1Union formation involves a number of stages, as does union dissolution, and new couples often spend an initial period in a non-cohabiting intimate relationship. Yet while certain couples never share the same dwelling, “living apart together”  has not developed widely as a long-term lifestyle option. Claude Martin in France, and Andrew Cherlin and Caitlin Cross-Barnet in the United States have studied a symmetrical phenomenon, that of couples who continue to live together while considering themselves to be separated. In this article, they draw together their analyses to describe an arrangement which, while marginal, reveals situations where residential separation is not possible, either because of the need to keep up appearances, often for the children’s sake, or because total separation is too frightening or living in separate homes is unaffordable. Beyond the differences between the two countries and the two survey fields, the authors analyse the ways in which persons who “live together apart” describe their loveless relationship that has led to explicit conjugal separation within a shared home.
2In the midst of the 2008 banking crisis and its ripple effect on the world economy, a phenomenon barely visible until then was highlighted by the media on both sides of the Atlantic: “The housing crisis is even forcing couples who want to separate into involuntary cohabitation. (…) The fear of not finding a new place to live is leading to untenable situations. Many couples are sharing the same home without actually ‘being together’”,  wrote Michaël Hadjenberg in an article on the Mediapart website published in May 2008. In December of the same year, an article in the Seattle Times entitled “Couples staying together because of poor economy”,  described the same phenomenon in the United States.
3Do these cases of “forced cohabitation” represent a whole new set of family situations which deserve to be identified and qualified as such? And if so, what should they be called: involuntary cohabitation? cohabiting separation? In any event, these situations bear witness to the constraints and fears associated with contemporary conjugal trajectories. The current economic and housing crisis may be aggravating this phenomenon, and French journalists and lawyers are detecting the first signs of such a trend.
4In the early 1990s, several sociological studies revealed the existence of new conjugal situations which we, along with others, termed “living apart together” (LAT) (Le Gall and Martin, 1988; Martin, 1994; Levin and Trost, 1999; Levin, 2004). Since then, the expression has been used to describe couples who, voluntarily or otherwise, on a temporary or permanent basis, and for a variety of reasons, do not live together and maintain two separate homes (Régnier-Loilier et al., 2009). One reason for these conjugal arrangements is linked to the way young couples begin their relationships, or to their working careers, which may oblige them to live in two different cities, for example. Another reason may be the desire to avoid exposing the children of a previous union to the presence of a new partner. By living in separate homes, time spent as a couple can be separated from family time, and the partners can choose the right moment to reform a new family, sometimes after a long waiting period (Martin, 2001). Such arrangements also exist among older adults who, after their children have left home, sometimes choose to live separately from their partner (Caradec, 1996a). In all cases, they must have the financial means to maintain two separate homes. These situations also raise questions about the “objective” indicators of conjugal life. Perhaps it is individuals themselves who should define whether or not they are in a conjugal relationship.
5Analysis of LAT raises questions similar to those we aim to discuss here. For example, the viewpoints of the persons concerned must be taken into account, since certain romantic and sexual relationships (teenage romances for example) are not considered by the protagonists as conjugal relationships. In all logic, the first requirement of an LAT relationship is that the persons concerned should see themselves as a couple, and even be perceived as a couple by others.  Another question concerns the significance of these practices, and the profiles of the persons concerned. Some authors, such as Irene Levin or Sasha Roseneil, see LAT relationships as a “new family form”, chosen by partners who wish to enjoy the intimacy of a relationship while maintaining their independence and their social networks by living apart (a both/and solution, Levin, 2004), or who give priority to friendships over romantic and sexual relationships (Roseneil, 2006). Levin thus posits that these behaviours are especially prevalent in societies where cohabitation is a widely accepted institution, as is the case in Scandinavian countries. Other scholars tend to focus on the extreme heterogeneity of so-called LAT couples, contrasting voluntary (or deliberate) separation with involuntary separation linked to a range of constraints, primarily those of the labour market (such as commuter marriages, Haskey and Lewis, 2006). By analogy, we propose to qualify the situation of estranged couples who carry on living under the same roof despite their desire to separate as “living together apart” (LTA).  These are households comprising at least two adults previously in a cohabiting relationship, who no longer see themselves as a couple but who continue to live together for a variety of reasons: one or the other former partner cannot afford to set up home elsewhere; fear of the financial consequences of separation; concern to maintain bonds with shared children; hope of returning to “normal” conjugal life.
6These very particular conjugal situations tend to be invisible and therefore difficult for sociologists to study.  In fact, they have very probably always existed in one form or other. On the basis of two exploratory surveys conducted in parallel in the United States and in France, this article aims to identify the reasons why, according to the protagonists, this type of situation arises, and to reflect on the meaning of cohabitation and of the family relationship (as a combination of conjugal and parental relationships). All in all, despite very different conceptions of marriage and cohabitation in France and the United States, LTA relationships and their significance for the persons concerned are quite similar on either side of the Atlantic. The testimonies of the persons concerned show how strongly financial circumstances weigh upon conjugal trajectories, and highlight the impact of the economic crisis and the fears it arouses. But they also reveal the strong value placed on parent-child relationships and on the parenting role.
I – Stay or leave? An irresolvable end to life as a couple
7Is there anything really new about these situations? Should we be alarmed, as is all too often the case, by a phenomenon whose only new feature is perhaps its association with the current economic crisis? One thing is sure: the idea that conjugal life may conceal situations of estrangement and that couples may stay together to avoid violating the dominant family norm is nothing new. This type of situation lies at the heart of the secular debate on divorce. In his polemical pamphlet written in the mid-seventeenth century, John Milton used these situations of marital mismatch as one of his main arguments in defence of divorce, vigorously contesting the principle of the indissolubility of marriage imposed by the Roman Catholic Church. Interpreting the first verses of Genesis, he refutes this ordinance of indissolubility, whose hypocrisy he denounces in these terms:
For although God in the first ordaining of marriage taught us to what end he did it, in words expressly implying the apt and cheerful conversation of man with woman, to comfort and refresh him against the evil of solitary life, not mentioning the purpose of generation till afterwards, as being but a secondary end in dignity (…); yet now, if any two be but once handed in the church, and have tasted in any sort the nuptial bed, let them find themselves never so mistaken in their dispositions through any error, concealment, or misadventure, that through their different tempers, thoughts, and constitutions, they can neither be to one another a remedy against loneliness, nor live in any union or contentment all their days, yet they shall (…) be made, spight of antipathy to fadge together, and combine as they may to their unspeakable wearisomeness and despair of all sociable delight in the ordinance which God established to that very end. What a calamity is this (…). All which we can refer justly to no other author than the canon law and her adherents (…).
9Numerous authors have looked for ways to preserve morality while recognizing marital discontent (Martin, 2006). The Enlightenment philosopher Christian Wolff (1679-1754) proposed that divorce by mutual consent be authorized for couples with no children, or whose children had completed their education, but that it be prohibited in all other cases – thereby obliging them to fulfil their parental duties. In the nineteenth century, Proudhon, loyal to the principle of indissolubility, “was forced to invent a legal remedy even more audacious than divorce itself: the estranged spouses would not be authorized to remarry, but would be free to relieve their solitude with mistresses recognized by law. He thus devised the status of legal mistress, much like that of secondary wife” (Carbonnier, 1988, p. 233). Things unquestionably become more difficult when the couple already has children. If divorce is possible, then its potential impact on the children’s well-being and development becomes a whole new area of debate.
10The literature also abounds with painful and pathetic stories of individuals imprisoned in conjugal misery due to a mismatch or a spouse imposed upon them to satisfy the demands of a “good marriage”. One such example is the tragic destiny of Gustave Flaubert’s Emma Bovary, who aspires to a world of amorous passion but instead discovers the cowardice of men. But the complexity and ambivalence of these loveless marriages are perhaps even better illustrated by Marcel Jouhandeau in his Chroniques maritales. Of marriage, he writes: “My mother, whose letters I am rereading, constantly talked to me as a young man about marriage as a downward spiral from which escape is impossible. One sees that it is a poison, but one becomes accustomed to its taste. How, then, can one renounce marriage without renouncing one’s very self?” (p. 163). So estrangement leads not to the end of the couple but rather to a routine, slow experience of falling out of love. Countless examples of such ambivalent feelings exist in fiction and in the cinema. We could mention Tolstoy (and his disgust for the bourgeois institution of marriage developed in The Kreutzer Sonata published in 1889), Mauriac (and, for example, Nœud de vipères [The Knot of Vipers] published in 1933), Lewisohn (The Case of Mr Crump published in 1926), but also Simenon (with Le Chat, [The Cat] published in 1967 and adapted for the cinema by Pierre Granier-Deferre in 1970 with Simone Signoret and Jean Gabin) and many more. With greater acuity than any sociologist, these authors describe the fragile equilibrium of marital life as it shifts between hopes, aspirations and disappointed dreams, between a thirst for passion and a desire for the comfort of habit and convention, between intimacy and distance, complicity and irritation, even hate and inescapable marital torment. The history of a marriage appears to unfold partly without its protagonists, as a history devoid of choice, built upon doubt and fear, upon the hope of compromise, a lesser evil. In The Big Picture, Douglas Kennedy, a popular contemporary author, describes the process of “demarriage” and evokes this tension between the appeal of recklessness and the comfort of responsibility:
We all crave latitude in life, yet simultaneously dig ourselves deeper into domestic entrapment. We may dream of traveling light but accumulate as much as we can to keep us burdened and rooted to one spot. And we have no one to blame but ourselves. Because – though we all muse on the theme of escape – we still find the notion of responsibility irresistible. The career, the house, the dependents, the debt – it grounds us. Provides us with a necessary security, a reason to get up in the morning. It narrows choice and, ergo, gives us certainty. And though just about every man I know rails against being so cul-de-saced by domestic burden, we all embrace it. Embrace it with a vengeance.
12This short excursion into literature suggests that the process which leads to involuntary cohabitation is long, and marked by doubt and ambivalence. So should the social sciences take an interest in this process of “falling out of love”? Probably yes in cases of cohabiting adults who once had a romantic relationship but no longer see themselves as a couple. While the pressure of conventions, social norms and beliefs doubtless played a major role in the past, the liberalization of divorce and its destigmatization have led to a radically different situation today. This raises the question of why such cohabiting non-couples continue to exist.
13A comparison between the United States and France may help to provide an answer, since the value placed upon marriage and cohabitation in these two countries is not identical. Likewise, conjugal trajectories may be marked by different forms of turnover: movements in and out of marriage in one case, and a lesser attachment to the marital institution in the other. Reflecting these differences of representation, this comparison might reveal very contrasting attitudes to the fact of “staying together” expressed by persons who no longer see themselves as a couple.
II – Different contexts in France and the United States
Marriage is central to American conjugal life
14As argued by Andrew Cherlin in a recent book (Cherlin, 2009), what sets the United States apart from many European countries in terms of family life, is the “strength of marriage as a cultural ideal”. This centrality of marriage is visible in both social behaviours and public policy. “The promarriage message is deeply embedded in American culture and history” (p. 182). And there has been a resurgence of interest in recent years, with the Bush administration releasing 300 million dollars to support marriage (marriage preparation classes, lower child benefits for births outside marriage). As he asserted in January 2004 in his State of the Union address, “our nation must defend the sanctity of marriage”, even if this entails an amendment to the constitution. 
15But this pressure to marry, combined with the desire to get the most out of married life while satisfying individual aspirations, has led to the American paradox: very high marriage and divorce rates that have produced what Cherlin calls the “marriage-go-round”. Although marriage rates have fallen in all developed countries over the last four decades, there is still a substantial gap between the American situation and that of many European countries, including France. In the United States, the marriage rate dropped from 10.8 marriages per 1,000 inhabitants in 1970 to 7.1 per 1,000 in 2008, but is still well above that of France, where it fell from 8 to 4.2 per 1,000 over the same period. The difference is equally large for divorce, with the United States again recording the higher rate (a crude divorce rate of 3.5 per 1,000 inhabitants in 2008 versus 2 per 1,000 in France). A similar contrast is observed for non-marital births, which account for around 40% of American births (half to cohabiting couples, and half to lone mothers) but more than 50% in France (mainly to cohabiting couples).
16Yet the trajectories of entry into adulthood in the United States also show large disparities in attitudes and behaviours across social groups. While young, college-educated, socially integrated adults still have relatively classic trajectories, comprising stable and predictable sequences – partner met during or immediately after college, marriage followed by the birth of one or more children – this is not the case among the working classes, nor even among the lower fringes of the middle class. In environments where cultural capital is low, children are often born before marriage and in many cases the marriage never takes place. Some young women in the most disadvantaged social groups even appear to see motherhood as a means of obtaining status and social recognition, and this may explain the large difference in the proportions of young lone mothers in the United States and in France (Martin, 2010).
17While the disadvantaged do not forfeit their right to have children, access to marriage, on the other hand, appears to be dependent upon a stable economic situation. This is so for men especially, who are expected to assume the breadwinner role. Indeed, financial insecurity appears to make many men “unmarriageable”, or worthless on the marriage market (Wilson, 1987 and 1996). In a survey of women in poor districts of Philadelphia, Edin and Kefalas (2005) said that the respondents did not trust the men available to them to earn a living, and also remain faithful, refrain from illegal activity, treat them as equals, and make their family their top priority, conditions the women felt were all necessary to make a partner marriageable.
18Trajectories are much more “disorderly” in lower socioeconomic groups: individuals may cohabit with a partner while at high school, college or in vocational training; they may have a child while still a student or in training; they may have children with more than one partner and may still be unmarried five, ten or fifteen years after the birth of the first child. Moreover, these fertile cohabiting relationships are unstable, much more so than marriage. So in the United States, unlike France, cohabitation is seen as a secondary form of union. And even among couples who have married, divorce is also more frequent among lower socioeconomic groups, leading to considerable variations in the risk of divorce across social classes. For example, one-third of first marriages among women with a high-school level of education end in divorce within the first five years, a very high rate for such a short period of marital life, compared with just 13% for women with a college degree (Cherlin, 2010).
19It is precisely these working-class couples  who, according to Andrew Cherlin, have the most unhappy marriages and the highest stress levels (Cherlin, 2009, p. 169):
The working-class wives said they had taken jobs because the families needed the money, not to develop careers, and many of them wanted to work fewer hours or not at all. Many thought that the amount of housework and child care they did was unfair compared to how much their husbands did. Husbands, for their part, complained that their wives came home from work irritable and tense and that their jobs interfered with the life of the family. Both wives and husbands were more likely to say that they had problems in their marriage because one of them got angry easily, was critical or moody, or wouldn’t talk to the other one. They were also more likely to say that of the thought of getting a divorce or separation had crossed their minds or that they had discussed the idea with family members or close friends.
21Recent demographic trends in the United States thus suggest that marital and family trajectories are becoming increasingly polarized. Educational level is not – or no longer – an obstacle to marriage, but rather a predictor of the chances of marrying, while low levels of education and income increase the likelihood that a young adult will opt for some form of temporary or longer-term cohabitation, often following a birth. Based on the survey on Time, Love and Cash in Couples with Children, Reed (2006) shows that almost three-quarters of respondents began cohabiting after a pregnancy. Cohabitation enables couples to share costs and to raise their child together. But the conjugal bond is weaker, and the distinction between cohabitation and singlehood is unclear. Roy et al. (2008) use the term “suspended relationships” to characterize these fragile, uncommitted, unstable and uncertain conjugal bonds.
The routinization of cohabitation in France
22The situation appears to be quite different in France, where alternatives to marriage (cohabitation or civil partnership [PACS])  and non-marital births are now widely accepted. There is no social divide in conjugal trajectories comparable to that of the United States, and when such social disparities are observed, they tend to be in a direction opposite to that observed across the Atlantic. For example, women in higher socioeconomic groups (in terms of occupational category or educational level) separate more frequently than women from lower groups. And while men’s behaviour is more homogeneous across socioeconomic groups, men from lower groups have lower risks of separation (Beaujouan, 2009).
23Likewise, cohabitation and non-marital births in France are not specifically associated with low-income groups. The recognition of cohabiting couples, whether or not they have registered a PACS civil partnership, narrows the gap between cohabitation and marriage, despite the fact that certain legal differences persist, and even though certain couples and families, such as gay and lesbian families, are excluded from marriage. One difference perhaps lies in the determinants of cohabitation in the United States, where this choice is more often dictated by a lack of resources than by a desire to form a union based on different attitudes or values. In France, by contrast, cohabiting couples, who are often highly educated, favour autonomy, freedom and relational quality over the status acquired through marriage. This may reflect the impact of the powerful pro-marriage ideology in the United States, which pushes the most marginalized people into stigmatized relationships, considered as negative, deviant or undesirable. Yet, it can be argued that cohabitation does not necessarily signify non-commitment; it is neither a suspended relationship nor an uncommitted relationship. Commitment is represented precisely by the fact of having children. And on this point, the situations of the two countries are perhaps more similar than it would seem.
24As observed in the United States, economic and social insecurity may now also be weighing more heavily on family trajectories in France. While the links between unemployment and divorce, or between a job loss and marital instability are perhaps less clear and, above all, less well documented in France than in the United States, the few available studies all attest to their influence (Paugam et al., 1993; Martin, 1997; Commaille, 1999). Growing employment insecurity, non-standard working hours and fears of a worsening job market also have similar effects, placing additional strain on marital and intrafamilial relationships.
25In short, the disparities observed between the United States and France in the practices of cohabitation or in the differences in marital trajectories across social groups should be analysed with caution. Despite different cultural and political contexts and contrasting social norms relating to marriage, we can make a first hypothesis that the socioeconomic and employment conditions of households have similar effects on marital and social trajectories in both countries. The fragility of marital bonds may thus be partly attributable to these external threats and pressures and may concern a broad range of social groups. The family is a reflection of these conditions, suggesting that it may not always be pertinent to focus on individuals’ attitudes and aspirations to grasp the main reasons for their behaviour. Second, we posit that one of the consequences of family transformations may be a shift from the family bond of the marital relationship (considered as fragile) to that of the parenting relationship (more unconditional).
III – Living together apart: The findings of two exploratory surveys
The Three-City Study in the United States
26The US data used in this article are drawn from the Three-City Study  conducted in three disadvantaged neighbourhoods of Boston, Chicago and San Antonio to examine the impact of welfare reform on the lives of low-income African American, Latino, Hispanic, and non-Hispanic White families (Winston et al., 1999). The survey comprises three interrelated components, including a quantitative survey of a large sample of 2,400 households and an in-depth ethnographic survey of 256 families. Here, we focus on 18 cases of LTA relationships identified via this ethnographic study.
27The 2,400 mothers and their children included in the Three-Cities Study were selected in the same low-income neighbourhoods as the families in the ethnographic study. The quantitative survey took place between June 1999 and December 2000. Among mothers who reported being neither married nor cohabiting at the time of the survey, 25 (around 2%) stated that they shared their dwelling with an unrelated man who was the father of at least one of the children or was of an appropriate age to be her partner. These women tended to be younger than average (around 28, versus a mean of 33 for the survey sample as a whole) and were more likely to be African American than White or Latina. The number of children in these households was not significantly different from the overall sample average, however.
28The ethnographic study was conducted between 1999 and 2002. Families were visited an average of once or twice per month for 12 to 18 months and then every six months over the following two or three years. Fieldnotes and interview transcripts were processed on a thematic basis, covering employment, childcare, living environment, intimate relationships. Using the respondents’ personal data, it was possible to identify 18 families who, at some point during the fieldwork, spent time in an LTA relationship. They are a group of women, aged 18-42, with the following characteristics: has previously lived with a partner, has had one or more children with that same partner, no longer considers herself to be in a relationship with that partner but is still living with that partner (who may be the father of one or more of the children in the household).
29Close attention was paid to the way these women referred to the cohabiting adult. Several expressions are used repeatedly, such as “roommate”, “tenant” and “baby daddy”. These women also described the nature of their relationship, often insisting that the male partner was not a member of the household even when he obviously lived in the home. “We live together, but we are not really together”.
30Most of these women said that they didn’t really trust these men, or trusted them only to fulfil certain roles, such as meeting financial obligations and/or doing their share of parenting. They gave three main reasons to explain this situation: housing problems, often linked to financial difficulties (“otherwise he would be in the street”); coparenting needs and the importance of fathering ties for the children; and last, a desire for social legitimacy by creating the appearance of a conventional family.
31Four women were engaged in LTAs solely for parenting reasons and stressed the importance of the man’s fathering role. Five other women used this same argument, but added that their partner had nowhere else to live. In such cases, the benefits are shared: the child(ren) keep their two parents and their familiar home environment, the mother receives help with parenting and the father remains in contact with his child(ren). For these mothers, an important advantage of this arrangement is the continued presence of the father, which averts one of the most devastating consequences of separation: fatherlessness. The women may nonetheless have a new partner, but they do not live with him. Maintaining a new “remote” sexual relationship is also a means of avoiding social disapproval for women with children born to different fathers.
32This was the case for Joanne, a White woman who shared her home with Sonny, a Hispanic with whom she had an 11-year relationship history riddled with instability and domestic violence. For Joanne, Sonny was no longer her partner, but had a vital role to play as the father of their six children. She says: “We’re not married. We’re bound by children”. Sonny assumed his fathering role by helping with shopping and sometimes watching the children or taking them to his mother’s when Joanne worked at a local fast food restaurant. Sonny made sporadic financial contributions to the household, but they were not Joanne’s primary motivation for allowing him to stay. Most importantly, his residence in the home allowed the children to have a present and involved father. Joanne had only had one boyfriend since their break-up, but she made sure that Sonny was the only father to her children and would never bring another man into the house because she thought it would be disrespectful and confusing to them. Joanne was clear that they were not intimate, however, as that would give Sonny the kind of control over her that she felt a husband or cohabiting partner expected to have.
33Similarly, Yasmin, a 23-year old Puerto Rican mother of a preschooler in Chicago, allowed her former boyfriend to live with her in part because she saw him as the “perfect father”. She did not trust him as a partner, however, because of his previous infidelities. So she laid down her conditions for allowing him to live in her home: he had to do what she told him to do, pay the bills, do the housework and look after the child. In fact, several women mentioned this “pay and stay” rule. While the man’s contribution is modest, the women often need it, as in the case of Tonya, who had put up with the presence of Curtis, her “recliner tenant” for a year but without ever giving him to the key to their apartment. Tonya who was welfare-reliant, and whose mother, sisters and elder daughter also lived in poverty, took a weekly rent of fifty dollars from Curtis, who would have otherwise been on the street.
34In LTAs, the women’s demands are different from those of the men. Women who provide a roof for the father of their children expect him to contribute financially to the household and do housework, but above all to help with childrearing and/or day-to-day childcare. Men who provide housing for the mothers of their children may ask in return for sexual favours from their former girlfriend, and for help with housework and, of course, childcare. These LTA relationships, often imposed by circumstance, may involve violence, and sexual violence especially. They certainly generate considerable frustration.
35In some cases, women accept these compromises because they feel obligated to the father of their children. Lizzy, a mother in Boston, intermittently allowed her ex-husband and the father of her two grown children to live in her apartment, though they had separated about 20 years earlier and had ultimately divorced. She was simply doing him a favour: “he just needed a place to stay, but he is not in my life”. He met no needs of hers, had no resources, and even brought potential danger to the household as he had made his living from various illegal enterprises. But their shared history was sufficient to justify her support: “You don’t give up on the father of your children”.
36In other cases, the main reason for this arrangement is to present as a conventional family to relatives and neighbours. This was the case for Marka, a Puerto Rican mother aged 26 who had an on-and-off relationship with her partner John, involving frequent domestic violence. Despite several attempts to get out of the relationship, she had a daughter with John at age 22. Conventional parenting ties were especially important to Marka, for her own sake as well as her daughter’s. Soon after the birth, John spent three years in prison, but was anxious to resume his relationship with Marka upon his release. Despite her doubts, and though she said “I don’t feel anything for him anymore”, she allowed John to move back in with her and later conceived another child with him. She was opposed to having children with several different men, so continued to live with John while keeping their relationship on a low key. It is noteworthy that Hispanic mothers placed considerable import on conventional families and the appearance of respectability.
37In this survey, the LTA couples belonged, by construction, to low-income groups, and only women’s viewpoints were recorded. The lack of material resources is a determinant of these ambiguous, unstable arrangements, not only because it prevents women from entering more stable relationships, such as marriage, but also because LTA is often the only way to avoid a worse fate (homelessness for certain men, and even for some of these women), or to create economies of scale while maintaining fathering ties. But the importance of the parenting dimension is striking. LTA relationships are often centered on shared parenthood.
An exploratory study in France
38The very different exploratory study conducted in France concerns persons in middle-income categories.  Like the American cases, the persons interviewed considered that they were no longer in a relationship with their partner, but still formed a family with their shared children. Unlike the Three-Cities Study which drew LTA couples from an extensive survey of low-income neighbourhoods, the specific purpose of the French study was to explore LTA relationships with a view to understanding the respondents’ reasons for their current situation. In-depth retrospective interviews were conducted to record their feelings about these arrangements and the reasons behind the “demarriage” process of mutual estrangement that led to these atypical family situations. The testimonies also reveal the respondents’ dilemmas, hesitations and fears. Despite undeniable differences between the French and American cases, the main arguments are very similar: parenting obligations on the one hand, and fear of financial hardship on the other.
39After twenty years of life together and almost seven years of “marital warfare” which ended in an LTA arrangement, Roland (age 49, high-school teacher) makes the following diagnosis:
Our daily life got worse and worse over the years. We couldn’t stand each other anymore. Things got so bad that I often thought about leaving. Several times, after a scene, I spent a night or two in a hotel, because we were so busy fighting each other that we even forgot about the children and flung insults around while they were right in front of us… But I was actually scared about being separated from the kids. She often threatened to take them far away, and to wage war from a distance via a really nasty divorce… Like, “I’ll really make you pay”. So I decided to wait it out, to stay with the kids and see them grow up… Now I realize that I was also being a coward. They’ve grown up with me, but in the midst of a marital battlefield and I’m just starting to see the damage we’ve done. I can’t count the number of scenes and shouting matches they would have avoided if we’d had the courage to break up… Now I’m seeing a psychologist and so are my daughters. It’s obvious that they’ve paid a heavy price and that they’re deeply scarred by their hellish family life… We have something that needs repairing and I’m sure that they hate us for being so cowardly and selfish.
41Roland focuses on his dilemma, but also on his guilt towards his children. He thought that waiting it out would be the least unacceptable way to avoid “all-out warfare” and to preserve his all-important role as a father. But this compromise solution – continuing to be a father but not a husband – also had consequences in terms of conflict, tension and unhappiness, for the children especially, and he needs to make amends. For Roland, breaking up would have been an act of courage.
42And this is what Jacques has to say. He is a 54-year-old restaurant owner, father of three children aged 20, 18 and 11, whose wife stayed at home for ten years after the birth of their first child before finding a job as a secretary. Jacques is still married and says he has been “stuck” in an LTA for almost ten years. He too describes the steady breakdown of the marital relationship and the couple’s irreversible descent into lovelessness, while perhaps, for him, avoiding the worst: the loss of fathering ties or even the risk of financial hardship.
We’ve always had a very stormy relationship. To start with, we were passionately, sometimes ferociously, in love or should I say “in fusion”… But with every new clash, every new crisis, and every new shouting match, this fusion which asked too much of us, which asked everything of us, turned into pain, and sometimes into violence. We’ve inflicted so many wounds on each other, only to arrive where we are today… Now, just the idea of touching her has become dangerous. The sexuality which once gave us so much pleasure now exposes us to a danger – that of making up – when we know from experience that making up through pillow talk doesn’t solve the basic problem. We’re simply not made for each other.
44Jacques stresses the importance of the children in the choice of staying together and also expresses his fear of the financial hardship that a divorce would bring.
There’s no doubt that we’re unhappy together. We’ve been sleeping in separate bedrooms for practically ten years and when we’re forced to share the same bed, we keep well apart. It’s mutual cold shoulder… But it’s hard for us to split up for thousands of reasons. The children, to begin with. We should’ve done it more than ten years ago, before the youngest was born. Today, I suppose I feel we should wait for the children to grow up and be independent. Things might be even worse if we separated… I’m also worried about running out of money. Divorce is a big leap into hardship… We’re frightened of the future, of losing our spending power, and even of becoming poor. But sometimes I think that’s not what counts. We stay together because we’ve built this dependence out of habit.
46The two following excerpts illustrate these concerns about the effects of divorce, the first focusing on fears of its adverse impact on the children, and the second on the financial consequences of separation.
Although I often think about divorce, it immediately reminds me about everything I’ve heard and read about the children of divorcees. How could I be selfish enough to ignore all these warnings about its disastrous effects on children, how it makes them suffer, fall behind at school, get into drugs, or worse, commit suicide? No, I haven’t the strength to take a decision like that, though our family life is terrible.
I remember vividly going to see a lawyer. It was about 15 years ago. I made it clear from the start that I wasn’t sure, and that I was frightened by the idea of divorce. After saying a few things about my daily life, our constant fighting matches, our separate bedrooms, and so on, I explained that I had an insecure job and that a break-up would make me destitute. The lawyer made things quite clear to me. She told me what was in store – selling our apartment, dividing up all our belongings, legal battles – all that sort of thing. I really felt as if a gaping hole was opening up in front of me and she could see that I was scared… At the end of the appointment she simply said: “I’ll see you again in ten years, I can tell that you’re not ready”. That was really painful. And things haven’t moved an inch since then: it was more than 15 years ago now.
49Faced with such dilemmas, LTA couples play a waiting game, sometimes in the hope that things will get better or that they will recover their freedom after the children have grown up and left home, but sometimes out of a reluctance to break with routine and familiar family life. Aside from a sense of marital failure and an emotional desert, everyday life goes on as before. The relative comfort of habit despite conjugal breakdown was a theme that featured in several of the testimonies we recorded. As was the case for some of the LTA couples in the United States, the attitudes of family, friends and neighbours may also play a role in keeping the couple together. Patricia, aged 49, a civil servant in an LTA for just over three years, explains:
Although our relationship is on low ebb, our friends and family make us feel as if we form a family nonetheless, and that we’re right to hold onto what is most important: a family for our children. Yes, it’s definitely the children who keep us together.
51In his survey of couples in retirement, Vincent Caradec (1996a) had already noted this tendency to find comfort in habit. Mr Berg mentions it explicitly when explaining his reluctance to leave his wife and the advantages of preserving a certain family unity for the children’s sake:
You know the other person. You know them inside out. You know everything about them, so it makes life more comfortable in the sense that you don’t need to say, for example: “No, I don’t like potatoes” or “I never eat french fries”. That’s a very routine example. There’s a sort of breathing mechanism which is complicated because you don’t get along any more, but simple compared to a new life with someone else, where you’d have to learn everything from scratch, rebuild things from the bottom up. It’s paradoxical what I’m saying, but it’s two entities that come together somewhere. The fact that you know the other person so well; you don’t love them anymore but you know them so intimately that it makes life simpler, because they know how to steer the course with you
We have children, and we try to give them an image of their parents which isn’t too negative… The big question is do we give a more negative image to them by staying together in a loveless relationship than if we separated? But we get the sense that we are useful somewhere, despite our arguments, which they’ve more or less got used to by now, so we want to go as far as we can with them
55At retirement, these non-choices and dilemmas come to the surface again. Vincent Caradec cites this letter to the advice column of Notre temps magazine:
I’m 60 years old and I’ve just retired. My husband, who I’ve never got on well with, has been retired for two years. All my life I’ve kept on going because of the children, but now they’ve left to live their own lives and I don’t know what to do. My husband is overbearing and coarse, sometimes even brutish. He has driven away all our friends, and I can’t see myself putting up with him all day long. On the other hand, getting separated at our age comes with so many problems! I’m also scared about how he will react if I bring up the idea… In your opinion, what can I do?
57In this case, a radical life-changing decision is not seen as a realistic option. It seems to be too late to start a whole new life. At this age, the fear of loneliness adds to the earlier fears of separation.
IV – Discussion and conclusion
58No definitive conclusions can be drawn from an exploration of this kind. However, it gives pointers for new avenues to follow in more systematic future studies. First, it sheds light on the usefulness and the limits of a comparison between France and the United States. Though the cases we have looked at so far in both countries are clearly not comparable (lone mother households in low-income neighbourhoods in the United States; a small sample of middle-income men and women in France), they nonetheless prompt a number of hypotheses. In the data collected so far, we have been struck by the similarities found in areas where more differences might be expected. In the American context of the marriage-go-round, we observed the duality of conjugal and familial trajectories. Staying in a loveless household is not compatible with this model, except in cases of necessity, which arise when the economic resources to marry are not available. In other words, LTA relationships may exist partly because the norm of marriage is inaccessible. In the French context, the social divide would appear, at first sight, to be less relevant for explaining the decision to marry or not marry. Yet although the ideal of the “good divorce” seems to be preferable in a context where the children’s wellbeing depends upon parental harmony, staying together in a loveless relationship also seems to be incompatible with the model, except, here again, when material well-being is threatened by separation. In sum, in the first case, marriage is only possible when certain financial conditions are met, while in the second, it is the option of divorce which may be unaffordable.
59Based on these initial remarks, there is clearly a difference between men’s and women’s viewpoints in such situations, as already noted for the trade-offs negotiated between LTA partners. But this gender perspective also applies to middle-income cases, because the financial impact of separation is not the same for men and for women, and neither are the opportunities for entering or remaining in employment. A systematic survey of comparable couples would be needed to examine the issues in more depth, given that middle-income Americans destabilized by the economic crisis are quite likely to adopt tactics similar to those of the French households we interviewed.
60Let us return to the common features of the two studies, and the further reflections they inspire. The first concerns the importance of the parenting role over and above the conjugal relationship. While relations between partners becomes increasingly fraught over the long term, the unconditional bond is between the parent and child. “You don’t divorce from your children” as the saying goes. This argument is used recurrently in the testimonies of LTA couples, in both the United States and France. The need to maintain parenting ties is one of the main reasons given to justify LTA relationships, and this doubtless goes hand in hand with the frequent social criticism voiced against parents for their ineffective, selfish and even irresponsible attitude towards parenting and childrearing duties. A remarkable number of measures to sanction certain parents (withdrawal of family allowances; parenting contracts, etc.) have been proposed in recent years, not only in France but also in other European countries, where a “parenting policy” is now taking shape, notably in the form of specific parenting programmes. Such policies are reviving the notion of a “family police” intended to hammer home the parents’ responsibility in promoting their children’s well-being and success, and likewise exacerbate their guilt if they fail in their task (Donzelot, 2005; Martin, 2003). In the United States, this pressure on parents stigmatizes lone mothers, mothers with children by several partners, mothers seen as responsible for their children’s violent and delinquent behaviour, and even devalues cohabitation, reputedly less stable than marriage.
61While social pressure and compliance with bourgeois family norms provide an explanation for loveless marriages in the nineteenth and early twentieth centuries, LTA relationships today have more to do with the childcare responsibilities imposed upon parents, even if they no longer see themselves as a couple. When a relationship fails, the partners can either separate and share parenting responsibilities by moving the children around, or set up an LTA household to ensure the continued presence of both parents. There is no doubt that both of these parenting options cause major upheaval in the daily lives of the children and parents concerned, but they provide two ways of addressing the need to remain a parent after life as a couple has ended.
62It is at this level, a second possible avenue of analysis, that the potentially intolerable material and financial consequences of separation become manifest. In the American cases analysed here, LTA is often the only option.  The fathers must be protected from homelessness and given a place in the home so that they can fulfil their fathering role. In France, where we studied persons in the middle-income range, the financial constraint is viewed from a different angle: here concern focuses on the financial consequences of divorce and separation, and on a “fear of losing social status” (Maurin, 2009). This is perhaps a specific feature of French LTA relationships: this fear of the future, this anxiety about loss of status and social exclusion. As pointed out by Eric Maurin, “Today, France is one of the western countries where people are most worried about their personal future and about the prospects for the country”. French society is afraid, and this anxiety underpins the individual strategies and public policies which end up creating a vicious cycle of fear. Each recession adds new vigour to this process” (Maurin, 2009, p. 89). And this fear of social demotion concerns those who believe they have the most to lose, namely the middle classes.
63There is perhaps no truly substantive difference between the low-income LTA couples in the United States and those interviewed in France, but rather a difference of intensity of need. While the American women appear practically obliged to engage in LTA relationships to keep what they view as essential (the fathering tie), those in France describe the process and the fears which ultimately lead to the same type of arrangement. While the former are guided by necessity, the latter are driven by fear. We are yet to see whether the major economic crisis now affecting both countries will provoke in the United States this same fear of social demotion and economic insecurity already felt by the French middle classes. Only a survey covering the same socioeconomic groups can answer this question.
64Are LTA relationships more frequent today than in the past? We may surmise that while the proportion of estranged couples who stay together is doubtless statistically smaller today than fifty years ago, given that divorce was much more stigmatized in the past and powerful social norms prevented married couples from separating, contemporary LTA couples may be more aware of their condition and of their conjugal unhappiness. Half a century ago, few dissatisfied couples envisaged divorce. It is even probable that the dilemma of their marital condition did not occur to them, and was certainly not a topic to be discussed with others.
65But in cultural terms, the experience of “consciously” living together apart may be more frequent today. In other words, while in statistical terms the proportion of unhappy couples who stay together is certainly lower today than in the 1950s, loveless couples who do not separate today are more aware of their condition than in the past. The demographic situation of living in an LTA relationship may be an old one, but the dilemma of such situations is perhaps a new experience. Indeed, now that divorce is relatively common, unhappy couples who do not separate are required to justify their “choice” to themselves and to others; such was not the case in the 1950s. The new question is: why am I staying with my partner when so many other couples choose to separate? This was probably not such an obvious dilemma fifty years ago, and so awareness of being in an LTA relationship was doubtless less acute. This hypothesis could also explain the recent need to name such relationships and talk about them in the press as if they were something quite new.
See article by C. Villeneuve-Gokalp, in Population, 1997, 52(5), pp. 1059-1081.
They describe the situation of a Denver couple who had divorced after six years of marriage but were forced to cohabit because they were unable to sell their house. They continued to live under the same roof, while keeping out of each other’s living space. “Couples staying together because of poor economy”, The Seattle Times, accessed on 18/04/2009.
Simon Duncan and Miranda Phillips distinguish in this way between dating LATs and partner LATs, with the former not seeing themselves as an established couple, unlike the latter (Duncan and Phillips, 2010).
An alternative expression might be “non-conjugal cohabitation”, although this could be confused with apartment sharing, a growing practice among young people needing a place to live, but which is not linked to conjugality or to the family.
Claude Martin attempted an initial exploration in an article that focused on the pre-divorce period and on the process leading up to divorce with a view to studying the effects of non-divorce on the children of estranged couples (Martin, 2007). He ended the paper with some questions on LTA families. This article cites certain excerpts of the material used for France in that article. Caitlin Cross-Barnet and Andrew Cherlin also made some initial analyses of LTA situations in the United States (Cross-Barnet et al., 2011).
Note also the legislation passed in Louisiana, Arizona and Arkansas from 1997 defending the model of covenant marriage, i.e. a marriage requiring the bride and groom to make a formal pledge to seek marital counselling and wait at least two years should one partner wish to divorce, the aim being to reduce marital instability.
i.e. couples where the partners have a high-school level of education but not a college degree, a manual worker qualification for men and clerical or sales worker qualification for women.
In 2008, 146,000 civil partnerships were registered (slightly above one for two marriages), in 95% of cases between different-sex partners (Prioux and Mazuy, 2009).
Given the difficulty of identifying LTA relationships, we used a snowballing technique, asking identified cases to point us to other cases (results available from the author). Six households were finally surveyed. We also examined LTA relationships already identified in a previous study of retirement-age couples by Vincent Caradec.
This is by no means unique to America. In some countries of Eastern Europe, the legal systems provide for a regime of post-divorce cohabitation in cases where the protagonists do not have the financial means to live separately.