Families facing the Italian lockdown: Temporal adjustments and new caring practices in shared physical custody arrangements

Objective: This paper investigates the impact of the 2020 Covid-19 related Spring Lockdown in Italy on families practicing shared physical custody (SPC) arrangements for their children. Background: Those family configurations partly challenge the dominant ‘mother as main carer model’ that characterizes Italian society. Here, we consider the lockdown as a “challenge-trial” (Martucelli 2015) to analyze the strategies that these families have developed to cope with lockdown, and to reveal the overarching structures that contributed to shape this experience of lockdown. Method: We draw on semi-structured interviews with 19 parents (9 fathers and 10 mothers), part of 12 families practicing SPC. Results: We propose a typology of custody re-organizations during lockdown and how this affected the division of parental involvement based on a) change/no change in sleepover calendars in favor of mother/father; and b) similar/different arrangements for siblings – a new practice that emerged and also has implications for the division of childcare between parents. Four types are identified where we emphasize new parenting practices and the role played by material housing configurations, relations and tensions between family members, as well as balancing work, school and childcare. Conclusion: We highlight the usefulness of applying a “challenge-trial” lens to the study of family life under lockdown, and the need to complexify research on gender equality in shared parenting and on sibling relationships in post-divorce families.


Introduction
This paper focuses on the challenges of the Covid-19 related Spring 2020 Lockdown in Italy for separated families living in the Turin area (Piedmont) and who were practicing shared physical custody arrangements (SPC) for their children when this lockdown was declared.
Italy was the first European country to be severely hit by the COVID-19 pandemic, with the first confirmed cases of community transmission found on February 21 st in the North of the country. The following day, the Italian government declared a local lockdown in several municipalities of Lombardy and Veneto, and people decreased their mobility in the whole Northern region, as companies opted for teleworking and people avoided public transport (Beria & Lunkar 2021). Between February 23 rd and 29 th , several Northern regions, including Piedmont, suspended public events and closed schools and museums, and on March 4 th all schools and universities in the country were closed. National lockdown -the first in Europe -was declared on March 9 th . All companies and working places were closed, except essential ones, and people were asked to remain at home except for work or absolute necessity (such as health reasons or food purchasing). From May 4 th on, lockdown measures were progressively relaxed -industry and retail progressively reopened, and movements across municipalities were allowed again for work or health reasons, as well as for visits to relatives. During this so-called 'Second Phase' (Beria & Lunkar 2021), the government announced that schools would remain closed until September. From May 16 th on measures were further relaxed, with most businesses, sports clubs and theatres re-opening, and freedom of movement allowed within regions. However, travel between regions remained prohibited until June 3 rd , 2020.
When the pandemic hit Italy, we were just about to start a second wave of data collection with families practicing SPC arrangements in the Turin area, in the context of a research project exploring children's socialization in post-separation families. At that stage we had an in-depth vision of the practices and routines characterizing those families' arrangements in 2018-2019, and we wanted to deepen certain issues and see how their arrangements had changed over time. But with the lockdown, additional questions emerged: what impact did lockdown have on those family arrangements? How did these families adapt to the lockdown situation? And what do these adaptations tell us about the structural factors and inequalities shaping post-divorce family life in Italy, a country still characterized by strong gender inequalities and women's prominent role in caring for children (Naldini & Solera 2018)? These are the questions we address in this paper. After presenting SPC in the Italian context and highlighting the persistence of the "mother as carer model" (Naldini & Santero 2019), we present our theoretical approach which consists in considering lockdown as a "challenge-trial" (Martuccelli 2015) that profoundly disrupted family routines, was imposed on individuals, and revealed some key elements of the social structures and inequalities that underlie and shape post-divorce family life in Italy. We then explain how the families we met adapted their custody arrangements during lockdown, and the implications those (re)organizations had for fathers' involvement in parenting.

SPC in the Italian context: the centrality of mothers as carers
In 2006, law 54/2006 introduced affidamento condiviso, or joint legal custody, as the preferred choice in parental separation cases. 1 The law states that "minor children have the right to maintain balanced and continuous relations with both parents, to receive care, education, and upbringing from both and to maintain significant relations with relatives from both parental family lines. (…) Parental authority is exercised by both parents" (Legge 54/2006 Art.1.1). However, this does not translate into an equality of residential time at each parent's dwelling. Indeed, what the Italian court set out in that law is a common right to bigenitorialità, or the children's right to have an equal and continuous relation with both parents, who are both entitled to legal custody in the form of parental responsibility. But there is no indication that they are required to share physical custody of their children, in the form of a relative equality of residential times at each home (de Blasio & Vuri 2019; Lenti & Long 2014;Murru forthcoming 2021). 2 In addition, Italian law provides the principle of assigning the dwelling where the family was living for a significant amount of time before the separation to one of the parents, following the principle of the children's right to stay in the familial dwelling. With joint legal custody (affidamento condiviso), the house can in theory be assigned to either of them, but statistics show that the family house continues to be majoritarily assigned to mothers (around 90%) as was already the case before the 2006 law (de Blasio & Vuri 2019; Lenti & Long 2014). 3 In practice, family law thus maintains a gendered hierarchy and gendered division of labor between parents that places a heavier workload and responsibilities on mothers, which is also reflected in separated families. Statistics indeed show that since the 2006 law, children have been in a very large majority in joint legal custody (89% in 2015, Istat), but shared physical custody, defined as "a care arrangement in which a child lives from 30 to 70% of the time with each parent after their separation or divorce" (Bernardi & Mortelmans 2021: 5), seems to remain quite marginal. 4 Barriers against SPC are numerous in the Italian context. As we have said, Courts tend to asign the family dwelling to the mother, meaning fathers must find and create a new living place for their children. The school system also reinforces the mother as main carer model, with dismissals at 1 or 2 pm from age 11 on, and a pedagogy heavily based on studying and exercising at home after school. Children thus have lunch at home after school, study from home in the afternoon, and carry heavy books on a daily basis -further complicating SPC logistics.
These examples highlight the influence of traditional normative models in the Italian context. Contemporary Italy is indeed still characterized by a familialistic mother-centered 1 Italy has a relatively stable crude divorce rate (from 1.6 in 2016 to 1.4 in 2019) that is slightly under the European rate (1.9 in 2016) (Eurostat). 2 As Bernardi and Mortelmans (2021: 6) note, legal custody is widespread across Western countries, but "it is not always followed up by SPC despite the strong advocacy it has received". 3 For a detailed account of the Italian legal framework for divorce, separation, and custody of children, see Murru, forthcoming 2021. 4 There is a severe lack of statistics on physical custody arrangements in Italy, but according to Steinbach et al., (2021: 151), in 2010, egalitarian SPC (50-50% arrangements) concerned 2.8% of children in non-intact families.
welfare state, where "women continue to be considered and continue to be the primary caregivers for children, while men continue to act as the main breadwinners of the family" (Cannito & Scavarda 2020: 802; see also Naldini 2015). Gender equality scores are lower in Italy than in the EU-28 average, as measured by the 2020 Gender equality index, with respectively 63.5 for Italy and 67.9 for the EU 28-average. The difference is even higher for care activities (with 63.5 for Italy and 70 for the EU-28) and work (63.3 for Italy compared with 72.2 for the EU-28). Drawing on the latest data from Istat, the Italian National Institute of Statistics, Manzo and Minello (2020) note that within working couples with children, women contribute 4.2 more hours to daily family work than men. Obstacles to fathers' involvement in childcare include, among other factors, the persistence of the mother as carer model, hegemonic masculinity (Crespi & Ruspini 2015), and barriers in the workplace (Murgia & Poggio 2011) -these factors being reflected in and supported by Italian law in cases of family dissolution. Families who practice SPC in this context are thus challenging the dominant gender model. Thomson and Turunen (2021) locate parents practicing SPC in Western countries at the forefront of the gender revolution, whether or not they are actually motivated by gender equality in earning and caring. As the existing body of literature shows, this custody arrangement not only encourages fathers' involvement in childcare, it also favorably impacts on mothers' participation in the labor market, as it reduces the temporal pressure of combining paid work, childcare, and leisure (Bernardi & Mortelmans 2021).

Lockdown as a challenge-trial for families practicing SPC
Building on French sociology, we conceptualize lockdown as a "challenge-trial" ("épreuvedéfi" in French) (Martucelli 2015), defined as "historical challenges, socially produced, and unequally distributed, that individuals must face" (Martuccelli 2006: 12). Challenge-trials profoundly disrupt the routines of life (Caradec 2007) and engage a vision of individuals as actors (Martuccelli 2006). It's also important to acknowledge that trials are not necessarily experienced negatively. As Sacriste (2019: 50) points out, "they can also be experienced in a playful, adventurous, disruptive way and, once resolved, can be liberating or emancipating". 5 Martucelli (2015) uses trials as an analytical operator not only to describe and understand how individuals cope with changes, but also to achieve a specific understanding of the structures and phenomena of society -including in the family sphere. "Trials" are indeed moments in which the places and roles occupied by family members are played out (Boutet & Le Douarin 2014), where social actors re-position themselves, and where social structures, relations and inequalities are revealed. Actually, many people's social and individual characteristics (like gender, age or material resources) make sense in relation to these challenges (Martuccelli 2006: 12).
In this paper, we operationalize challenge-trials through a focus on the daily practices through which people 'do' family (Morgan 2011). Family practices are the activities that family members do in relation with each other, and through which they (re)affirm, (re)produce and (re)define their family relationships. Childcare practices thus represent 5 Original in quote in French, translated by the authors.
an important set of family practices. Time is also an important dimension of the 'doing family' approach (Morgan 2020). Family practices indeed reproduce and construct everyday time, for instance through the setting, management and coordination of timetables within the home, the creation of temporal routines and, in the case of SPC arrangements, the definition and management of the rhythm of alternations between homes. The focus on family practices thus offers a fruitful lens to analyze how lockdown disrupted daily routines concerning the division of child custody and childcare between parents, and the new practices families developed in response to those disruptions. The lockdown represents a particular type of "challenge-trial": unlike divorce (Martuccelli 2006;Wagener 2013) or old age (Caradec 2007), it is not an institutionalized trial that punctuates the life course. Rather, it is an unprecedented constraint imposed on entire populations by governments -in response to a pandemic no one was prepared for -, and that does represent a major challenge for individuals, families, and contemporary societies. In this sense, it may be considered as particularly significant for this historical period and for life trajectories. As Mari et al (2020) note, this pandemic profoundly disrupted the emotional, working and social lives of millions of people, and had key implications for families who were confronted with telework, the impossibility of being (re)united with relatives, and constant co-habitation with their household members. In their study of 20 family networks of Neapolitan students, Fraudatario et al (2020) suggest that the strategies they developed to (re)adapt to lockdown depend on, and are influenced by, a broad series of factors such as family composition, living space arrangements, familiarity with new communication technologies, area of residence, the family's type of support system, the distribution of care loads and the organization of marital roles. In fact, there is wide consensus in the social sciences on the extent to which the experience of confinement and its social, economic and health-related consequences are not only differentiated according to key markers of social inequality but may also exacerbate them in the long-term (see for instance, Arpino et al. 2020;Bessière et al. 2020;Dubost, Pollack & Rey 2020). Existing studies of work-family balance in the Italian lockdown context suggest that in nuclear families, fathers working remotely have tended to increase their involvement with their children, particularly when their partners continue to go to their usual place of work (Cannito & Scavarda 2020, Del Boca et al. 2020Mangiavacchi et al. 2020). However, overall, these studies point out that women have spent even more time on housework and childcare, and the division of care duties has remained largely gendered and unequal, including in the more egalitarian families, and has even been exacerbated in less egalitarian ones (Cannito & Scavarda 2020;Del Boca et al. 2020). For Cannito and Scavarda (2020), this is largely due to dominant models involving the centrality of paid work in men's lives, and the expectation that women should prioritize care over paid work when care needs increase.
To date, no published studies have looked into divorced or separated families during the Covid-19 lockdown in Italy. This is an important gap, as the lives and routines of separated families have been particularly disrupted by lockdown, and their capacity to sustain SPC arrangements is a case in point. This is certainly the case for monoparental families, but it is also an important issue for those practicing SPC -which we focus in this paper. The closure of entire sectors (schools, businesses, companies), telework and distance learning erased the boundaries between school, work, and family life, disturbed the rhythms and temporalities of those social spheres, and brought them all together in the space of the dwelling. This space was often constrained as, in Italian cities, most families live in apartments (located in condominiums). This is linked to cultural modes of inhabiting as well as to the real-estate market, where these types of dwellings represent the bulk of residential offers (Filandri et al. 2020). In addition, lockdowns were avowedly designed to immobilize people in their 'homes' and limit social contacts. Alternating child custody between households is problematic in that context. Although Italy rapidly allowed minor children to maintain physical contact with both parents, fears of bringing the virus back home or of transmitting it to the other household may have encouraged parents to change their child custody arrangements, possibly opening up the way for increased involvement of fathers or, on the contrary, reinforcing the centrality of mothers as carers. That said, if alternating children between households may have represented a constraint or something difficult/impossible to manage in the context of a lockdown, it may also have been experienced as a resource (e.g., allowing children to "move", or to provide them with better housing conditions). A French study focused on children aged 8-9 shows for instance that children in SPC had fewer sleep problems, fewer socio-emotional difficulties and were less likely to have poor relations with their parents during the French Spring lockdown than children living with both their parents, or with just one parent only (Thierry et al. 2021).
Studying how parents practicing SPC faced the lockdown "challenge-trial" thus not only allows us to dig into the practices that were deployed, but also to reveal the weight, and role, played by social structures and inequalities in shaping those negotiations and practices.

Data collection and analysis
In this paper, we analyze the accounts of 12 families living in the Piedmont Region (mostly the city of Turin 6 ), among which 19 parents were interviewed (9 fathers and 10 mothers). These families were initially recruited and interviewed in 2018-2019, in the wider context of the ERC Starting Grant project "MobileKids" that focuses on children's socialization in SPC arrangements. 7 Recruitment criteria were centered around the geographical delimitation just mentioned, the age of the children (at least one child between 10 and 16), and the amount of time spent in each dwelling (calculated in terms of nights slept in each place) ranging from a 30-70% distribution to 50-50%. Here, we draw on follow-up interviews with parents (interviewed separately), conducted during the 2020 6 Among these 12 families, 1 lived in the rural area, 2 in peri-urban neighborhoods, 2 had one parent in the city center and the other in a peri-urban neighborhood, and the other 7 lived in the city center of Turin. 7 See www.mobilekids.eu. This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program under grant agreement No 676868. This paper reflects only the authors' view. The European Commission is not responsible for any use that may be made of the information it contains.
Spring lockdown. 8 Due to the severe lockdown, we described in the introduction, these interviews took place in a virtual form, using a secured institutional version of the Microsoft Teams platform. Having children in SPC, we explored how their daily lives were affected by the Covid-19 lockdown. That is to say, did they continue to use the same temporalities and rhythms for home swappings for their children to age 18 or change the days when children would be living at each parent's home? Were the logistics of moving different? How did they manage their children's remote schooling? And did eventual changes during the lockdown make them reconsider how they practiced SPC and envision organizational changes for the post-lockdown? 9 These interviews were transcribed and analyzed following an inductive thematic analysis, using Nvivo software. We first analyzed each interview separately, in order to progressively identify key analytical categories (or nodes) and enrich our coding system. We then conducted a transversal analysis of the interviews, to highlight elements of convergence and divergence across our families. Note: * Not interviewed. For the purpose of this research, we needed the interview of at least one of the parents, but both were not obliged to participate if they did not wish to (as long as they gave consent for their child to participate). So, these parents are the ones that authorized their child to participate in the MobileKids project but did not want to be interviewed themselves from the first wave of data collection on.

8
Interviews of the parents analyzed in this paper were conducted from May to July 2020 (with the majority in May and early June). 9 Something that did not emerge from our data is the process of negotiation that led to each family's choice of organization during lockdown. This is the case for two main reasons: our second wave of interviews was also aimed at posing follow-up questions regarding everyday life outside of lockdown -which entailed part of the interview time also being devoted to those discussions -and parents mostly reacted to the lockdown in a very pragmatic way, making quick decisions rather than engaging in open and complex negotiation processes.    Table 1 presents the profiles of the participating families. Prior to Lockdown, SPC was egalitarian in five families (50-50). Five families practiced quasi-egalitarian SPC (60% at the mothers'-40% at the fathers') and two families had opted for 70% at the mothers' and 30% at the fathers' arrangements. Notably, the 'cycles of care' (Masardo 2009: 201) of those families -that is to say, how often children moved between homes -were all organized around short alternations between both places in order to maintain frequent physical contact with both parents, including those practicing a 30-70% arrangement (as children spent every Wednesday at their father's). This contrasts with what is observed in other countries (where the most common cycle of care for families practicing egalitarian SPC is weekly) (see Thomson & Turunen 2021).
There is some diversity in terms of the socio-economic levels characterizing our families, which are defined based on levels of education, work contracts and income. If parents range from lower middle-class to upper middle-class, the majority of them are middle/upper-class, reflecting the global tendency of SPC to be adopted first by more affluent, highly educated middle classes (Reckseidler & Bernardi 2021). There is a diversity of family configurations in our study (such as only children, children with siblings, or children with step-siblings), but we have more boys-only than mixed or girlsonly siblings, and most children are over 10. Conflict levels between parents range from: high in two families -where parents communicate the strict minimum, don't meet in person, and are involved in tense judicial battles; medium (4 families) -at least one of the parents adapts his/her behavior in communicating with the ex to avoid tensions but they have no problem meeting face to face, eventual judicial battels are involved; to low (all the other families) -parents practice a serene and frequent communication involving the children and meeting in person is no problem.
Finally, several factors in Table 1 stress that mothers remained the main caregiver in these post-separation arrangements. First, apart from the 5 families practicing egalitarian SPC, the mother was always the parent the children spent the most time with. Second, in 7 out of 12 families, mothers earned less than fathers, and 4 of those mothers also worked part-time (only one father worked part-time). Two other factors that are not listed in the table should be mentioned too: the mother is the one to which the family dwelling was assigned to majoritarily; and 8 families where parents lived nearby dealt with school constraints by systematically organizing homework at the mother's dwelling -where children thus kept all their books and came back to everyday after school, even if it wasn't where they slept that day. This also implies that food was constantly made available for children and most laundry done there -thus reinforcing the mother in her caregiver role. Those children thus tended to consider their mother's dwelling as their "primary residence", and their father's dwelling a "secondary residence".

Facing the lockdown challenge-trial in SPC arrangements: family practices and adaptations
Our analysis of how family has been 'done' in the pandemic context reveals that the lockdown challenge-trial had contradictory effects on gender inequalities between parents: deeply shaping and reinforcing the unequal, gendered division of caring in some cases, while challenging it in others. We demonstrate this, first, through a typology of four patterns of (re)organizations of sleepover schedules, in which, for each type, we point out relevant aspects and changes in family practices relating to : (1) the articulation of paid work, childcare and school work during lockdown ; (2) material constraints linked for instance to housing space or internet connections and how families dealt with them; and (3) relational dimensions, including the management of tensions and conflicts. Second, we concentrate on the gendered patterns of parenting across the four types and examine structural factors in a cross-sectional manner. These two steps allow us to analyze the concrete practices put in place to 'do family' under lockdown beyond the temporal organization of home swaps, and thus to identify changes in parenting practices both in the amount of time spent together and in what is actually done during that time together.

(Re)-organizations of SPC under Lockdown
We built a typology of the ways our 12 families (re)organized their custody arrangements and how this affected their caring practices, that crosses two dimensions: a) whether there has been a change/no change in the calendar of sleepovers at each parents', and in case of change, whether children have spent less, more, or the same number of sleepovers at their fathers'; and b) whether families kept similar, or designed different arrangements for siblings -a new practice that emerged from the parents' accounts, and that also has implications for the division of childcare between parents. As shown in table 2, four types emerged from this analysis: (1) putting SPC on hold and reinforcing the mother as sole/main carer; (2) maintaining a (relative) status quo; (3) increasing paternal involvement through split siblings; and (4) reinforcing the mother as main carer with the establishment of different childcare arrangements among the siblings. 40-60% Split siblings, with the eldest son staying mostly at the mother's and the youngest siblings continuing to alternate together in a 50-50 arrangement.
5.1.1 Putting SPC on hold and reinforcing the mother as sole/main carer Lockdown represented a break with SPC as practiced in four families, with mothers assuming the sole physical custody of children. In Family Gialla, this re-organization of their formerly egalitarian arrangement was shaped by material constraints and opportunities. Mother and daughter were spending a week of vacation in the grandparents' house up in the Alpine mountains of Piedmont when the lockdown was declared. As they normally lived in an apartment in central Turin, remaining in the mountains seemed a much more comfortable option as they had a yard and could enjoy nature walks. The mother, while she was no longer working during the lockdown (as she worked in the cultural sector where everything was shut down), was glad to have her parents' help in caring for and occupying her 11-year-old daughter. The father was also supportive of this arrangement, acknowledging that it was best for the girl. In Family Verde, the decision to suspend their 60-40 arrangement was related to a relational factor as tensions already existed between the father and the older child before the pandemic. The lockdown thus offered an excuse to stay away for some time from a difficult relationship (the younger sister following her big sister's decision). Here, the mother, a teacher, had to cope with teleworking. But she could count on her new cohabiting partner's help to supervise the children while she taught classes online.

Mother (Family Gialla): we stayed in the mountains and obviously {the father} agreed that
For two other families, the decision to put SPC on hold was motivated by fear of spreading the virus. The father of Family Bianca lived next door to his aged mother and cared for her. So, when the mother and three children contracted the virus, the ex-partners decided it would be safer to keep the children at their mother's. Material considerations also came into play, as her house was much bigger (each child had his own room which was not the case at the father's) with a nice backyard. In Family Corallo, the father who got back from abroad at the moment of lockdown, was quite stressed about the sanitary context, the family thus deciding to suspend the home swaps for the time of the lockdown.
In all four of those families, the mothers' centrality as carers was reinforced. Beyond the factors already pointed out, work-family-school balance also came strongly into play here. These mothers were all self-employed and already arranged their schedules around their children's needs. During lockdown, two of them (Gialla and Bianca) were practically unemployed, their professional sector closing, and were thus fully available for their children, one (Verde) could count on her new partner's help when she needed to work, and the last one (Corallo) did not need to adapt her (tele)working schedule as her 17-yearold son didn't need supervision for school. She did worry though that "full-time" homemaking was tiring, even if she also enjoyed some of the aspects of everyone being together: Mother (Family Corallo): more than anything else, it's been challenging from a household viewpoint in the sense ... the burden of always having to cook three times a day was challenging. But you have to do it to survive. On the other hand, it's been ... fun as I said before. Definitely a personal change, if also a bit more tiring since you have to share the space, spending time all day long, 'cause the day never ends...
In these four families, SPC was thus suspended either to ensure children's wellbeing (by providing them a better-suited relational or material environment during lockdown) or for sanitary reasons. This arrangement reinforced the mother as main carer, in a context where the four mothers could easily adapt their working schedule to their children's caring needs.

Maintaining a (relative) status quo
Three families maintained the same custody arrangement as pre-covid, motivated by past or present parental conflicts. The relational dimension was thus key for them. In Family Arancione, pre-covid intra-parental relations were marked by serious conflicts. Communication was already reduced to strict necessities and never face to face, which made renegotiating the home swapping calendar impossible. Pre-and during lockdown, sleepover shifts were thus organized following this scenario: sleeping at Mom's on Mondays and Tuesdays, alternating Wednesdays, at Dad's on Thursdays and Fridays, and alternating weekends (if for instance it is the Mom's weekend, then it is the Dad's Wednesday). Interestingly, new caring practices did emerge though.
Step-parents took on a new role, as both parents worked in the Health sector and were required to continue working during lockdown. Since both of these parents had repartnered, the three children aged 6 to 16 stayed at home with their stepparents, which was new as they'd never spent days together alone pre-covid. This family also made use of the baby-sitter that worked for them pre-covid to help out certain days (at either dwelling).
Relying more heavily on his new partner's help with childcare did create tensions in the father's household.

Father (Family Arancione): {my partner}, she wasn't working and home a lot; she suffered more from the lockdown. She found herself almost in a cage. Sometimes she said 'but I who didn't want, I didn't want to baby sit your children, I found myself baby-sitting your children". Eh that is, nobody asked you to but the situation was that way…
The father did though also engage in new caring practices, by investing a lot of time and energy in helping his children with distance learning, which meant changing the habits of his son, who was not used to asking him for help, and relied a lot on his older sister to supervise his homework. In Family Rosa, while intra-parental conflict was still high, parents did make some small adaptations to deal with the lockdown challenge-trial but without having to change the actual agreed upon egalitarian shift of residence. This was due to several factors, all connected to the work-family-school balance. The physical closure of schools played a role, as the transition between residences could no longer take place there. In fact, the children used to spend Mondays and Tuesdays at Mom's; they'd usually go to their Dad's right after school on Wednesdays and go back to their mother's on Fridays after school if it was the weekend they spent there (alternating weekends). Since the father was required to work in hospitals, he could not be available noon to maintain the usual arrangement in that context, to his great regret. In this ex-couple, he was the one who'd really fought to establish egalitarian SPC pre-Covid and wanted it kept up. Working remotely from home, the self-employed mother was stressed about her workload but could be more flexible in organizing her schedule. She also emphasized that balancing work and family was facilitated by her teenage boy who soon became autonomous in managing his own school schedule (thus only having to supervise the younger one). Thus during lockdown, because of school closures and working requirements, the father would pick his boys up on Wednesday evenings at their mother's before dinner (so the children would now spend the whole day at the mother's), would arrange to have very few patients on Thursdays so he just had to leave for two or three hours (leaving the children alone briefly), and on Fridays of the mother's weekend, drop the kids off in the morning before going to work (here too, the children thus spent an extra day every other week at their mother's instead of arriving in the afternoon after school).

Father (Family
In this quote, the father explains the delicate arrangements he had to make at work and with his ex-partner to spend time with his children: Father (Family Rosa): {the mother} continued to work from home, not having to go to the office, while I continued to work in the hospital even a bit more than before, because then they'd ask us to do shifts and other things. We worked things out, and instead of taking the boys as from the document of the Court, meaning when school gets out (...), {on Wednesdays} I picked them up at her place in the evening before dinner. And then on Fridays when they were with her, normally she would pick them up {in the afternoon} at school, but she agreed to have them on Friday morning instead. So, in short {on the Fridays they stayed with me,} I would either take a recovery or a day off. There was a little more flexibility from this point of view, which was ok for me, and was ok for her.... So, while Family Rosa kept the same calendar of home swappings, the daytime the children spent with their mother increased, and hence so did her care work.
Finally, in Family Nero, the ex-partners, who were now on good terms, had established a 60-40 arrangement after engaging in an intense conflict and judicial battles during the first years of separation. They thus considered this arrangement was set in stone and not open to re-negotiations. Children continued to go at Dad's on Mondays and Tuesdays, at Mom's on Wednesdays, Thursdays, and Fridays, and alternating weekends. Parents were teleworking but this did not feel like a burden as the children were old enough to take care of themselves.
Present or past parental conflicts thus limited the room for manoeuvre in these three families, particularly with respect to the calendars of home swaps. However, some changes did emerge, with the mother taking up more daytime care in one case, and the step-parents and the father engaging in new caring practices in the other cases. Workfamily-school balance played a key role in these three cases.

Increasing paternal involvement through split siblings
In our study, four families changed the calendars of sleepovers and divided children between parents, leading to an increased involvement of fathers. Family Azzuro has two teenage boys, aged 16 and 18 who used to reside 70% of the time at their mother'salthough the older one was starting to enjoy more flexibility in alternating homes precovid. The day everything closed down, it happened that the brothers were residing apart (the older had chosen to go to his father's instead of his mother's). Motivated by a desire to respect the injunction to circulate a minimum during lockdown, parents and children together decided that things would remain pretty much that way for the time of the lockdown. So, each boy established his 'main residence' at the mother's and father's respectively, and only paid small visits or spent short stays in the other one. SPC was thus suspended, but with each parent taking on main responsibility for one child. This new arrangement does challenge the mother as main carer model, as the father actually increased his caring role for one of the children. Constantly having a child at home was possible for these parents who were both teleworking, and wasn't seen as a burden because the boys were very autonomous in their schoolwork.
For its part, Family Marrone shifted from a 30-70 arrangement to a 50-50, one week/one week format -which, as it has been stressed, is an unusual arrangement in the Italian context. This combined with the decision to split both children, aged 15 and 17, between mother and father. So SPC was maintained and more equal, but with a separation of siblings, the mother coming to the father's every Sunday evening to "exchange children". This new arrangement was motivated mainly by material conditions, including housing space (the father's apartment being very small) and internet connections (weak at the mother's place). The mother needed to teach remotely, the father worked in IT for a private company, the children had to follow their own lessons as well, so it was crucial for all of them to obtain a stable internet connection and enjoy some private space. Interestingly, a new family practice also emerged. During the mother's Sunday visit to the father, the children enjoyed spending a bit of time together, and the father ended up offering the mother to stay for dinner. A new routine of Sunday "family meal" thus emerged, something that never happened before the pandemic.
Family Viola changed due to relational aspects and material constraints. In this passage, the mother summarizes her reasons for choosing the new arrangement.
Mother (Family Viola): for the first two weeks, then also due to a problem of availability of... computers to provide connections for school as well as our work from home... we just had to separate, as they themselves requested, 'cause they couldn't stand each other. In fact, when they were all in {the father's} place...the apartment being so small ...a few clashes broke out so we separated them so that on Mondays and Tuesdays one was with me and the other went there, and on Wednesdays, we exchanged them.
As pointed out by the mother, in this family of three boys aged 12, 16 and 18, the two eldest who were fighting a lot asked their parents if they could alternate separately. That issue was aggravated by the father's living in a small apartment with only one bedroom for the children, whereas, in both homes, they were all struggling with internet connections whenever everyone needed to be online at once. So, on their children's insistence, while maintaining the same shifting calendar, the parents decided to disjoint their older sons' alternating rhythms, with each parent having at least one child at home at all times for the time of the lockdown. As for the youngest, they also used the same shifting format but starting a day later, so he'd have a few days with each of his older brothers.
Moreover, the parents who were both required to tele-work, stressed that they enjoyed always having someone at home to fight feelings of loneliness in a context of reduced social contacts. In order to adapt to continuously having at least one child at home, the father who worked in the banking sector, maintained a fairly similar tele-working schedule but mentioned having a lighter workload than before. The self-employed mother had more flexible working schedules and could adjust to the children's needs.
In Family Blu, children were also instrumental in the division of siblings between households, but in the context of a very fluid organization of their 60-40 sleepover schedule. Before the pandemic, there were some fixed shift days but the two boys, aged 15 and 17, were free to ask to change them if they needed/wanted. This pre-covid flexibility was then totally embraced during the lockdown. What motivated the comings and goings in this case, matching everyone's need for a stable internet connection (the father was teleworking full-time, not the mother as her workplace was closed), was the need to move around and get a breath of fresh air. This flexibility was experienced positively by the family, as seen in this quote.
Father (Family Blu): so actually, with the lockdown the logistical organization changed a bit. Because it happened a lot, let's say they came more often, they were here, sometimes one of them was here, sometimes not, also because there were problems with the internet bandwidth because one of them had a class with zoom, the other one a class with zoom, I work and so on.... (…) but all in all, it didn't go so bad, in the sense that...no I would say that it went reasonably well, in short, they were moving a bit, they were not completely blocked, closed in the place, always in the same home, always motionless and so it was quite tolerable, ...
Here too, the brothers ended up living apart on some days (not always) to avoid conflict, to secure stable internet connections, and also based on the help they needed with homework (the father being able to help with science and math).
In this type, each parent thus constantly hosted one child (with various calendars of home swaps), in most cases under the impulse of the children themselves. These new care arrangements were mainly motivated by material and relational aspects. They allowed the parents to equally divide the care burden while enjoying a continuous presence at home. 5.1.4 Reinforcing the mother as main carer with the establishment of different childcare arrangements among the siblings Finally, in Family Rosso, the calendar of home shifts pre-covid followed a very fluid and unorganized pattern for the oldest son and a more fixed calendar for the smaller two boys, with children spending approx. 60% of their time at the mother's. During lockdown, the youngest continued to alternate as planned (two weekdays at the father's and alternated weekends). But the eldest was already alternating following his desires and needs and considered his mother's home his main place of residence, where he had invested most materially. Indeed, he didn't have any sort of space of his own at his father's, whereas he had his own room at his mother's. The older son thus saw his Dad's home more as a convenient location, being near where he'd go to relax with his friends. So, when everything shut down and occasions for seeing his friends vanished, he decided to remain mostly at his mother's where he felt more at home -and his parents didn't oppose it. The mother, being a physical therapist, was unable to work during lockdown. The father worked in the social sector and had to go to work physically some days (but otherwise mostly off work).
Like in the previous type, the child's own desires were thus instrumental in shaping the lockdown care arrangement. In this type, parents paid attention to the different needs of younger and older siblings. That is to say, maintaining a sense of routine for the younger ones and allowing autonomy to the older one in following his desires -which resulted, in this case, in an increased investment of the mother -, with a strong influence of material and relational dimensions.

Changing gendered parenting practices
Our typology shows that some families reacted to the lockdown challenge-trial by reorganizing home swap schedules in ways that reinforced the mother as sole or main carer. This happened in type 1 where SPC arrangements were put on hold; in type 4, where one child was always at the mother's while the younger two alternated, and also in one situation of apparent status quo in type 2, where family Rosa increased the presence of children at their mother's during daytime. In type 3 however, fathers increased their share of parenting by continuously accommodating at least one child and, in addition, moving from a 70-30 arrangement to an egalitarian schedule of shifts in the case of family Marrone.
Splitting children between homes also changed family practices and relations during time spent together. Some parents would enjoy cooking with one child, and then tending the balcony plants with another. Conversation topics as well as leisure activities varied from child to child. This created a different dynamic for the parents as well, being used to alternating between days with children at home and days alone. Mothers who assumed more care of their children during the lockdown also engaged in practices that nourished their mother-child relationships. It is important to note that these mothers didn't necessarily see temporary sole custody as an extra -and so negative -burden. They also described this as an opportunity to spend time with their children on leisure activities (such as watching movies, cooking, or sorting out and tidying home "stuff") that they never had the time to do during busy pre-covid days with everyone running around.

Mother (Family
Mother (Family Rosa): this fact of staying at home; we started to clean. We have a nice big cellar but it was crammed full, so {my older son} started emptying it, he took the books out, arranging them. Yesterday we did a lot, now we're thinking a little bit about the garden. We clean house together, like Saturday, Sunday, (…) when we are all together, the three of us start cleaning from top to bottom. We don't have much money so we always try doing things ourselves, but... but dedicating time to things is positive too, a bit more relaxed, without running, the frenzy, rushing the way we used to from morning to night, it was very tiring.
From the data of the first wave, we know that several parents used to organize their workload, leisure activities, as well as the days where they'd spend time with their noncoresident new partner, on days the children didn't sleep at their place (a practice also observed in other countries -see for instance, Hachet 2014). It's something some participants longed to find again post-lockdown.
Mother (Family Marrone): I think we'll go back to the usual organization and, frankly, I hope so too... because that allowed me, for example, to have free time, where I was alone and... it was better balanced.
Aside from sleepovers, it is important to point out that all children, except those who continuously resided at their mother's, ended up spending full days with their father, thus increasing his involvement in childcare and in doing tasks usually performed by women. For instance, fathers who were used to only having their children at home on evenings, nights and mornings before school, suddenly had to take care of their children all day, supervise remote learning and homework, and see to cooking and cleaning more than before. In some families, care and domestic duties also weighed more heavily on daughters than sons -girls being more helpful than boys. Care work increased even more in cases where siblings were split from each other, as the fathers constantly had to care for at least one child at home. So, the experience of always having children at home also implied a whole new practice of parenting, especially for fathers. This increased involvement of men was appreciated by mothers who could get some relief from their childcare duties, especially for those used to having their children back at their place every day after school to do homework: Mother (Family Viola): Not coming home after school anymore meant that the boys were where they were, and so {the father} had to make them lunch... for the days they were with him. So, I was relieved of some of the lunches with the boys and managing the boys in the afternoon.
Several structural dimensions shaped and constrained these gendered parenting practices. Several scholars have stressed the importance of materiality in housing, especially in the case of SPC arrangements (Merla & Nobels 2021, 2019Merla et al. 2021;Palludan & Winther 2016). As mentioned previously, Italian courts tend to allocate the family dwelling to mothers after separation. This allows mothers and children to live in a place that was previously suited to accommodating them, and de facto facilitates children feeling more at home there (although the process of feeling at home is complex and relies on a whole series of factors -see Merla et al. 2021). The mothers' dwelling thus seems more attractive and suited in a lockdown context, reinforcing mothers as main carers. That said, housing conditions also prompted some families to adapt shared custody in a sense that increased fathers' involvement in childcare.
With regards to the Italian context, and in particular the urban landscape of Turin, it is important to recall that, apart from one, all of the families we encountered lived in apartments. Most families with more than one child shared rooms in both homes or, in some cases, had separate bedrooms at one parent's, and shared at the other's. Only one family with siblings had separate rooms at both homes. In the context of a lockdown, as described, it was particularly challenging for families composed of 3 children who shared a single bedroom at their father's. This usually meant that it was impossible for all of them to be alone in a room to attend online classes (or for the parent to work alone in a closed space). When it was possible, some of those families opted to split siblings between mother's and father's, which, in turn reinforced paternal involvement in childcare. The need to secure a stable internet connection also prompted teleworking parents with children attending online classes to either increase children's presence in the dwelling with the best internet connection (thus increasing the burden for that parent), or else split siblings up between homes.
Gender inequalities in the Italian labor market also played an important role. Many mothers were used to adapting their work schedule to their children's needs (either by working part time or through being self-employed and having flexibility in organizing schedules) and continued to do so in lockdown. In contrast, very few fathers worked parttime or were self-employed, and most fathers were expected to continue working remotely, or at their workplace for a few of them. However, school closure challenged the centrality of mothers in separated families, supported by the Italian school system. During lockdown, the school day no longer ended around lunch break and the burden of carrying books back and forth to school was no longer there. This had consequences for all our participants, apart from the 3 families that practiced the exact same calendar of home swapping. They all ended up (to a varied extent) extending the days the children lived in each home consecutively, thus too increasing the time that fathers spent with their children and their roles in home schooling. It is also important to note that the burden of work-care-school balance heavily depended on children's age. In our study, most children were teenagers at the time of lockdown and were rather autonomous with regard to their schedules of online lessons and homework. This thus allowed parents to telework during their children's "school day".
If the lockdown represented a time for experimenting with new organizations and schedules, most families were in the end looking forward to resuming to the pre-covid custodial arrangement -or at least saw no other possibility once school would resume in a physical form. In only two families (Verde and Marrone) did the parents reflect on the possibilities of modifying the home swapping calendar after the lockdown, based on their positive experience in lockdown.

Conclusion: Directions for future research
Analyzing the experience of lockdown through the lens of the "challenge-trial" proved particularly fruitful for the study of family life in times of covid, in at least three ways. First, it turned our attention to the structural factors that shape and constrain family practices in challenging times, such as, gender inequalities in family law, working conditions or the school system. Second, by stressing the importance of re-negotiations and re-positionings, it allowed us to show that the lockdown trial led parents to reposition themselves and rethink their respective places. Third, our results confirm that trials are not necessarily experienced negatively. Our data is marked with moments of anxiety related to the circulation of the virus, especially when caring for frail grandparents, or with regard to balancing teleworking and supervising children's schoolwork. But it also was experienced in a more positive way. Here, the context of shared custody was particularly useful in limiting the negative impact of lockdown on the wellbeing of family members, as it allowed parents to divide the burdens (and joys) of the lockdown across two households, to deal with material constraints, and to manage or even reduce parent-child and sibling tensions. This latter point underlines the need to move more firmly beyond simplistic, normative (and often alarmist) visions of post-separation family life, which are still very much present in literature on this topic (Zartler 2021). Here we'd like to point out two aspects that are particularly challenged when taking a qualitative approach to family practices. First, there is a tendency in research on SPC to consider that shared parenting automatically challenges gender inequalities in parenting (Davies 2015). The general assumption is that divorced fathers who house their children overnight will automatically take on a larger share of care duties, including those usually performed by mothers (Vrolijk & Keizer 2021). 10 We partly endorsed that vision, as we used sleepovers as a measure of paternal involvement. Yet, testimonies of parents describing the new care practices fathers have had to put in place show that gender inequalities persisted in precovid SPC arrangements. We also found that out because we looked into what happened daytimes rather than solely focusing on sleepovers. Some parents also emphasized the caring role played by their new partner, thus showing that care tasks may be taken over by other women in the father's household. Finally, older children, and daughters in particular, also relieved fathers of certain tasks, for example by tidying the house or helping their younger siblings with schoolwork. These observations point towards the need to take daytime and sleepovers into consideration, to trace what family members do during these times together, and to shift the prominent focus in SPC research from shared parenting to "shared care families", as suggested by Hayley Davies (2015: 11) -a conceptualization better able to take the variety of persons involved in what is considered shared parenting into account.
Second, the new practice of splitting siblings that emerged during the lockdownsometimes at the request of the children themselves -challenges prevailing visions of sibling relationships in post-divorce families, a topic also largely invisible in the literature on family dissolution and post-divorce family life (Unterreiner 2018; for an exception, see Winther & Larsen 2021). There is a general assumption amongst family experts that splitting siblings is automatically detrimental and should be avoided at all cost (Marquet & Merla 2019). Some researchers have put forth in the 1990s and early 2000s that this arrangement increases the difficulty to cope with parental separation, supports loyalty conflicts and parental alienation, and impairs the possibility to create strong bonds with siblings on the long term (for a literature review, see Hawthorne 2000; see also Drapeau et al. 2000). Yet, other research challenges those assumptions. For instance, Bruce Hawthorne (2000) found that sibling relations not only are not necessarily weakened in split siblings but are also portrayed by some children as stronger than pre-separation, particularly when siblings have regular occasions to see each other across households. If our intention here has not been to promote this type of arrangement -which was, in our case, put in place for a short period of time and in families where regular contacts between households were maintained -our study calls for a need to complexify sibling relations (Gulløv & Winther 2021). This implies recognizing that they can be conflictual, and that siblings may wish to live apart from each other at particular times, without necessarily jeopardizing their family relationships.