extend these insights by adopting Mason's (2018) innovative theoretical work on affinities as a lens through which to advance understandings of how people experience the making and unmaking of home. Mason (2018) defines affinities as 'potent connections' that are not 'static relationships between fixed entities' but instead should be conceived of as 'connective charges, forces and flows' (p. 123) between people and their environments. Shifting the focus from spatially bounded homes and neighbourhoods to 'the dynamics of connection' (p. 180) means that home and neighbourhood are no longer conceived of as a context or a fixed 'material territory' (p. 168). Furthermore, just as Baxter and Brickell (2014) argue that homemaking and unmaking can occur simultaneously, Mason (2018) explains that affinities are not necessarily 'inherently positive and enjoyable' but can also be 'toxic and fearful' (p. 186).
We examine home through the lens of affinities in conjunction with QSA of longitudinal data where experiences of home emerged unprompted within narratives about relationships to place. The analysis shows how multiple and even contradictory experiences can be given to home, and how these are not only contained within the dwelling but are felt as potent connections across a range of domains and relationships. In other words, as observed by several scholars, constructing a sense of home involves different scales, such as the neighbourhood and the city (Ahmed, 1999;Blunt and Dowling, 2006;Hammond, 2004;Hillcoat-Nallétamby and Ogg, 2014;Massey, 1994). In response to Jarvis and Mountain's (2021) call to attend to the interrelationships between time and space in resisting idealisations of home, we explore how the potent connections that help make and unmake home change and shift over time, fluctuating between 'enjoyable' and 'toxic' (Mason, 2018: 186). In doing so, we build on May and Muir's (2015) argument that changes in social networks and the availability of social support affect a sense of belonging to place over time. More specifically, we focus on how people experience potent connections in the context of their expanding or shrinking daily lives, as their routines and forms of sociability in and outside the home change over time and how this, in turn, influences how they experience their dwelling.
In exploring such dynamic connectivities, our article responds to Boyle et al.'s (2015) call for 'a more holistic understanding of neighbourhoods as . . . social spaces of ageing' as a way to 'better understand and support ageing in place' (p. 1503). Influenced by Massey's work, we conceptualise neighbourhoods as a 'constellation of social ties' (Massey, 1994: 154). Using an affinities approach, we push this analysis further, exploring the potency of relational connections that are entwined with the material and temporal experiences of place, all of which help form what constitutes a 'neighbourhood' and its atmosphere. We now turn to the study on which our analysis is based, discussing how our longitudinal QSA and case-study approach allows for a holistic analysis of older people's changing sense of home over time.

The study
This article draws on detailed QSA of longitudinal data archived with Timescapes: An ESRC Qualitative Longitudinal Initiative. The data were collected as part of the 'Step Change' project which began in 2010 and ended in 2013, involving a qualitative longitudinal panel study with interviews at yearly intervals with participants living in the UK city-region of Greater Manchester. Participants were asked about their everyday lives, travel, and transport use as well as changes to their neighbourhood, health, sense of belonging, and social networks (Miles et al., 2018). The rich content of the interviews and the fact that they offered extensive exploration of older participants' experiences of place prompted our interest to explore the data further through the lens of ageing. Although almost a decade has passed since the data were collected, the involvement of one of the authors of this article, Camilla Lewis, in the original research allowed for both 'proximate knowledge of the data and a form of critical distance' (Irwin, 2013: 298). Analysing data at a 'remove' in this way can also provide opportunities for 'other kinds of insight, engagement and analytical apprehension' (Hughes et al., 2021: 4). Being able to bring insights from re-analysis into conversation with theories of home unmaking developed since the completion of the initial research project allows us to 'ask new questions of old data in order to purposively disrupt embedded epistemologies' (Dodds et al., 2020: 272).
The four cases analysed in this article derive from our broader study of how ageing in place is experienced in different types of neighbourhood. To this end, we decided to focus on four neighbourhoods in Greater Manchester with different socio-economic profiles; however, the focus of this article is to explore the ways that the experience of home changes over time. As such, socio-economic differences are not systematically considered across the four cases (see Lewis and Buffel, 2020, for further discussion on the impact of neighbourhood inequalities on ageing in place). Our sample of 28 participants was derived by selecting from the 'Step Change' dataset those participants who met the following criteria: they lived in one of four chosen neighbourhoods; they were over the age of 50 years, reflecting the fact that in the more deprived areas of the city, life expectancy is lower than the national average (Bullen et al., 2016); and they took part in all three waves of the study.
We carried out secondary analysis of the data, exploring what they could tell us about experiences of ageing in place over time. The interview transcripts were analysed using NVivo, with a coding framework developed from key themes in the literature and recurrent themes within the data, including: place, community-belonging, physical and mental health, social networks, and temporality. Although participants were not directly interviewed about their experiences of home, the topic emerged unprompted in the data. Furthermore, the participants discussed their homes in relation to age-related factors such as their levels of mobility and how their relationship to their neighbourhoods had changed as they got older. In other words, the participants experienced their homes not as bounded, but in relation to the wider environment. Because the experience of home in later life has not previously been examined with longitudinal qualitative data and because longitudinal data illustrate change over time (Sou and Webber, 2019;Vogl et al., 2018), we decided to focus this article on the dynamic nature of home as it relates to ageing place.
To be able to focus in detail on the interlinked processes of ageing in place, home, and physical ageing, we decided to select a sample that was homogeneous in relation to the following key factors. In relation to ageing in place, we selected those participants who had lived in the same area for most of their lives. In relation to home, we included participants who had a strong desire to remain living in their homes. In relation to physical ageing, we selected those participants whose diminishing health and mobility meant that they spent increasing amounts of time in their homes over the three interviews. This process of selection resulted in four cases that met all three criteria: Jean, Violet, Sandra and John, whose names have been changed for the purposes of anonymity.
Our analysis focused on how and why the participants' experiences of home changed over time and how their understandings of home related to ageing in place. Thanks to the longitudinal nature of the data, the case studies allowed us to explore the impact of specific life events on the changing experience of home over time (Feagin et al., 1991;Jarvis and Mountain, 2021).
The re-analysis of qualitative data is cause for debate. Some express epistemological concerns of a loss of both the context in which the data were collected (Mauthner et al., 1998) and relationships between researcher and research participants (Heaton, 2004). As noted above, the involvement of one of the authors in the original study means the loss of relationships is not a prominent issue in our analysis. Furthermore, our analysis pays close attention to the biographical, local, and social contexts of our four participants, and we understand these contexts not as 'fixed, separate, exterior, containers' but as 'the dynamic relational nexuses of the broader social worlds of which they form an integral part' (Hughes et al., 2021: 3). In this way, we view the data as a resource in itself, existing, as Hammersley (2010) notes, 'independently from the research project' (p. 5.2). We also query to what extent our secondary use of the Step Change data entails a loss of the relationship between researcher and research participant, given that the interviews were conducted by several researchers (see Bishop, 2007;Hughes et al., 2020).
Our approach to secondary analysis draws from Moore's (2007) observation that it is not fruitful to draw a clear distinction between the 'use' and 're-use' of data, given that both entail a process of 'reordering' that helps reveal new dimensions of the data. We argue that when good justification is offered for why the data are suitable to be put to 'new uses beyond those underpinning their original generation' (Hughes et al., 2020: 568), as we have done above, careful secondary analysis is justified.
We first present a discussion of Jean and Violet, whose narratives of homemaking emphasised the role of continued engagement with social ties and activities within and beyond the home, through both local social connections and activities within the home (Baxter and Brickell, 2014;Hammond, 2004;Massey, 1994). We then move on to discuss John and Sandra, for whom a narrative of unmaking home and a decreasingly porous sense of home was evident due to dwindling affinities and feelings of isolation within the home. Finally, we discuss the four cases together, highlighting the factors that contribute to the un/making of home in the context of ageing in place.

Entanglements of house, neighbourhood, and social networks
The cases of Jean and Violet presented in this section highlight the role of strong affinities in the form of an enduring sense of home that stretches beyond the dwelling to include the wider neighbourhood, even though both women spent increasing amounts of time at home.

Jean
Jean was white, aged 58 at the time of the first interview and lived in a relatively affluent neighbourhood with a high proportion of owner-occupied properties, with her husband, in a house which they owned. She had lived in the same neighbourhood since she was a child. Jean had retired due to ill-health the year before the first interview from her longterm job as an administrator in a bank. Jean's sense of home was deeply connected to the wider neighbourhood, extending beyond her house (Hammond, 2004;Massey, 1994). When asked if she felt a sense of belonging to the area, Jean replied:, 'Very much so, especially this area, this road, it's very much a village feel . . . It's a long road and we all know one another'. Having lived in the same neighbourhood for most of her life, Jean had cultivated long-standing connections with other residents by regularly 'being out, seeing people, talking to them'. She also recounted frequents acts of neighbourliness, including feeding her neighbours' cats when they were away and calling into her neighbours' homes for cups of tea. Jean's friendly relationships with those living on her street highlight the porosity of home whereby the connective charges that contribute to a sense of home are not contained within the dwelling (see Mason, 2018). In Jean's case, the development of affinities was not solely the outcome of the passing of time, but also an active and ongoing process of social engagement in the neighbourhood that requires work (see Visser, 2019: 7).
Jean spent increasing amounts of time at home over the 3-year duration of the study due to changes in her and her husband's health, which restricted their mobility. In the second interview, Jean said, 'What has changed? Two years ago, my husband became very ill. It's just pulled us in a bit. So we tend to stay more at home, which is quite nice'. While she remained engaged with the wider neighbourhood, Jean placed greater importance on ensuring her house felt 'like a home; homely and welcoming', for example, by decorating rooms and gardening. These activities allowed Jean to exercise control within her home (Douglas, 1991) and to continue to feel a positive affinity with the space.
The longitudinal data showed that Jean worked to balance the seemingly contradictory forces of 'pulling in' and connecting beyond the home to ensure that her decreased mobility did not result in a restrictive sense of home (Mason, 2018;Visser, 2019). She continued to maintain, as best she could, the social relationships she had built up over time. These helped support a sense of home which felt positive. Jean's case illustrates well Mason's (2018) assertion that affinities are not 'static relationships between fixed entities' (p. 123) and that the potency of these is not contained in particular spaces. Violet's case further develops this discussion of the porous nature of home in the context of decreased mobility, in the absence of a spouse.

Violet
Violet, a white woman, was 77 at the time of the first interview and lived in a flat which she owned in sheltered accommodation in a demographically similar neighbourhood to Jean, although with a slightly older population than the national average, where she had spent most of her life. She lived alone following the death of her husband 10 years before. She had struggled with physical health problems throughout her life and had moved to sheltered accommodation with her husband 20 years earlier due to a serious illness. During her adult life, Violet had worked in the newsagents which she and her husband owned while bringing up her children and caring for her parents, who lived with her family towards the end of their lives. She had two daughters, one who lived locally and one in another region, whom Violet saw more frequently.
Although Violet struggled with declining physical health over the three interviews, she remained engaged in activities and relationships beyond the home, regularly visiting her local church, and travelling to see her daughter who lived a 3-hour drive away. Working in her family's newsagents and attending the same church throughout her life contributed to a deep sense of belonging to the local area. She explained, 'you do feel part of those communities'. In the first interview, she explained how she made the decision to move into a flat in sheltered accommodation with her late husband: 'we'd more or less ruled out that we wouldn't move away from the area because we felt that friends were an essential part of life for us'. Violet's desire to age in place was influenced by the potent connections she had developed with people living in the local area over time. The longitudinal nature of our data, including the life histories participants recounted in the first interview, means that our analysis can retain 'the holistic and meaningful characteristics of real-life events' (Yin, 1989: 14). We are, for example, able to understand how people adapt to specific events, such as bereavement or change in health, and how this is reflected in their changing experience of home over time.
In the second interview, Violet was asked what she would miss if she were to move. She replied, 'the people who live here and friends, health facilities. And I think the effort you would have to make to settle in a new community'. For Violet, her connective charges with the local neighbourhood as well as her relationships with other residents were entangled. Her sense of belonging to her home was reinforced not only by her sense of belonging to the wider neighbourhood, but was also strongly connected to her sense of community within the sheltered accommodation. She said, 'there's a certain amount of social activity within the building which you wouldn't get in a lone house, which is good'. Violet's case illustrates how a move from one house to another does not necessarily always disrupt a sense of home, rather, living in sheltered accommodation offered Violet the possibility of greater sociability. While she lived alone and missed 'having somebody to socially react with at home and to talk to' following her husband's death, there were opportunities to socialise with others living in the building, which she stressed would not exist in a 'lone' house.
Violet spent increasing amounts of time in her flat over the 3-year duration of the study due to reduced mobility, as she recovered from a broken leg after a fall in between the first and second interviews. In the second interview, she described the various hobbies she engaged with in her flat, including playing the piano and making gift cards. But Violet also continued to feel connected to those living around her. Describing the period of recovery following her injury, she recounted how 'everybody rallied around . . . brought me food for a couple of weeks and did shopping for me'. This informal support that she received was in part a result of the work she had put into cultivating these relationships. Violet was actively engaged in the community there, applying for funding for the communal garden, and leading the redesign of the gardens, working to get consensus from residents who were, as she described, 'not 27 likeminded people'. Despite reduced mobility outside the home, the length of time she had lived in the area (Rowles, 1983), combined with the enduring nature of the potent connections she had within and beyond the sheltered accommodation (Mason, 2018), ensured that Violet did not become isolated.
In contrast to Jean's and Violet's experience of maintaining positive affinities with both their homes and their neighbourhoods despite declining health and mobility, we now move on to discuss the cases of Sandra and John, whose shrinking social networks and declining health formed a central component of unmaking of home, whereby both came to feel increasingly isolated within their homes.

Shrinking social networks and a diminishing sense of home
Sandra's and John's accounts highlight the way that 'toxic' affinities (Mason, 2018) can affect the porosity of the home. Sandra's and John's engagement in social relationships and activities beyond the home in the form of volunteering and socialising in pubs dwindled over the duration of the research. Both participants expressed an increasing experience of their homes as a site of isolation as the ties between their homes and the wider neighbourhood weakened.

Sandra
Sandra was white and aged 61 at the time of the first interview, and had lived alone in a rented council house for 22 years in the same neighbourhood as Jean, where she had lived for most of her life. She had worked as a part-time secretary for short periods of time while bringing up her children. Sandra's three adult daughters lived locally, and while she described her relationship with them as close, she did not see them as often as she would have liked.
Sandra's sense of home became less connected to the neighbourhood over time. Over the three interviews, she spent more time in her house, rarely seeing friends with whom she had regularly socialised in pubs and bars in Chorlton in the past. She described her relationship with the neighbourhood as strained, explaining how she disliked both her next-door neighbours, and only had one neighbour whom she could turn to for help. Sandra's main connection beyond the home came in the form of regularly volunteering at a museum in Manchester. By the time of the second interview, Sandra had stopped volunteering to care for her ex-husband. After his death, she had not returned to her role, describing a sense of despondency: 'it's because of me taking time out for [former husband] and then afterwards it was the loss'. When asked in the second interview if there had been any changes in the places she regularly visited, Sandra responded, 'I've not been going as many places as I did . . . sometimes I've really not been going out the door which is not good 'cause I'm a very sociable person'. The implications of this decreased activity were significant for Sandra, resulting in an altered experience of time spent alone at home as underlined by a sense of isolation. In contrast to the first interview, where she expressed a more positive experience of home, her later evaluation of spending time at home as 'very boring' and her emphasis that she should 'get out more' can be read in the context of her reduced capacity to socialise and her felt distance to her social and familial contacts.
The longitudinal analysis reveals that as her affinities with people and place diminished, Sandra's experience of home at the beginning and end of the study stood in stark contrast. In the first interview, she said, 'I'm a bit of a loner really, I'm quite happy with my own company'. However, by the second interview, she felt more cut off due to changes in her physical and mental health which meant that she spent most of her time at home and no longer invited friends around. Sandra attributed these changes to depression, which losing her ex-husband had contributed to. By the third interview, she described feelings of isolation, her three daughters being the only people she saw. Sandra said of her home as a social space: 'I used to [have friends round to her house] but I don't have any more', emphasising later on in the interview: 'I've got no friends at the moment'.
Sandra's reflections on her changing experience within the home, from one of chosen solitude to isolation reveals the unmaking of home through the damage to the material and emotional components of home, in two ways. First, because of her ex-husband's increasing care needs, she felt unable to continue volunteering. The loss of this connection in turn led to an unmaking of her sense of 'insideness' (Rowles, 1983) in the wider context of place. Second, this reduction in activities and engagements beyond the home meant Sandra's experience of home became unwoven from the potent connections that had previously contributed to her sense of 'insideness' (Mason, 2018). John's case further illuminates how the disappearance of social ties within and beyond the home can lead to the home becoming a site of isolation as his 'spatial horizons recede' (Jarvis and Mountain, 2021: 12).

John
John, aged 57 at the start of the study, was white, and had lived in an inner-city neighbourhood with high levels of deprivation alongside rapid gentrification. He had lived alone in sheltered accommodation run by a social housing provider for 5 years and had lived in the area for most of his life. He had taken early retirement due to ill-health several years before, after working in a range of short-term, precarious jobs. His physical and mental health deteriorated over the three interviews. Consequently, although John's sense of belonging to his local area was strong, partly due to the length of time he had lived there, he spent much of his time in his flat over the three interviews, rarely seeing his two adult sons, although they lived close by.
Like Sandra, in the first interview John expressed a desire for solitude and described his flat as a space of safety that provided him with everything he needed where he could do 'what I want, when I want'. There was thus at least initially a strong element of homemaking present in how John spoke about his flat. However, the tone changed over the duration of the research as John's poor health restricted his mobility and the amount of time he was able to spend outside his flat, which limited his already reduced social interactions. The connection between deteriorating physical and mental health and social isolation has been well documented by previous research (Jarvis and Mountain, 2021;Shankar et al., 2017). By the second interview, John was reflecting on how 'meaningless' he felt his life was, given how little he did with his time both inside and outside his home. This, coupled with the dwindling visits from his children and friends, contributed to his home becoming less porous and more isolating. He said to the interviewer in the second interview: 'it's why I'm going to be depressed when you go, I don't really do anything'. By the third interview, John no longer saw his friend or his brother, two of the few people he most regularly saw at the time of the first interview.
How John spoke about his neighbours also changed across the three interviews. In the first interview, he referred to the block of flats as 'a community in itself', albeit one that respected his need for solitude: 'if you want to be a miserable old so and so and shut your door and not bother anyone, they'll leave you alone'. While there were amenities and activities available in the sheltered accommodation, John preferred to be in his flat. In the first interview, he said, 'I do occasionally get a dinner and bring it up because you can either eat it downstairs or, you know, you can put it on a tray and bring it up'. By the last interview, he no longer mentioned any sense of community, instead referring to his neighbours as 'just people I know, aren't they. They're not important to me'. Unlike Violet, John did not talk about informal support offered by his neighbours, for instance with carrying shopping bags up the stairs, which he described as a challenge due to his declining physical health.
John's relationship with the only local pub was one of the remaining significant places that connected him to the wider neighbourhood, and the pub acted as an extension of his home. Across all three interviews, he talked frequently about going to the pub to collect food and drinks to take home and enjoy in front of the television: 'I just go to this pub across the road here, she [the pub owner] lets me bring a pint home'. John explained how his old friends largely drank elsewhere -so the pub was no longer a particularly important site of social interaction but occasionally he enjoyed socialising with postal workers who drank there after their shift at the nearby sorting office. Thus, although John did engage with the surrounding environment (Hammond, 2004), it was clear that he did so only briefly, preferring to spend time in his flat. The repeat interviews demonstrate that while John's sense of belonging to the wider area did not hinge on close social relationships (May and Muir, 2015), the porosity of his home reduced over time, with only the pub and the block of flats remaining an important, if shallow, source of connection (Mason, 2018).
John's case illustrates that affinities can somewhat paradoxically be experienced as simultaneously uncomfortable and enjoyable (Mason, 2018). While John's sense of isolation grew stronger over the duration of the research, it was nevertheless underpinned by his preference for solitude that led him to voluntarily retreat from communal spaces and opportunities for social contact. But he also compared his present solitary existence with the sense of community he felt in his previous home: 'the sense of community, oh it was fantastic, I think they took to me 'cause I was a single parent as well'. This suggests that on some level, John yearned for a stronger sense of commonality with neighbours, further emphasising the sense of unmaking of home that he was currently experiencing.
Below we discuss the four cases together and conclude by arguing that ageing in place research must pay more attention to the experiences of potent connections between home and the wider neighbourhood, given how much this can reveal about how ageing in place may be experienced.

Discussion and concluding remarks
This article has presented a novel analysis of the changing experience of 'home' in later life. Analysing longitudinal data of four cases through the theoretical lens of affinities has allowed us to examine in detail how homes are made and unmade over time in the context of ageing in place. Analysing un/making of home over time reveals that ageing in place must be viewed as a dynamic and multi-dimensional experience, shaped by changes over the life course. The analysis presented in the article has made two arguments in response to the questions: 'How does the experience of home change over time?' and 'How do understandings of home relate to ageing in place?'.
First, we challenge the notion of home as haven which underpins ageing in place policies, by showing that home can be experienced as both a haven and a site of social isolation and that people's sense of home can simultaneously be made and unmade (Baxter and Brickell, 2014;Jarvis and Mountain, 2021). Over the 3-year period of the study, all four participants spent increasing amounts of time at home due to worsening health. However, as the analysis has shown, the experience of home changed in different ways. Jean and Violet continued their homemaking though their 'activities and interests' (Stones and Gullifer, 2016) within their dwellings (e.g. redecorating rooms, playing piano, and making gift cards) and in the wider community. Home unmaking was a more prevalent theme in Sandra's and John's experiences of ageing in place, due to their dwindling social ties, decreased engagement in activities beyond the home, and the deterioration of their physical and mental health, resulting in their sense of home becoming less porous and more isolating over time. The analysis of the four cases demonstrates that while ageing in place may be desired by many older people, this preference can co-exist with the contrasting feeling of unhomeliness and isolation within the home, as making and unmaking of home can 'occur at the same time' (Baxter and Brickell, 2014: 135).
Our second argument addresses the assumption that a sense of home strengthens over time. Instead our innovative theoretical and methodological approach has helped us to illustrate the important role played by potent connections with both people and places comprising a sense of place and home (Mason, 2018: 123). Mason's affinities approach has allowed us to unpick the relationships between home and the wider environment in experiences of ageing in place, looking anew at how the two intertwine. A sense of belonging to home must be understood in the context of a sense of belonging to place, defined in this article as an enmeshment of relationships, routines, and familiarity with place. Violet and Jean felt a sense of enduring 'insideness' (Rowles, 1983) in their homes and the wider neighbourhood, due not just to the length of time they had lived in the area, but also the lasting nature of the connective charges which their local environments, including their home and neighbourhood, gave rise to over time (Mason, 2018). By contrast, the analysis of Sandra's and John's cases reveals that potent connections between their homes and the wider neighbourhood became unmade over time (Baxter and Brickell, 2014). Although Sandra and John expressed a desire to age in place, the affinities which had supported a strong and porous sense of home in the past became 'divested, damaged or even destroyed' over time (Baxter and Brickell, 2014: 134), resulting in a dwindling sense of 'insideness', which, in turn, fed into an increasing sense of isolation at home (Rowles, 1983).
This article offers a challenge to the notion that ageing in place necessarily enables continued feelings of homeliness. We argue that by paying attention to affinities as potent connections that are not contained in any one space, it is possible to understand the processes whereby people's sense of home becomes connected to or disentangled from the wider environment, as well as whether these affinities are experienced as toxic and enjoyable. Our analysis has also shown that experiences of home are complex and can be contradictory. The nature of our sample means that our data speak to the experiences of those who have lived in the same area for most of their lives, who wish to remain living in their homes and whose mobility has reduced over time. We call for more longitudinal research to be conducted to understand how those who have moved house frequently, those who do not wish to remain living at home and those who remain physically mobile experience home in the context of ageing in place. Furthermore, given that our sample comprises white respondents, further research is needed to understand changing experiences of home in later life among racialised minorities, in addition to deeper exploration of the role of socio-economic inequalities on experiences of home in later life. Better understandings of how a sense of home is made and unmade over time could be used to develop more adequate forms of support for people in later life as they make decisions about ageing in place.
Vanessa May is Professor of Sociology and Co-Director of the Morgan Centre for Research into Everyday Lives at the University of Manchester, UK. She is Co-Editor-in-Chief of the journal Sociology. Her research interests include the self, belonging, temporality, ageing, family relationships, and qualitative methods. Vanessa has published in a number of journals including Sociology, Sociological Review, Time & Society, and British Journal of Sociology. She is the author of Connecting Self to Society: Belonging in a Changing World (Palgrave Macmillan) and co-editor of Sociology of Personal Life (2nd edition, Macmillan).
Camilla Lewis is Senior Lecturer in Ageing and Urban Studies at Newcastle. Her research centres on the themes of ageing, urban change, inequalities, housing, belonging, and community with a strong methodological focus, spanning a variety of ethnographic, sensory, and longitudinal approaches. Her publications have provided theoretical analysis on the everyday experiences of inequalities in urban environments.

Date submitted 26 November 2020
Date accepted 1 March 2022