Ambivalence and Living Apart Together in Later Life: a Critical Research Proposal

http://www.diva-portal.org

Postprint

This is the accepted version of a paper published in Journal of Marriage and Family. This paper has been peer-reviewed but does not include the final publisher proof-corrections or journal pagination.

Citation for the original published paper (version of record):

Connidis, I A., Borell, K., Ghazanfareeon Karlsson, S. (2017) Ambivalence and Living Apart Together in Later Life: A Critical Research Proposal. Journal of Marriage and Family, 79(5): 1404-1418 https://doi.org/10.1111/jomf.12417

Access to the published version may require subscription.

N.B. When citing this work, cite the original published paper.

Permanent link to this version: http://urn.kb.se/resolve?urn=urn:nbn:se:hj:diva-36741 Ingrid Arnet Connidis Western University Klas Borell Jönköping University∗ Sofie Ghazanfareeon Karlsson Mid Sweden University∗∗

Ambivalence and Living Apart Together in Later Life: A Critical Research Proposal

Most unattached older persons who would like The intimate relationships of older persons in an intimate partnership do not want to remarry the Western world are characterized by continu- or be in a marriage-like relationship. A growing ity and change. Marriage still dominates, and, as trend is to live apart together (LAT) in an ongo- more individuals live longer and healthier lives, ing intimate relationship that does not include many marriages survive well into old age. Nev- a common home. We address the debate about ertheless, notable changes have occurred. In the whether LAT constitutes a new form of intimate past, older people were generally affected indi- relationship in a critical assessment of research rectly by contemporary shifts in intimate ties, on LAT relationships that applies ambivalence for example, by their adult children’s divorces and concepts from the life course perspective. (Cherlin & Furstenberg, 1986; Creasey, 1993). We conclude that among older but not younger Now, an increasing number of older persons, adults, LAT relationships are generally a stable especially the “young old” (65 to 74 years of alternative to living with a partner, negotiated age) are taking an active role in these processes in the context of current social institutions and (Connidis, 2010). The young old of today were arrangements. We propose research questions teenagers during the sexual revolution of the that address later life living apart together as 1960s and were deeply involved in the dramatic an innovative alternative intimate relationship. changes that characterize family life in the West We encourage comparative work on the unique (DeLamater, 2012; Lin & Brown, 2012). They challenges of later life living apart together,their now represent what we would describe as the implications for other family ties, and their con- graying of the family revolution. A growing nection to social and cultural arrangements. number of people either arrive at old age already divorced or divorce in later life. About a quarter of the divorces in the United States in 2010 were among individuals older than the age of 50, and, Department of Sociology, Western University, 1151 among the old, divorcees now outnumber wid- Richmond St., London, Ontario, Canada, N6A 5C2 ows and widowers (Brown & Lin, 2012). Many ([email protected]). divorcees wish to have intimate relationships, ∗Department of Social Work, Jönköping University, as do many widows and widowers (Calasanti Gjuterigatan 5 Jönköping, Sweden. & Kiecolt, 2007; Carr, 2004; Moorman, Booth, ∗∗Department of Social Work, Mid Sweden University, & Fingerman, 2006), but few are interested in Kunskapens väg 8 Österund, Sweden. remarrying. In the United States, only 0.6% of Key Words: aging, ambivalence, dating, family dynamics, those aged older than 65 are remarried (Cruz, gender, intimate relationships. 2012; see also Mahay & Lewin, 2007). Journal of Marriage and Family (2017) 1 DOI:10.1111/jomf.12417 2 Journal of Marriage and Family

These trends are reflected to some degree is dominated by smaller exploratory studies in current gerontological and family research; and lacks common terminology and elabo- the intimate relationships of older unattached rated theoretical threads. A central dynamic or single persons can no longer be described as of all relationships is how they are negoti- a neglected topic of research, as was the case ated. Ambivalence is a particularly useful a quarter of a century ago (McElhany, 1992; concept for exploring how the contradictions see also Bulcroft & Bulcroft, 1991; Cooney & between expected and new ways of engaging in Dunne, 2001). Although still relatively limited, committed intimate relationships are negotiated. research on intimacy in later life has expanded to Ambivalence occurs at multiple levels, including include a broader range of topics, for example, the micro level of individuals and relationships, late-life romantic relationships (e.g., Malta & the meso level of social institutions including Farquharson, 2014), including the experiences those concerning intimate relationships, and of older women (e.g., Dickson, Hughes, & the macro level of structured social relations, Walker, 2005) and older widows and widowers sociolegal, economic, political, and cultural (e.g., Carr, 2004), the impact of aging on gay arrangements (Connidis, 2012, 2015; Connidis and lesbian romantic relationships (e.g., Averett, & McMullin, 2002b, 2002a). The concept of Yoon, & Jenkins, 2011), intimacy in long-term ambivalence emphasizes “the coexistence of care (e.g., Frankowski & Clark, 2009), later life contradictory sentiments, expectations, and online dating (e.g., Coupland, 2000), second forces as characteristic of family and of social couplehood (e.g., Koren & Eisikovits, 2011; life” (Connidis, 2015, p. 77) and explores how Koren & Simhi, 2016), and living apart together, these contradictions are and can be negotiated that is, intimate relationships that do not involve by individuals and through social change. Thus, sharing a common home. ambivalence goes beyond mixed feelings about Our aim in this critical review is to examine a partner or about LAT to considering how the living apart together as an alternative intimate contradictions of LAT itself are constructed and relationship among older persons and to offer a negotiated in a multilevel context. conceptual framework and research agenda that Applying ambivalence to LAT relationships will further comparative research. Central to our helps to bridge various efforts to explore LAT discussion is the debate on whether living apart and LAT-like relationships such as dating and together (LAT) is a new method of intimacy or to connect LAT to other family relations and simply a step on the way to cohabiting or mar- to social institutions and arrangements. Cultural rying. Addressing this debate requires specifica- variations in the degree to which marriage is tion of the life stage under study. We argue that entrenched as the acceptable form of intimate the unique situations of LAT later in life are in relationship make ambivalence a useful concept keeping with LAT as a new family form. As a for comparative research on LAT later in life. relatively new type of intimate partnership, LAT Striking a balance between having a committed faces unique challenges; couples must negoti- relationship and maintaining autonomy (Karls- ate their relationships in social contexts that may son & Borell, 2005; Upton-Davis, 2012)—a question the legitimacy of LAT to the point of central contradiction of LAT—is a source of viewing it as a deviant arrangement in some ambivalence negotiated by individuals and cou- countries. We begin by presenting ambivalence ples in the context of social arrangements that and concepts from the life course perspective as promote marriage and, more recently, cohabi- a useful framework for studying LAT. We apply tation. Contradictions at the level of structured this model to a critical assessment of the LAT lit- social relations also underscore the differential erature, starting with definitional variations that appeal and accessibility of LAT as an option, hinder progress on this topic. We conclude by for example, based on gender and class. Recent proposing research questions that we believe treatments of agency that emphasize its connec- will advance the comparative study of living tion to social constraints are central to the con- apart together in later life, what we term LLAT. cept of ambivalence (Connidis, 2012, 2015) and to the life course perspective (Settersten, 2003). Placing LAT relationships in a life course per- Conceptual Framework spective helps to clarify the unique features of As in many new, rapidly expanding areas LLAT. The concepts of life stage and linked of social science research, the study of LAT lives (Heinz, 2001) are especially relevant to Ambivalence and Later Life LAT Relationships 3 studying LAT across the life course. Circum- which LAT is considered an alternative intimate stances and reasons for forming unions at dif- tie or a temporary situation. ferent life stages vary markedly (Upton-Davis, 2012). Among younger adults, many still aim for forging a secure economic future and hav- What Is LAT? ing children, goals that make cohabitation and Notable cultural differences in the extent to marriage more probable objectives. At this life which LAT is an identified and accepted inti- stage, the primary family links of single per- mate relationship are reflected in linguistic sons are to parents and siblings, ties that in variations that inhibit a common focus and most cases are either sources of support (parents) research agenda. In the United States, expres- or resource neutral (siblings). Consequently, for sions used to describe long-term intimate many at this life stage today, LAT is regarded as relationships between older singles are often a temporary arrangement either because couples borrowed from teenage culture. When older are still determining whether they are commit- U.S. respondents in Talbott’s (1998) study of ted enough to move in together or because they romantic relationships used expressions such as are living with parents while working toward “dating” or “going steady,” they often qualified financial independence and being able to live them with reservations about their inappropri- together. ateness. “It sounds nuts,” said one of Talbott’s In contrast, among older single persons, most respondents (1998, p. 446) “for an old lady have already had a long-term union, usually to have a boyfriend” (see also Benson, 2013; marriage, that ended either in the death of a Benson & Coleman, 2016a; Brothers, 2015; partner or dissolution of the relationship. The Carr, 2004; for Great Britain, see Coulter & Hu, majority of these unions resulted in children 2015; Haskey, 2005). Similarly, LAT partners in and grandchildren who remain the primary a qualitative Canadian study (Kobayashi, Funk, family links for older parents. Building financial & Khan, 2016) struggled to find a good term for security and having more children are rarely their partner, disliking the terms boyfriend or the goals of having an intimate relationship. girlfriend because of their failure to reflect the Indeed, some state-sponsored income programs strength of their relationship. for seniors penalize recipients who marry. In contrast, LAT relationships have an onto- Instead, the desire for companionship and inti- logical status in Sweden, Norway, the Nether- macy are primary motivators. Unlike younger lands, and Belgium. In Norway and Sweden, adults, the primary family link of single older the expression sarbo or särbo, meaning “to live adults—children—involves ongoing respon- apart” have been in common use for some time sibility. Older parents typically continue to and describe long-term intimate relationships offer support and protect their resources for the that do not include a shared home. The term current and eventual use of their children. Some LAT was first used in the late 1970s by aDutch of the young old also have parents, but by this journalist referring to a movie in which two stage of life, their very old parents are more people shared a common relationship but not likely to be receiving than giving support. a common home. Besides being an acronym, In sum, ambivalence and concepts from the the word lat in Dutch also means stick (Levin, life course perspective encourage connections 2004). In Belgium and the Netherlands, LAT between multiple levels of analysis and attention has become generally accepted in everyday use. to the dynamics of relationships beyond indi- When the language provides us with a typi- viduals and dyads. The concept of ambivalence fied meaning of this relationship, LAT becomes focuses our attention on the dynamics of LLAT part of accepted, commonsense knowledge (see relationships, their impact on extended family Schutz, 1972) and can be discussed, considered, dynamics, and their connection to social and cul- and assessed as a realistic alternative to other tural arrangements. Life stage and linked lives forms of intimate ties (Borell & Karlsson, 2003). create quite different circumstances across age Against this backdrop, it is not surprising cohorts that are pivotal to understanding vary- that such a large proportion of the research ing approaches to intimate relationships, includ- on older LAT couples is from Belgium (e.g., ing LAT, across the life course. Yet, as we shall Lyssens-Danneboom & Mortelmans, 2015), see, studies of LAT relationships often fail to dif- the Netherlands (e.g., de Jong Gierveld, 2004), ferentiate among age cohorts and the degree to Norway (e.g., Levin, 2004), and Sweden (e.g., 4 Journal of Marriage and Family

Karlsson & Borell, 2002). Only recently have & Mays, 2009). Even in the few cases in researchers in Australia (Upton-Davis, 2012, which LAT relationships have been included in 2015), Canada (Funk & Kobayashi, 2016; such surveys or have been studied in other Kobayashi et al., 2016; Turcotte, 2013), and the statistically representative samples, different United States (Benson, 2013; Benson & Cole- definitions and measures hinder international man, 2016a, 2016b; Brothers, 2015) considered comparisons. LAT a category of intimate relationship in later Estimates of the incidence of LAT rela- life worthy of study. tionships in Australia, North America, and Differences in defining LAT reflect diverse West European countries lie in the range cultural, period, and age-based assumptions of 7% to 10% of the population (e.g., Rei- about what constitutes both intimacy and com- mondos, Evans, & Gray, 2011 [Australia]; mitment. An important challenge to researchers Duncan & Phillips, 2011 [Britain]; Milan & is to develop a definition of LAT that facilitates Peters, 2003 [Canada]; Levin, 2004 [Norway]; the study of such relationships regardless of dif- Strohm et al., 2009 [United States]). Young ferences in everyday understanding. We favor a adults, typically defined as between 20 and definition of LAT that separates LAT as a unique 24 years of age, often constitute the largest intimate tie considered an end in itself from LAT group categorized as LAT partners in these as a temporary arrangement, ideally on the way surveys. In Australia (Reimondos et al., 2011), to marriage or cohabitation. This follows earlier France (Régnier-Loilier, 2015; Régnier-Loilier, calls for distinguishing between LAT as a sub- Beaujouan, & Villeneuve-Gohalp, 2009), and stitute partnership rather than a stepping stone Southern Europe (Billari, Rosina, Ranaldi to an established form of partnership (Ermisch & Romano, 2008), half or more of the LAT & Siedler, 2008; Upton-Davis, 2012). We define population is estimated to be younger adults. LAT as a chosen, intimate relationship between Although more common now (see Asendorpf, partners who are committed to LAT and to each 2008), LAT relationships among older persons other for the long term and who live in separate still involve relatively few individuals in mid- homes (see Benson & Coleman, 2016b; de Jong and later life. In Canada, 2.3% of those older Gierveld, 2002; de Jong Gierveld & Peeters, than 60 years of age were LAT partners (Tur- 2003; Levin, 2004; Upton-Davis, 2012). Some cotte, 2013), and 4.3% of those aged 50 to include the additional provisos that such LAT 64 years in the Netherlands (de Jong Gierveld, relationships are sexually active and publicly 2015) and 5% of those older than 60 years of known (Levin, 2004; Stevens, 2004) and monog- age in Sweden were LAT partners (Bildtgård amous (de Jong Gierveld, 2002). In our view, & Öberg, 2015). In a statistically representative the realities versus ideals of sexual activity and sample of unmarried, noncohabiting U.S. adults monogamy in intimate relationships are subjects aged 57 to 85, 18% of those aged 57 to 64, 14% to be studied rather than assumed as defining of those aged 57 to 85, and 11% of those aged features of them. For example, marriages that 75 to 85 years agreed that they “currently have are not sexually active or in which partners are a romantic, intimate, or sexual partner” (Brown sexually active with others are still considered & Shinohara, 2013, p. 4). Similar to most sur- marriages, even if they violate assumptions veys, this study does not determine how many about ideal marriages. of those who have an intimate relationship see it as a long-term arrangement or their subjective definition and view of the relationship. LAT and Life Stage LAT relationships among older people are far more significant than nominal data may imply. Incidence of LAT Studies from Germany (Asendorpf, 2008), the Variations in which criteria are deemed neces- Netherlands (de Jong Gierveld, 2015), and Swe- sary to qualify as LAT result in different sample den (Bildtgård & Öberg, 2015) indicated that populations (Duncan & Phillips, 2011). Statis- LAT was especially relevant to those who started tics about families and intimate relationships are a new relationship in their old age. In Sweden often based on households, so the distribution (Bildtgård & Öberg, 2015), of those who had of LAT relationships has not generally been started a new intimate relationship after their included in national population surveys (Borell 60th birthday, 7 of 10 were LAT partners and & Karlsson, 2003; Strohm, Seltzer, Cochran, fewer than 3 of 10 were cohabitants, leaving a Ambivalence and Later Life LAT Relationships 5 very small number who had chosen marriage (for long-term committed relationship. As Duncan similar results in the Netherlands, see de Jong and Phillips (2011) concluded, steady girlfriend Gierveld, 2004). and boyfriend relationships among younger adults who did not live together and commuter relationships when jobs in different locations LAT: New Family Form or Stepping Stone? forced couples apart, are not new. In some of the literature, LAT relationships are Many LAT relationships among those with seen as one of the defining aspects of contem- coresident, dependent children are also tempo- porary differentiation of intimacy and family rary and involuntary. In this case, consideration life. This is particularly evident in the work of of a third party is the impetus for LAT. Single the pioneer researchers of LAT. As a research parents, in practice often mothers, choose not to concept, LAT was originally launched with a share their home with their partner to preserve claim to describe “a historically new family the boundary of the family home and to provide form” (Levin, 2004, p. 223; see also Levin & continuity for and avoid conflict with their cores- Trost, 1999). Does a critical review of subse- ident children (de Jong Gierveld & Merz, 2013). quent empirical research support such a univer- Single parents may also be concerned about the sal claim? Following a life course perspective, reaction of others such as parents or former part- we expect LAT relationships to be characterized ners who may consider it inappropriate to have a by a high degree of heterogeneity in terms of live-in partner or view it as grounds to withdraw subjective meaning, function, and durability that financial support. Another third party—a parent relate in part to life stage (see Benson & Cole- who requires care—leads some adult children, man, 2016b; Coulter & Hu, 2015). in practice often adult daughters, to delay liv- Qualitative and quantitative studies coming with a new partner while looking after their bine various versions of LAT and multiple age elderly parent (see Régnier-Loilier et al., 2009; cohorts, making it difficult to focus on LAT Reimondos, 2011). With longer lives, these chil- as a committed alternative relationship related dren may be old themselves. to life stage. Studies that are ostensibly about Waiting to cohabit while in a LAT relation- LAT are often about LAT as a stepping stone ship creates its own contradictions, and couples to marriage or cohabiting. For example, in a must negotiate the ambivalence of their goal to British study of LAT relationships, defined live together with their current realities. Older as “currently in a relationship with someone persons too may be reluctant to engage openly you are not living with here” (Carter, Duncan, in nonnormative romantic relationships such as Stoilova, & Phillips, 2015, p. 579), only 6 of the LAT, but they are more likely to keep them 50 respondents would fit our definition of LAT secret or provide justifications for them (Koren because they were the only ones who considered & Eisikovits, 2011; Kobayashi et al., 2016) than LAT a preferred and committed relationship. to be deterred from LAT altogether. Here, the Young adults who LAT usually intend to ambivalence created by engaging in a relation- establish a common home, but cannot do so ship that is not socially accepted is managed at present because of practical barriers such as through accounts and secrecy. a housing shortage, unemployment, or differ- Regarding durability, although LAT relation- ent locations for work or school (Reimondos ships between young to middle-aged partners et al., 2011 [Australia]; Liefbroer, Poortman, are often transitional, the LAT relationships of & Seltzer, 2015 [France and Germany]; Dom- older couples are typically long term. In Canada mermuth, Noack, & Wiik, 2009 [Norway]). (Turcotte, 2013), young adults (20–29 years) Recurring recessions have meant that young averaged 2.3 years together, whereas older adults who cannot establish themselves in the adults (older than 60 years) averaged 7.5 years. labor or housing markets are often obliged to In a Swedish study, all 116 LAT partners aged return to their parents’ home instead of setting 60 to 90 years described their relationship as up a household with their partner (Connidis, long term, with a median length of 7 years and 2014). Those who intend to live with their a range from 1 to 28 years (Karlsson & Borell, partner once current impediments are resolved 2002). Differences in relationship duration or because they are uncertain about their rela- are even more marked in longitudinal studies. tionship are more accurately viewed as being In a study in France (Régnier-Loilier, 2015), in a testing period or passing phase than in a only one in 10 of the young couples was still 6 Journal of Marriage and Family

LAT 3 years after the first interview, and virtu- once a week, with two thirds seeing their partner ally none 3 years later. The majority had either daily or several times a week. moved to a common household (36% with the Variations in the terminology that British same person) or separated. In contrast, only 7% LAT partners who view themselves as cou- of seniors had moved in with their partner, more ples use to describe their relationship suggest than half of them were still in their original LAT greater commitment among older couples; relationship after 3 years, and about a third after they more often used the terms partner or 6 years; the balance experienced the death of or husband or wife, and younger persons more separation from their LAT partner. often used girlfriend or boyfriend (Duncan, Carter, Phillips, Roseneil, & Stoilova, 2014). In Swedish (Karlsson & Borell, 2002) and Belgian Choosing to LAT: Later Life LAT Relationships (Lyssens-Danneboom & Mortelmans, 2015) as a New Family Form studies of LAT, fidelity was generally seen as a prerequisite for the relationship, a view The greater durability of older couples’ LAT also supported by a survey of LAT partners relationships reflects their preference for this in Britain in which 87% thought it would be type of intimate tie. At this life stage, LAT is wrong if a LAT partner had sex with someone neither primarily a practical necessity nor a step else (Duncan et al., 2014; see also Carter et al., toward establishing a common home. A British 2015). In a U.S. study (Brothers, 2015) even study of LAT that identified one category of sexually active older LAT couples viewed sex participants who both preferred and were com- as a less significant component of their relation- mitted to LAT (Carter et al., 2015) suggests that ship than they did when in marriages earlier in commitment has two elements: commitment to their lives. The assumption of greater freedom LAT as a form of intimate relationship and com- coupled with monogamy between couples who mitment to one’s LAT partner. In an Australian do not live together requires an added degree study (Reimondos et al., 2011), more than 70% of trust—another paradoxical element of LAT of older LAT partners had chosen to LAT over (Kobayashi et al., 2016). any other type of relationship. A British study Cohort differences in LAT echo earlier U.S. that distinguished between partner LAT couples research on dating that found almost half of and dating LAT couples found that those aged respondents aged 55 to 75 years and older 65 to 97 years fell entirely in the partner LAT described their dating relationship as “steady” category, and those aged 55 to 64 years were and only a small number considered it a step twice as likely to be partners rather than daters toward marriage, leading the authors to conclude (Duncan & Phillips, 2011). that “dating is taken seriously by older adults Committed or intrinsic LAT partners in Bel- but not as a means to marriage,” especially by gium have features that parallel normative views women (Bulcroft & Bulcroft, 1991, p. 258; see of marriage as involving a strong sense of being also Bulcroft & O’Connor, 1986). Watson and in a couple with a deep emotional investment in Stelle (2011) found that “young old” women did commitment, loyalty, being together, and mutual not see dating as a step toward remarrying, even respect (Lyssens-Danneboom & Mortelmans, in cases when dating developed into a steady, 2015). For most of the 116 older LAT partners long-term relationship; a steady, long-term in a Swedish study, the overwhelming motive for relationship, not marriage, was their goal. Older LAT was the intimacy of the relationship; LAT adults in the United States seriously pursue partners regarded their relationships as deeply romantic partners (Stephure, Boon, MacKin- intimate, marked by mutual trust, understand- non, & Deveau, 2009), and, when compared ing, and the sharing of confidences (Borell & with the personal ads and dating profiles of Karlsson, 2003). Almost all of the LAT partners younger adults, they are less future oriented and had daily contact with one another via telephone, more focused on present concerns (Alterovitz & more than one third (36%) met almost daily, and Mendelsohn, 2013). another half met a couple of times a week (Karls- Counter to normative assumptions about mar- son & Borell, 2002). Most of the older LAT cou- riage, older Belgian LAT couples also valued ples in a British study (Coulter & Hu, 2015) lived autonomy, time on their own, economic inde- within an hour of each other, with two thirds less pendence, and maintaining territorial boundaries than 30 minutes apart, and saw each other at least (Lyssens-Danneboom & Mortelmans, 2015). Ambivalence and Later Life LAT Relationships 7

Older Swedish couples also emphasized auton- chosen for its good fit with their current life stage omy as a prime motivator for LAT (Borell and their links to adult children. Participants & Karlsson, 2003). Similar to LAT couples, in a qualitative U.S. study of 59- to 89-year- later life dating women “prided themselves on olds in LLAT relationships (Brothers, 2015) their independence and did not want to give it viewed marriage as a relationship associ- up” (Dickson et al., 2005, p. 73). Analyses of ated with the life stage of having children. A personal online ads generally confirm that older 65-year-old widowed mother of two observed women are seldom prepared to compromise the following: their independent lifestyle with a new marriage (e.g., Levesque & Caron, 2004; McWilliams & Ithinkthatifyouareraisingchildren...itmakes Barrett, 2014). good sense to have a committed relationship that The low level of organization of LAT rela- works together to raise these kids. But outside of tionships and the absence of many of the com- that, I don’t really see where marriage has a reason. (Brothers, 2015, p. 64) mitments and ties that characterize marriage (Yodanis & Lauer, 2014) put the focus on the Brothers (2015) concluded that older persons emotional side of the relationship (Borell & in LLAT relationships were engaged in a new Karlsson, 2003). In theory at least, LAT simulta- family form that contrasted with the focus on neously gives both partners autonomous access gendered family formation and child rearing in to their resources and home along with access to their earlier marriages. strong mutual commitment and intimacy. Part- A multilevel conception of ambivalence ners defined as being in intrinsic LAT relation- encourages us to link agency to structure and ships in a Belgian study (Lyssens-Danneboom emphasizes variations in opportunities for exer- & Mortelmans, 2015, p. 19) “operate as two cising particular choices. LAT relationships autonomous individuals with their own objec- among older couples reflect class-based differ- tives (maintaining autonomy and freedom) and ences that make the capacity to run two homes cooperated . . . to achieve a common goal (the more likely among those with more resources. mutual provision of intimacy, companionship Among younger couples, LAT often occurs and support).” LLAT relationships exemplify because a couple cannot afford their own home one of the features of Giddens’ (1991) ideal type and, therefore, lives with others (often parents) of “pure” relationship; to a relatively high degree until they can (Coulter & Hu, 2015). In cases they depend on “satisfactions or rewards generic where coresidence in old age means forfeiting to that relationship itself” (p. 224). public funding, then those with fewer resources Among older persons, priorities shift as the may be more likely to choose to LLAT. responsibilities of raising children and paid work Gender also shapes the appeal of LAT later diminish. A limited future increases a preference in life. The pull of linked lives and responsibil- for affectively rich relationships (Carstensen, ities to others is gendered, with women more 1995). LLAT is well suited to this psycholog- likely to experience consequent limits on agency ical process. Emotional content is brought to in the types of intimate relationships that are the fore, and, in contrast to marriage, the auto- compatible with these commitments (de Jong matic development of increasing commitments Gierveld & Merz, 2013; Upton-Davis, 2015). and responsibilities is not assumed (de Jong Women who can afford to maintain a separate Gierveld, 2004; Karlsson & Borell, 2002). At home appear to be more motivated than men the same time, among Swedish LLAT partners, to be in LLAT relationships as a way of avoid- reciprocal emotional support was accompanied ing the gendered arrangements that they expe- with almost daily practical support (Karlsson & rienced in earlier marriages (Brothers, 2015; Borell, 2002), and they received more support Upton-Davis, 2012). Swedish research (Karls- from their LLAT partner than from any other son & Borell, 2002) showed that women were person in their network, including children and significantly more motivated to choose LAT grandchildren (Karlsson, Johansson, Gerder, & to secure independence and avoid a traditional Borell, 2007). gendered division of labor. Whether women Most older persons who LAT are previously regarded their previous marriage as “happy” or married parents (Karlsson & Borell, 2002; “unhappy,” they often saw their shared home as Karlsson et al., 2007; see also de Jong Gierveld representing responsibility rather than freedom & Merz, 2013). Unlike younger adults, LAT is and work rather than relaxation. 8 Journal of Marriage and Family

Among repartnered adults aged 50 and older Proposed Research Agenda in the Netherlands (de Jong Gierveld & Merz, Our critical review establishes that LAT in older 2013), women, older participants, and parents age is characteristically different from LAT at were more likely to LAT than to remarry or earlier life stages, making LAT more in keep- cohabit. Women were also motivated to LAT ing with a new family form than a transitory as a way of protecting their financial assets coping mechanism in later life. Research on dat- (de Jong Gierveld, 2002). A Canadian quali- ing in the United States shows parallel life stage tative study of LAT couples involving mostly differences. Although younger adults generally divorced partners (Funk & Kobayahsi, 2016) view and experience LAT as a temporary situa- also found that women saw LAT as a way to tion, older ones typically seek it out as an end in avoid gendered household tasks and partner itself and aim to negotiate long-term, stable LAT nurturing and to protect autonomous decision relationships. For this reason, we propose the making about finances and the home. Men did acronymn LLAT to emphasize the unique fea- not make parallel observations. Australian work tures of LAT in later life. by Upton-Davis (2015) involving women 45 Does the fact that LAT is usually a durable years of age and older showed the appeal of relationship for older couples but a flexible LAT as a transformative option that allowed arrangement for younger ones (Coulter & Hu, for a more egalitarian intimate relationship 2015) mean that older couples in LLAT relation- marked by independence and a positive view of ships never marry? Not necessarily, but just as oneself. marriage is considered marriage even though it The role of women in establishing LAT rela- might end in divorce or widowhood later on, so tionships can be seen as a result of the influ- a committed LLAT relationship is a LLAT rela- ence that earlier experiences have on choices tionship even if it might result in marriage later made in later life (Elder, 1994). If their previ- on. In the meantime, for both LLAT and mar- ous marital households were a constitutive force riage, the accepted definition of the situation at in the reproduction of traditional gendered rela- the time has consequences for the partners in tions, their own household today is a resource the relationship and for those in their social net- base from which they may avoid an asymmetri- works. The possibility of change does not deny cal distribution of household labor and unequal the power of currently living apart together or demands of caring for a partner (Borell & Karls- being married if the couples involved consider son, 2003; Brothers, 2015). The ambivalence their relationship a long-term arrangement. of enjoying a committed intimate relationship Our focus has been on LLAT relation- that threatens egalitarianism in household labor ships in which both partners have chosen to is resolved by LAT. LAT relationships can also be in a committed relationship that protects be a method of resolving ambivalent views autonomy and limits obligations. Choice or about cohabiting (Roseneil, 2006) and compet- agency refers to acting on one’s own behalf ing desires to both stay in a preferred neigh- in the context of available alternatives, that is, borhood and engage in an intimate relationship agency-within-structure (Connidis, 2012, 2015; (Coulter & Hu, 2015). Settersten, 2003). LAT partners who defend Social policies that rest on the assumption their choice to LAT rather than marry because that couples have their own safety net—each of their emotional commitment or financial, other—are motivators to LAT, especially among parental, or legal situations were described in women. In this case, institutionalized policy a Canadian study as displaying interpretive tends to reinforce gender and class relations. For agency (Kobayashi et al., 2016). Unveiling the example, in Sweden, decreased access to public fact that some women enter into LAT relation- social services has made it increasingly difficult ships because they feel constrained or vulnerable for couples who live together to get assistance. rather than doing so as a consequence of reflex- Accordingly, Swedish women who LAT tend to ive and strategic choices (Duncan, 2015) does see a home of their own as a protection against not negate the fact that they have chosen to potential societal demands that partners provide LAT as their best available alternative. Seeing the care that would otherwise be the responsi- such vulnerability or constraint as motivators bility of local welfare services. For those who unique to LAT also ignores the reality that many cannot afford to forfeit public funding, there is women (and men) enter marriage for similar also a push toward LAT. reasons. The unique contradictions of choosing Ambivalence and Later Life LAT Relationships 9 to LAT when another form of relationship is merging LAT with normative views of what con- preferred and their negotiation are topics for stitutes a romantic relationship are character- further research. ized as ambivalent. The absence of an accepted LLAT can be viewed as a way of resolving the term for LAT in the United States noted by ambivalence that older persons may have about partners trying to find appropriate titles for one both cohabiting and marrying, but LLAT has its another (Benson & Coleman, 2016a; Brothers, own contradictions that require negotiation. The 2015) reflects cultural variations in the extent to term itself suggests the contradiction inherent in which LAT is a recognized and accepted inti- negotiating a committed, intimate relationship mate tie and in the likelihood of institutionaliz- (togetherness) without coresiding (apartness) in ing LAT, comparative topics that warrant further cultures where coresidence has been a defin- study. These variations in turn help to account ing feature of intimate ties (Stoilova, Roseneil, for different levels of ambivalence associated Crowhurst, & Santos, 2014). As is true of all with LAT. families of choice, LAT relationships involve Explorations of alternative intimate relation- ambivalence in the tensions between critique ships that lapse into traditional ideas of “good and transformation and between belonging and partnerships” as those in which partners feel exclusion (Pidduck, 2009). obliged to care for one another (see, e.g., de Among couples that include one partner who Jong Gierveld, 2015) highlight the socially con- prefers to LAT and another who would rather structed ambivalence that couples who LLAT live together (see, e.g., Duncan et al., 2014), the must negotiate. The assumption that being a ambivalence of negotiating a committed rela- good partner includes the obligation to provide tionship in separate households is compounded extensive care challenges the legitimacy of inti- by contradictory goals between partners. For mate ties where this is not a premise. Yet if part- LAT couples in which both partners want to LAT ners choose LAT in part because they prefer not because they value a committed intimate rela- to assume care obligations, then they are being tionship combined with autonomy and limited the kind of good partners they had agreed to obligation, the ongoing challenge is to negotiate become by not expecting to either give or accept this paradox. care. What is the experience of LLAT partners As is often true of applications of ambiva- when the need for care arises? lence to family ties, reference to ambivalence At the heart of LAT is the contradictory aim in LLAT relationships is usually at the micro of having a long-term, committed, intimate level of mixed feelings. Some LLAT partners relationship while preserving autonomy and and couples experience ambivalence about the minimizing obligations to give and receive appropriateness of LAT (Benson & Coleman, care. Circumstances that threaten the balance of 2016b) and about care exchanges (de Jong intimacy and independence, such as the declin- Gierveld, 2015). Such ambivalence is focused ing health of a partner, are likely to heighten on the micro level of psychological feelings but ambivalence in LLAT relationships (see Con- can be related to institutionalized expectations nidis, 2015). Indeed, LAT partners themselves regarding what constitutes a committed rela- express ambivalence about the impact that tionship and to broader structural issues such as future care needs may have on their relationship gender and class. (Kobayashi et al., 2016). What happens if a Ambivalence about LLAT is likely to vary partner’s situation changes and more extensive based on the degree to which a culture and soci- support is needed? How do couples negotiate ety view marriage as the proper or ideal intimate this contradiction and how are other family relationship. In Great Britain, marriage remains members implicated in this negotiation? the overwhelming favorite (Duncan et al., 2014). There are no longitudinal studies that can Negotiating LLAT in the context of established show the actual extent of care commitments understandings of what comprises a real and between LLAT partners over time. Studies in socially accepted committed relationship cre- Sweden (Karlsson et al., 2007) and the Nether- ates ambivalence for some individuals who LAT, lands (de Jong Gierveld, 2015) were more or particularly in countries where LAT is a rel- less hypothetical, as the respondents had gener- atively new alternative. In a U.S. qualitative ally not been faced with the challenge of a seri- study of 25 LAT partners (Benson & Coleman, ously ill partner. When asked to envisage a future 2016b), those who are still coming to terms with scenario in which their partner is seriously ill, 10 Journal of Marriage and Family respondents had higher expectations of receiv- substantial numbers of older persons, particu- ing care from their LLAT partner than from their larly women, structural barriers to this option relatives, and none of them would consider end- must be explored. ing their LLATrelationship because their partner There is a great need for a broader network became ill (Karlsson & Borell, 2002; Karlsson perspective on the implications of LLAT rela- et al., 2007). Most, however, envisaged caring tionships for the larger constellation of fam- for an ill LAT partner a few days a week or a ily ties—the issue of linked lives. The desire few hours a day, not full-time. to protect the inheritance of one’s children by Belgian data (Lyssens-Danneboom & Mortel- not living with an intimate partner often moti- mans, 2015) showed that LAT partners consid- vates LLAT. Adult children can be influential in ered their partners as providers of emotional decisions to LAT rather than live with a part- support and care, but there were signs of ambiva- ner (de Jong Gierveld & Merz, 2013). One’s lence about the ability to meet future care needs. own or a partner’s children can sabotage plans A British study (Duncan et al., 2014) that relied to live together and may decrease contact with on hypothetical views of future care provision parents when children hold negative views of found that slightly more than one quarter of their parents engaging in LLAT relationships. those in chosen LAT relationships believed that Comparisons with continuously married part- their partner would care for them were they “ill ners found that repartnered couples, especially in bed.” Men were more likely than women those who cohabit or LLAT have weaker ties to believe this and, unlike the Swedish study, with their children (de Jong Gierveld & Peeters, both men and women were more likely to say 2003). The risk of greater conflict that repartner- that a family member would provide such care. ing in a cohabiting relationship poses (Schenk & In practice, some LLAT partners had received Dykstra, 2012) leads some parents to LAT as a extensive support when ill. way to resolve the ambivalent goals of engaging Research in the Netherlands suggests that in an intimate tie and maintaining closeness with the attitudes held by LLAT couples about care children. exchanges are not matched by their behavior The ambivalence of LLAT relationships in the face of actual illness (de Jong Gierveld, reverberates across family ties, raising questions 2015). LAT partners who had not yet faced the about the boundaries and expectations of others situation of a seriously ill partner often expressed regarding the LLAT partner of a family member. reservations about providing unlimited care, but Key assumptions made by LLAT partners in those who actually had a very ill partner usually qualitative studies (e.g., Brothers, 2015; de Jong provided the same degree of care as would mar- Gierveld, 2015) are that they have made the nec- ried spouses. This suggests that for some LLAT essary arrangements for health care, that their couples, the resolution of ambivalence is found children will take care of them if needed, and in favoring the commitment of the relationship that their partner’s children will care for their over the autonomy of not providing care. partner. From the standpoint of adult children, Comparative research should explore vari- the fact that a parent is LLAT cannot be assumed ations in how the ambivalence of autonomy to mean future care for their parent if needed. and intimacy are resolved when the need for What are the implications of this assumption care arises, a transition that is a revealing site for negotiating a relationship with a parent’s for studying ambivalence (Connidis, 2015). The LLAT partner? Is there an effort to develop link between policy and LLAT and its impact an emotional bond in the absence of assuming on negotiating caregiving is an important com- an instrumental one? Does the expansion of ponent of this topic. As Upton-Davis (2012) networks that LAT potentially allows increase observed, LLAT also has policy implications for the number of potential care providers (Cherlin, housing if more old people continue to live alone 2010), or is there ambivalence among adult even when in an intimate relationship and for children about offering care to their parent’s health care arrangements given that one can- LLAT partner? not assume that having a partner means receiv- Is a LLAT partner “part of the family” or “my ing care. To date, LLAT among heterosexuals parent’s partner”? Variations in the response are tends to be more common among those with likely to parallel those among step ties—much more resources (Upton-Davis, 2015). Given that will depend on the duration of the relationship, LLAT appears to be a positive experience for the extent of shared activity, and the degree of Ambivalence and Later Life LAT Relationships 11 emotional attachment that develops over time lesbian couples’ perceptions of the ambivalent (Connidis, 2010). A qualitative British study views and actions of other family members (Stoilova, Roseneil, Carter, Duncan, & Phillips, regarding their sexual identity and same-sex 2017) suggested that strong emotional attach- relationship (Reczek, 2016). Exploring both ment and the view of a LLAT relationship as the actual ambivalence experienced by family ongoing enhanced viewing a LLAT partner as members of LLAT partners and LLAT partners’ part of the family, which often extended to perceptions of ambivalence on behalf of family the LLAT partner’s family members. In cases members would help us to better understand the where LLAT partners do eventually provide care relational character of ambivalence. As well, despite a starting assumption that this would ambivalence would be situated in family net- not occur, are the family members, especially works and broader social arrangements rather children, expected or likely to support their than in the feeling states of individuals. parents’ caregiving efforts? Given that some As is true of today’s LLAT relationships, ex-wives care for their ex-husbands, primarily significant numbers of cohabiting same- and as an expression of their commitment to helping opposite-sex couples once preferred their unoffi- their children (Cooney, Proulx, Snyder-Rivas, & cial relationships precisely because they wanted Benson, 2014), perhaps children of parents who to avoid both the ideological and practical LLAT become implicated in caregiving chains trappings of marriage. The institutionalization that contradict expectations of filial obligation. of cohabitating and same-sex relationships in How is such ambivalence resolved? many countries may foreshadow similar changes The paradoxical concern about public and regarding the legal status of LAT, another poten- legal status in relationships that are sought tial source of ambivalence. An asset of applying after for their autonomy and limited obliga- ambivalence to LLAT relationships is to high- tion is a source of structured ambivalence as light the complexity of choices to LAT and LAT couples negotiate committed relation- to link the choices that individuals make to ships that lack legal standing (Duncan, Carter, the larger social and cultural structures that Phillips, Roseneil, & Stoilova, 2012; Lyssens- influence, constrain, or facilitate their choices Danneboom, Eggermont, & Mortelmans, 2013; (Connidis, 2015). If living with a partner did Lyssens-Danneboom & Mortelmans, 2015). not have legal implications for the subsequent Lyssens-Danneboom and Mortelmans (2015) property and resource rights of that partner, a found legal insecurity among LLAT partners couple might choose to cohabit. Alternatively, who were concerned about not having a public if LLAT were more socially accepted, more identity as a couple and about whether they individuals might choose this option. Ideally, if would have access to their partner should there LLAT becomes more institutionalized, it will be a medical problem or accident. At the same not lose its potential as a force for social change, time, LLAT partners generally accepted that, including more egalitarian gender relations upon their partner’s death, it would be their (Upton-Davis, 2015). partner’s family members and not themselves We are not arguing that ambivalence applies who would be responsible for making subse- only to LLAT but, rather, that LLAT has some quent arrangements. The potential cost of legal unique ambivalent features. Comparing LAT exclusion is exemplified by a man who was with marriage, Lyssens-Danneboom and Mortel- sole care provider to his LLAT partner up to her mans (2015, pp. 19–20) concluded that, unlike death, but was then denied partner status by her marital spouses who are bound to behave “as children (de Jong Gierveld & Merz, 2013). expected” because they have agreed on “a set Legal status issues parallel long-held para- of cultural and behavioural guidelines,” LAT doxes in committed same-sex relationships in partners are at risk that they “will enter the past and present jurisdictions where marriage is relationship with different intentions and expec- not a legal option for sanctioning the relation- tations.” Current divorce rates seem evidence ship. The ambivalence of their social standing as enough that many individuals also begin mar- couples may reverberate across the family net- riage with intentions and expectations that are work when other family members make efforts a poor fit with those of their spouse. Clearly, to claim the legitimacy of a family member’s LLAT partners have a greater burden of proof gay or lesbian relationship (Connidis, 2003). to be recognized as a couple based on the Recent research identified mid-life gay and content of their relationship, but marital partners 12 Journal of Marriage and Family too must negotiate contradictory expectations of than assumptions made about them as ideal their relationship. types. Regarding methods, future explorations of gender and LLAT will benefit from studies that include men and women. Dyadic studies (see References Benson & Coleman, 2016b) that focus on the Alterovitz, S-R., & Mendelsohn, G. A. (2013). Rela- interaction between LLAT partners would help tionship goals of middle-aged, young-old, and to address how gender plays out in LAT rela- old-old internet daters: An analysis of online per- tionships and whether differing perspectives on sonal ads. Journal of Aging Studies, 27, 159–165. LAT are a unique source of ambivalence or a https://doi.org/10.1016/j.jaging2012.12.006 challenge to resolving it. Family constellation Asendorpf, J. B. (2008). Living Apart Together: Age- studies that include a LLAT couple’s extended and cohort-dependence of a heterogeneous life family, particularly children, would broaden our style (Living Apart Together: Alters—und Kohor understanding of family relationship dynamics tenabhängigkeit einer heterogenen Lebensform). and the implications of LLAT beyond the cou- Kölner Zeitschrift fur Soziologie and Sozialpsy- ple. To truly explore a life course perspective chologie, 60, 749–764. https://doi.org/10.007/s11 577-008-0035-4 and the negotiation of ambivalence in LAT rela- Averett, P., Yoon, I., & Jenkins, C. L. (2011). Older tionships over time and in response to life transi- lesbians: Experiences of aging, discrimination tions, longitudinal studies are needed. Studies on and resilience. Journal of Women and Aging, 23, LLAT should relate the experience of partners, 16–32. https://doi.org/10.1080/08952841.2011 couples, and families to larger social and cul- .587742 tural contexts that enhance a comparative under- Benson, J. J. (2013). From living apart, to standing of this intimate tie. Both qualitative living-apart-together: Older adults develop- and quantitative research shed valuable light on ing a preference for LAT (Doctoral dissertation). this topic. Retrieved from https://mospace.umsystem.edu/ xmlui/bitstream/handle/10355/44653/research .pdf?sequence=1 Benson, J. J., & Coleman, M. (2016a). Older adult Conclusion descriptions of living apart together. Family Rela- Isolating LAT relationships in later life for tions, 65, 439–449. https://doi.org/10.1111/fare .12203 focused study will enhance our understanding Benson, J. J., & Coleman, M. (2016b). Older adults of this intimate tie as a new family form, its developing a preference for living apart together. challenges, and its implications for family ties, Journal of Marriage and Family, 78, 797–812. communities, and social change. Living in LAT https://doi.org/10.1111/jomf.12292 relationships means different things at different Bildtgård, T., & Öberg, P. (2015). Changing forms stages of the life course. We conclude that of intimacy among older people in late modern among older but not younger adults, LAT is society ( Förändrade imtimitetsformer bland äldre i generally a stable alternative to living with a det senmoderna samhället). Sociologisk Forskning, partner, not a stepping stone to establishing a 51, 5–32. common home. Billari, F. C., Rosina, A., Ranaldi, R., & Romano, C. (2008). Young adults living apart and together Applying ambivalence and life course con- (LAT) with parents: A three-level analysis of the cepts encourages a multilevel approach to Italian case. Regional Studies, 42, 625–639. https:// research that links the negotiation of LLAT doi.org/10.1080/00343400701543173 relationships to the family networks, social Borell, K., & Karlsson, S. G. (2003). Reconceptualis- institutions, structured social relations, and ing intimacy and ageing. Living apart together. In: cultural environments in which they are chosen S. Arber, K. Davidson, & J. Ginns (Eds.), Gender and worked out. As a new form of intimate and ageing: Changing roles and relationships (pp. tie, LLAT challenges institutionalized intimate 47–62). Buckingham: Open University Press. relationships and is more open to an unclut- Brothers, D. (2015). “Doing” LAT: Redoing gender and family in living apart together relationships tered exploration of how intimate relationships in later life (Doctoral dissertation). Retrieved are negotiated. We believe that pursuing our from http://rave.ohiolink.edu/etdc/view?acc_ proposed research agenda will extend our under- num=miami1429881367 standing of all intimate ties, including marriage, Brown, S. L., & Lin, I. F. (2012). The gray divorce as we study the realities of relationships rather revolution: Rising divorce among middle-aged Ambivalence and Later Life LAT Relationships 13

and older-adults, 1990–2010. Journals of Geron- 558–567. https://doi.org/10.1111/j.1741-3737. tology, Psychological Sciences and Social 2002.00558.x Sciences, 67, 731–741. https://doi.org/10.101093/ Cooney, T. M., & Dunne, K. (2001). Intimate rela- geronb/gbs089 tionships in later life. Current realities, future Brown, S. L., & Shinohara, S. K. (2013). Dating rela- prospects. Journal of Family Issues, 22, 838–858. tionships in older adulthood: A national portrait. https://doi.org/10.1177/019251301022007003 Journal of Marriage and Family, 75, 1194–1202. Cooney, T. M., Proulx, C. M., Snyder-Rivas, L. A., https://doi.org/10.1111/jomf.12065 & Benson, J. J. (2014). Role ambiguity among Bulcroft, K., & O’Connor, M. (1986). The impor- women providing care for ex-husbands. Journal of tance of dating relationships on quality of life Women and Aging, 26, 84–104. https://doi.org/10 for older persons. Family Relations, 35, 397–401. .1080/08952841.2014.859502 https://doi.org/10.2307/584367 Coulter, R., & Hu, Y. (2015). Living apart together Bulcroft, R. A., & Bulcroft, K. A. (1991). The nature and cohabitation intentions in Great Britain. and functions of dating in later life. Research Journal of Family Issues. Advance online publica- on Aging, 13, 244–260. https://doi.org/10.1177/ tion. https://doi.org/10.1177/0192513X15619461 0164027591132007 Coupland, J. (2000). Past the “perfect kind of age”? Calasanti, T., & Kiecolt, K. J. (2007). Diversity among Styling selves and relationships in over-50s dating late-life couples. Generations, 3, 10–17. advertisements. Journal of Communication, 50, Carr, D. (2004). The desire to date and remarry 9–30. https://doi.org/10.1111/J.1460-2466.2000 among older widows and widowers. Journal of .tb02850.x Marriage and Family, 66, 1051–1068. https://doi Creasey, G. L. (1993). The association between .org/10.1111/J.0022-2445.2004.000.78.x divorce and late adolescent grandchildren’s rela- Carstensen, L. L. (1995). Evidence from a life-span tions with grandparents. Journal of Youth and theory of socioemotional selectivity. Current Adolescence, 22, 513–529. https://doi.org/10. Directions in Psychological Science, 4, 151–156. 1007/BF01537713 https://doi.org/10.1111/1467-8721.ep11512261 Cruz, J. (2012). Remarriage rate in the U.S., 2010 Carter, J., Duncan, S., Stoilova, M., & Phillips, M. (Family Profile no. FP-12-14). Bowling Green, (2015). Sex, love and security: Accounts of dis- OH: National Center for Family & Marriage tance and commitment in living apart together rela- Research. tionships. Sociology, 50, 576–593. https://doi.org/ de Jong Gierveld, J. (2002). The dilemma of repartner- 10.1177/0038038515573689 ing: Considerations of older men and women enter- Cherlin, A. (2010). Demographic trends in the United ing new intimate relationships in later life. Age- States: A review of research in the 2000s. Journal ing International, 27, of Marriage and Family, 72, 403–419. https://doi 61–78. https://doi.org/10. .org/10.1111/j.1741-3737.2010.00710.x 1007/s12126-002-1015-z Cherlin, A. J., & Furstenberg, F. (1986). The new de Jong Gierveld, J. (2004). Remarriage, unmarried American grandparent. New York: Basic Books. cohabitation, living apart together: Partner rela- Connidis, I. A. (2010). Family ties & aging. 2nd ed. tionship following bereavement or divorce. Jour- Los Angeles: Pine Forge. nal of Marriage and Family, 66, 236–243. https:// Connidis, I. A. (2012). Theoretical directions for doi.org/10.1111/j.0022-2445.200400015.x studying family ties and aging. In R. Blieszner & V. de Jong Gierveld, J. (2015). Intra-couple caregiving H. Bedford (Eds.), Handbook of families and aging of older adults living apart together: Commit- (2nd ed.; pp. 35–60). Denver, CO: Praeger. ment and independence. Canadian Journal on Connidis, I. A. (2014). Age relations and family ties Aging, 347, 356–365. https://doi.org/10.1017/ over the life course: Spanning the macro–micro S0714980815000264 divide. Research in Human Development, 11, de Jong Gierveld, J., & Merz, E. (2013). Parents’ 291–308. https://doi.org/10.1080/15427609.2014 partnership decision making after divorce or .967050 widowhood: The role of stepchildren. Journal of Connidis, I. A. (2015). Exploring ambivalence in fam- Marriage and Family, 75, 1098–1113. https://doi ily ties: Progress and prospects. Journal of Mar- .org/10.1111/jomf.12061 riage and Family, 77, 77–95. https://doi.org/10 de Jong Gierveld, J., & Peeters, A. (2003). The .1111/jomf.12150 interweaving of repartnered older adults’ lives Connidis, I. A., & McMullin, J. A. (2002a). with their children and siblings. Ageing & Ambivalence, family ties, and doing sociology. Society, 23, 187–205. https://doi.org/10.1017/ Journal of Marriage and Family, 64, 594–601. S0144686X02001095 https://doi.org/10.1111/j.1741-3737.2002.00594.x DeLamater, J. (2012). Sexual expression in later life: Connidis, I. A., & McMullin, J. A. (2002b). Soci- A review and synthesis. Journal of Sex Research, ological ambivalence and family ties: A critical 49, 125–141. https://doi.org/10.1080/00224499 perspective. Journal of Marriage and Family, 64, .2011.603168 14 Journal of Marriage and Family

Dickson, F. C., Hughes, P. C., & Walker, K. L. Restructuring work and the life course (pp. 3–22). (2005). An exploratory investigation into dating Toronto: University of Toronto Press. among later-life women. Western Journal of Com- Karlsson, S. G., & Borell, K. (2002). Intimacy and munication, 69, 67–82. https://doi.org/10.1080/ autonomy, gender and ageing. Ageing Interna- 10570310500034196 tional, 27, 11–26. https://doi.org/10.1007/s12126- Dommermuth, L., Noack, T., & Wiik, K. A. (2009). 002-1012-2 Samlivsetablering–nye og vedvarende mønstre: Karlsson, S. G., & Borell, K. (2005). A home Gift, samboer eller bare kjærester?” [Union of their own: Women’s boundary work in formation–new and constants pattern: married, LAT-relationships. Journal of Aging Studies, cohabiting or just boyfriend/girlfriend?]. Sam- 19, 73–84. https://doi.org/10.1016/j.jaging.2004 funnsspeilet, 23, 13–19, Retrieved from http:// .03.008 www.ssb.no/ssp/utg/200901/ssp.pdf Karlsson, S. G., Johansson, S., Gerdner, A., & Borell, Duncan, S. (2015). Women’s agency in living apart K. (2007). Caring while living part. Journal of together: Constraint, strategy and vulnerability. Gerontological Social Work, 49, 3–27. https://doi The Sociological Review, 63, 589–607. https://doi .org/10.1300/J083v49n04_02 .org/10.1111/1467-954X.12184 Kobayashi, K. M., Funk, L., & Khan, M. M. (2016). Duncan, S., Carter, J., Phillips, M., Roseneil, S., & Constructing a sense of commitment in “living Stoilova, M. (2012). Legal rights for people who apart together” (LAT) relationships: Interpretive “live apart together”? Journal of Social Welfare agency and individualization. Current Sociology. and Family Law, 34, 443–458. https://doi.org/10 Advance online publication. https://doi.org/10 .1080/09649069.2012.753731 .1177/0011392116653237 Duncan, S., Carter, J., Phillips, M., Roseneil, S., & Koren, C., & Eisikovits, Z. (2011). Life beyond the Stoilova, M. (2014). Practices and perceptions of planned script: Accounts and secrecy of older per- living apart together. Family Science, 5, 1–10. sons in second couplehood in old age in a soci- https://doi.org/10.1080/19424620.2014.927382 ety in transition. Journal of Social and Personal Duncan, S., & Phillips, M. (2011). People who live Relationships, 28, 44–63. https://doi.org/10.1177/ apart together (LATs): New family form or just 0265407510385430 a stage? International Review of Sociology, 21, Koren, C., & Simhi, S. (2016). “As long as it’s good”: 513–532. https://doi.org/10.1080/03906701.2011 An intergenerational family perspective of bridg- .625660 ing gaps between reality and ideality of second Elder, G. H. (1994). Time, human agency, and social couplehood as a problem and as a solution. Ageing change: Perspectives on the life course. Social & Society, 36, 716–740. https://doi.org/10.1017/ Psychology Quarterly, 57, 4–15. So144686X4001482 Ermisch, J., & Siedler, T. (2008). Living apart Levesque, L. M., & Caron, S. L. (2004). Dating pref- together. In M. Brynin & J. Ermisch (Eds.), erences of women born between 1945 and 1960. Changing relationships (pp. 29–43). Hoboken, Journal of Family Issues, 25, 833–846. https://doi NJ: Routledge. .org/10.1177/0192513X03258300 Frankowski, A. C., & Clark, L. J. (2009). Sexual- Levin, I. (2004). Living Apart Together: A new family ity and intimacy in assisted living: Residents’ per- form. Current Sociology, 52, 223–240. https://doi spectives and experiences. Sexuality Research and .org/10.1177/0011392104041809 Social Policy, 6, 25–37. https://doi.org/10.1525/ Levin, I., & Trost, J. (1999). Living Apart Together. srsp.2009.6.4.25 Community, Work and Family, 2, 279–293. https:// Funk, L. M., & Kobayashi, K. M. (2016). From doi.org/10.1080/13668809908412186 motivations to accounts: An interpretive anal- Liefbroer, A. C., Poortman, A., & Seltzer, J. A. ysis of “living apart together” relationships in (2015). Why do intimate partners live apart? Evi- mid- to later-life couples. Journal of Family dence on LAT relationships across Europe. Demo- Issues, 37, 1101–1122. https://doi.org/10.1177/ graphic Research, 32, 251−286. https://doi.org/10 0192513X14529432 .4054/DemRes.2015.32.8 Giddens, A. (1991). Modernity and self-identity: Self Lin, I-F. & Brown, S. L. (2012). Unmarried Boomers and identity in late modern age. Cambridge, UK: confront old age: A national portrait. The Geron- Polity Press. tologist, 52, 153–165. https://doi.org/10.1093/ Haskey, J. (2005). Living arrangements in contempo- geront/gnr141 rary Britain: Having a partner who usually lives Lyssens-Danneboom, V., Eggermont, S., & Mortel- elsewhere and living apart together (LAT). Popu- mans, D. (2013). Living apart together (LAT) lation Trends, 122, 35–45. and law: Exploring legal expectations among Heinz, W. (2001). Work and the life course: A LAT individuals in Belgium. Social & Legal cosmopolitan-local perspective. In V. W. Marshall, Studies, 22, 357–376. https://doi.org/10.1177/ W. R. Heinz, H. Kruger, & A. Verma (Eds.), 0964663913478960 Ambivalence and Later Life LAT Relationships 15

Lyssens-Danneboom, V., & Mortelmans, D. (2015). Research, 17, 121–132. https://doi.org/10.1016/j Living apart together: Longing for the couple, .alcr.2012.01.004 enjoying being single. Family Science, 6, 11–22. Schutz, A. (1972). The phenomenology of the social https://doi.org/10.1080/19424620.2015.1009932 world. Portsmouth, NH: Heinemann. Mahay, J., & Lewin, A. C. (2007). Age and desire Setttersten, R. A., Jr. (2003). Propositions and to marry. Journal of Family Issues, 28, 706–723. controversies in life-course scholarship. In https://doi.org/10.1177/0192513X06297272 R. A. Settersten Jr. (Ed.), Invitation to the life Malta, S., & Farquharson, K. (2014). The initiation course: Toward new understandings of later life and progression of late-lite romantic relationships. (pp. 15–45). Amityville, NY: Baywood. Journal of Sociology, 50, 237–251. https://doi.org/ Stephure, R. J., Boon, S. D., MacKinnon, S. L., & 10.1177/144078331242254 Deveau, V. L. (2009). Internet initiated relation- McElhany, L. J. (1992). Dating and courtship in later ships: Associations between age and involvement years: A neglected topic of research. Generations, in online dating. Journal of Commuter-Mediated 16, 21–23. Communication, 14, 658–681. https://doi.org/10. McWilliams, S., & Barrett, A. E. (2014). Online 1111/j.1083-6101.2009.01457.x dating in middle and later life: Gendered expec- Stevens, N. (2004). Re-engaging: New partnerships in tations and experiences. Journal of Family late-life widowhood. In K. Davidson & G. Fennell Issues, 35, 411–436. https://doi.org/10.1177/ (Eds.), Intimacy in later life (pp. 47–64). London, 0192513X12468437 UK: Transaction Publishers. Milan, A., & Peters, A. (2003). Couples living apart. Stoilova, M., Roseneil, S., Carter, J., Duncan, S., & Canadian Social Trends, Summer, 2–6. Phillips, M. (2017). Constructions, reconstructions Moorman, S. M., Booth, A., & Fingerman, K. (2006). and deconstructions of “family” amongst people Women’s romantic relationships after widowhood. who live apart together (LATs). The British Jour- Journal of Family Studies, 27, 1281–1304. https:// nal of Sociology, 68, 78–96. https://doi.org/10. doi.org/10.1177/0192513x06289096 1111/1468-4446.12220 Pidduck, H. (2009). Queer kinship and ambivalence: Stoilova, M., Roseneil, S., Crowhurst, I., & Santos, A. Video autoethnographies by Jean Carlomusto and C. (2014). Living apart relationships in contempo- Richard Fung. GLQ: A Journal of Lesbian and rary Europe: Accounts of togetherness and apart- Gay Studies, 15, 441–468. https://doi.org/10.1215/ ness. Sociology, 48, 1075–1091. https://doi.org/10 10642684-2008-031 .1177/0038038514523697 Reczek, C. (2016). Ambivalence in gay and les- Strohm, C. Q., Seltzer, J. A., Cochran, S. D., & Mays, bian family relationships. Journal of Marriage and V. M. (2009). Living apart together relationships Family, 78, 644–659. https://doi.org/10.1111/jomf in the United States. Demographic Research, 13, .12308 177–214. https://doi.org/10.4054/DemRes.2009. Régnier-Loilier, A. (2015). Neither single nor in a 21.7 couple in France: What became of them three and Talbott, M. M. (1998). Older women’s attitudes six years later? In D. Mortelmans, I. Pasteels, A. towards men and remarriage. Journal of Aging Régnier-Loilier, D. Vignoli, & S. Mazzuco (Eds.), Studies, 12, 429–449. https://doi.org/10.1016/ Analysis of determinants and prevalence of LAT S0890-4065(98)90028-7 (pp. 29–53). Rostock, Germany: Max Planck Insti- Turcotte, M. (2013). Living apart together. Ottawa: tute for Demographic Research. Statistics Canada. Régnier-Loilier, A., Beaujouan, É., & Villeneuve- Upton-Davis, K. (2012). Living apart together rela- Gohalp, C. (2009). Neither single, nor in a couple: tionships (LAT): Severing intimacy from obliga- A study of living apart together in France. Demo- tion. Gender Issues, 29, 25–38. https://doi.org/10 graphic Research, 21, 75–108. https://doi.org/10 .1007/s12147-012-9110-2 .4054/DemRes.2009.21.4 Upton-Davis, K. (2015). Subverting gendered norms Reimondos, A., Evans, E., & Gray, E. (2011). of cohabitation: Living apart together for women Living-apart-together (LAT) relationships in over 45. Journal of Gender Studies, 24, 104–116. Australia. Family Matters, 87, 43–55. https://doi.org/10.1080/09589236.2013.861346 Roseneil, S. (2006). On not living with a partner: Watson, W. K., & Stelle, C. (2011). Dating for older Unpicking coupledom and cohabitation. Sociolog- women: Experiences and meaning of dating in later ical Research Online, 11, 1–24. Retrieved from life. Journal of Women & Aging, 23, 263–275. http://www.socresonline.org.uk/11/3/roseneil.html https://doi.org/10.1080/08952841.2011.587732 Schenk, N., & Dykstra, P. A. (2012). Continuity Yodanis, C., & Lauer, S. (2014). Is marriage individu- and change in intergenerational family relation- alized? What couples actually do. Journal of Fam- ships: An examination of shifts in relationship type ily Theory & Review, 6, 184–197. https://doi.org/ over a three-year period. Advances in Life Course 10.1111/jftr.12038