Grandmothers Offering Regular Auxiliary Care for Their Grandchildren: An Expression of Generativity in Later Life?

The study explores the meaning that grandmothers who offer regular childcare attach to that experience and the extent to which the notion of generativity might explain the meaning of the experience of caring for a grandchild. Twenty-four Spanish grandmothers aged 60 and over who provided at least 12 hours a week of childcare were interviewed. Interviews were transcribed verbatim and qualitatively analyzed. The results suggest that grandmothers regard their childcare input as being part of their parental duty of helping in times of need. A range of consequences of caregiving were mentioned, generally positive and satisfactory. Our participants apparently began their caregiving task as an expression of parental generativity, although rewards associated with childcare are linked to grandparental generativity.


INTRODUCTION
The combination of increasing life expectancy and declining fertility has changed the structure and relationships within the family in developed countries. For instance, it is now expected that three or even four generations will share a significant part of their lives together, increasing the number of grandparents and their opportunity to play a significant role in their grandchildren's lives (Bengtson, 2001;Szinovacz, 1998). However, the definition Caregiving Grandmothers and Generativity 293 and expectations associated with being a grandparent remain fuzzier than those of other family roles such as parent or child. As a result, the role meanings and typical behavioral patterns of grandparents have attracted considerable research interest. These studies acknowledge the complexity of such a role, proposing typologies to define qualitatively different ways of grandparenting (see, for instance, Cherlin & Furstenberg, 1985;Neugarten & Weinstein, 1964;Triadó, Villar, Solé, Osuna, & Pinazo, 2005;Van Rast, Verschueren, & Marcoen, 1995) and exploring how different types of grandparenthood are influenced by factors such as gender, proximity, or family lineage (Chan & Elder, 2000;Thomas, 1989).
In all these typologies, caregiving is recognized as one of the main functions performed by many grandparents. Although in Western societies grandparents follow a rule of noninterference, by which they remain apart from tasks that are deemed to be part of the parental role, grandparents do undertake functions that are more traditionally associated with parents if the family needs their support. This has led some authors to label grandparents as "family watchdogs," whose presence serves to stabilize the family in times of crisis (Troll, 1983). However, the nature, intensity, and seriousness of the need that leads grandparents to play a more active role in the family, particularly as regards childcare, may vary considerably. Grandparents' involvement as caregivers for their grandchildren can be seen as a continuum: At one end there would be their role as an occasional and auxiliary resource to which parents resort in case of need, whereas at the other there would be full-time care of grandchildren in times of family crisis or when parents are not available at all.
As regards the latter, studies show that grandparents play an increasing role at times of parental divorce, remarriage, and in single-parent families (Kennedy & Kennedy, 1993;Lussier, Deater-Deckard, Dunn, & Davies, 2002). Grandparents may even take on the role of a parent, either legally or informally, if parents are unwilling or unable to provide care due to various circumstances such as parental death, severe mental disorders, substance abuse, or imprisonment (Goodman & Silverstein, 2002;Minkler, 1999). In these cases, grandparents and grandchildren share the same household, either with or without parents, and grandparents often assume full-time responsibility for their grandchildren. In the United States it is estimated that at least 2 million grandparents, most of them grandmothers, assume a custodial or full-time surrogate parental role (U.S. Bureau of Census, 2008). Raising children is a demanding task, particularly in the adverse psychosocial circumstances that generally force grandparents to raise their grandchildren. Thus, there is a considerable literature describing how being a surrogate parent has a negative impact on grandparents' own physical and mental health, as well as its relationship with financial problems and a restriction of social networks (Bachman & Chase-Lansdale, 2005;Hayslip & Kaminsky, 2005;Jendrek, 1993;Lumpkin, 2008;Minkler, 1999;Minkler & Fuller-Thomson, 1999). However, the experience also typically brings a number of positive outcomes, including an enhanced sense of purpose in life and positive feelings derived from the companionship of a grandchild (Pruchno, 1999).
In contrast to the extensive research on grandparents fostering their grandchildren, their role as auxiliary caregivers in low-risk families, regularly offering care support but only for limited times and tasks, has been far less studied. This research gap needs to be filled for at least two reasons. The first is the pervasiveness of the auxiliary caregiving role among grandparents in developed societies. Due to the greater female participation in the workforce, there is a regular need for childcare in many families, with grandparents being an obvious resort, if they are available. Indeed, grandparents have proved to be a key resource in enabling parents to reconcile work and family responsibilities (Gray, 2005;Koslowski, 2009). In the United States, for example, 58% of grandmothers and 38% of grandfathers living nearby provide some type of childcare assistance to their adult children (Guzmán, 2004). In Europe the trend is similar. According to the Survey of Health, Ageing and Retirement in Europe (SHARE), 58% of grandmothers and 49% of grandfathers in 10 different European countries provided some care for their grandchild during a 12-month period (Hank & Buber, 2009).
It should also be noted that although grandfathers are involved in childcare, the involvement of grandmothers is more frequent, and they provide more hours of care than do grandfathers. This gender gap is particularly pronounced if only regular carers are taken into account (Hank & Buber, 2009). In addition, childcare offered by grandmothers tends to be more centered around parents' work hours and is less leisure-oriented (Guzmán, 2004). Such gender differences in childcare mirror traditional gender roles, women in general being more care and family oriented than men; this is consistent with other studies that found that, in general, grandmothers (and particularly maternal grandmothers) have more contact with their grandchildren (Uhlenberg & Hammill, 1998) and are more satisfied with their role (Thomas, 1986) than are grandfathers.
A second reason why research is needed on auxiliary caregiving in grandparenthood is related to the consequences for grandparents of their regular involvement in childcare, an issue that has received little attention so far. Although the degree of negative outcomes is likely to be lesser here than in the case of custodial grandparents, it needs to be acknowledged that the assumption of parental responsibilities by aging grandparents is clearly a developmentally "off-time" task, one that occurs at a time when grandparents may be experiencing a decline in their own health and stamina. Consequently, older adults may find the responsibility of offering regular childcare stressful and a hindrance to the rebuilding of life after retirement, when they are supposedly free of work and family duties. A recent study, for example, found that grandmothers who provide childcare for nine or more hours a week had higher rates of heart disease than did those who provided no care (Lee, Coldiz, Berkman, & Kawachi, 2003). Also, in the Spanish context, some authors (Guijarro, 2001) argue that health problems and psychological stress are more common in grandmothers whose support as caregivers is overdemanded by their adult children. This situation has been labelled the "slave grandmother syndrome." On the other hand, caring for grandchildren without being the main person responsible for their upbringing and education could also be a source of satisfaction for grandparents, who may obtain benefits in terms of enjoyment, well-being, a sense of purpose, and personal growth. In this regard, generative interests in older age could account for the optimistic view of auxiliary caregiving in grandparenthood. Generativity refers to a "concern for establishing and guiding the next generation" (Erikson, 1963, p. 267) and was originally proposed to be the main challenge and developmental task for the middle-aged. In that stage of life, parenting is the most obvious way to channel generative interest, although it can equally be expressed through teaching, volunteering, and any activity that seeks to promote and nurture future generations. Through generative behaviors, people contribute to the maintenance of their family and community and, at the same time, develop personal qualities and goals that enhance their competence and well-being (Villar, in press).
In his later work, Erik Erikson (Erikson, Erikson, & Kivnick, 1986) expanded the notion of generative interest beyond middle age, introducing the concept of "grand-generativity," which refers to the concern of older people to nurture, guide, and care for younger generations, with the role of grandparent providing perhaps the best opportunity of being generative in older age. Thus, generativity could be related both to satisfaction derived from the caregiver role and to the motivations that lead grandparents, and particularly grandmothers, to play that role. As regards the former, Thiele and Whelan (2008) found that generative concern was a strong predictor of grandparental satisfaction in a sample of noncustodial grandparents. In the same vein, Budini, Gattai, and Musatti (1999) report how a sample of grandmothers who regularly cared for their grandchildren derived from that role feelings of being important and useful to the family (for their children and grandchildren), as well as a sense of personal enrichment and the enhancement of new or rediscovered capabilities.
However, findings about the role of generativity as a motivation to offer care for grandchildren are somewhat conflicting. Whereas Hoppmann and Klumb (2010) report that grandparental generativity goals related to wanting to leave an imprint on others' lives and wanting to feel needed were positively associated with grandparents' desire to provide childcare, Materne and Luszcz (2010) failed to find evidence supporting a relationship between generative concern and commitment when studying the spontaneous offer of a sample of grandparents to provide care for their grandchildren. Budini et al.

296
F. Villar et al. (1999) found that grandmothers in their study saw their involvement in childcare as being taken for granted and a natural decision that did not need an explicit discussion or decision. Clearly, more studies are required to clarify the role of generativity in the reasons that drive grandparents to offer childcare.
In light of the above, the objectives of the present research were twofold. First, we sought to explore the meaning that grandmothers who offer regular auxiliary care to their grandchildren attach to that experience, taking into account the reasons that led them to offer childcare, the terms under which such care is provided, and the positive and negative outcomes of the experience. Second, we aimed to ascertain the extent to which the notion of generativity can explain the meaning of the experience of caring for a grandchild and, particularly, the reasons for offering this care and the associated positive outcomes.

Participants
Participants were 24 grandmothers aged 60 years and over (mean 69.6, SD = 6.95) living in Barcelona and who were not currently working outside the home. They were chosen through purposive sampling, and in order to be included they had to care for at least one grandchild younger than 13 years, to whom they devoted at least 10 hours a week spread across at least five of the seven days. Thus, we looked for grandmothers who acted as caregivers in a rather intensive and regular way, but without being the main caregiver (parents were the main caregivers in all cases) or sharing the household with their grandchildren. We also tried to have the same number of married and nonmarried grandmothers Most participants took care of more than one grandchild (see Table 1). Of these grandchildren, 22 were female and 23 male (total grandchildren = 45), with 29 belonging to the maternal family lineage and 16 coming from the paternal side. Their mean age was 4.51 (SD = 2.84). The participants estimated that they cared for their grandchildren an average of 20.3 hours a week (SD = 11.08). Six of the participants reported taking care of their grandchildren during all the mother's working hours in a full-time job.
None of the grandmothers was employed at the time of the study, and 14 of them had never worked or had had only short working experiences prior to marriage. The level of education among participants was low, even by the standards of older generations in Spain (see IMSERSO, 2009). Eleven grandmothers reported having completed only primary education, while another four had not finished their compulsory schooling period. The remaining grandmothers had secondary education. At the time of the study, 12 participants were married and 12 were widowed. Instruments A semistructured interview was developed to explore the participants' experience as caregivers, including questions referring to care tasks that were assumed by the participant, motivations for looking after their grandchildren, and the consequences of caregiving for the participant's life. The complete interview comprised eight questions, but only responses to three questions were analyzed in the present study. These questions were as follows: • How did you become a caregiving grandmother? • How do you try to care for your grandchildren? Is it similar to the way their parents care for them? • How do you regard your experience as a caregiver? What are the rewards and costs of being a caregiving grandmother?
Before being interviewed, participants read and signed a consent form. All the interviews were audiotaped and transcribed, and all proper names and locations were removed from the transcription to ensure confidentiality. Sections of transcriptions corresponding to answers to the target questions were then isolated and subjected to content analysis. The aim here was to identify key themes in the interview data (Strauss & Corbin, 1990). The data were then analyzed using an interpretive approach to identify emergent themes. These themes categorized the data according to the three questions studied: ideas about the point at which they decided to become a caregiver grandmother, how they see their involvement in this care (characteristics and limits of this role), and their views regarding the positive 298 F. Villar et al. Avoiding the risks of older age Finding company and a purpose in life "Slave" grandmothers and negative consequences of being a caregiver grandmother (see Table 2). To ensure consistent data coding, two researchers coded the same data set and cross-checked their coding throughout the data analysis to ensure reliability (Minichiello, Aroni, Timewell, & Alexander, 1995). There were no major differences in coding, and any minor differences were resolved by discussion between the researchers. Consistent with qualitative research, we were more interested in the quality of the coded text and how grandmothers express their ideas of caregiving, rather than in assigning numbers to the transcripts.

Procedure
The first six grandmothers who participated in the study were contacted through two civic centers in the city of Barcelona. These centers, which are funded by the city council, aim to provide leisure, social, and educational activities for older people. Participation in these activities is generally free. With the consent of the center manager, the researchers advertised the study in the civic centers and gave a contact phone number. These participants were asked to give the contact information of other grandmothers who did not attend any civic center and, at the same time, met the criteria to be included in the study, as described in the Sample section. Nine grandmothers were recruited following that snowball strategy. The sample was completed by recruiting five grandmothers from parks and other public spaces, where many grandmothers take their grandchildren to play.
These five grandmothers were chosen so that the number of married and widow grandmothers were the same in the final sample. The strategy in gathering participants from multiple sources was aimed at obtaining meanings and opinions about the research topics that would be as varied as possible.
After making sure that grandmothers met the participation criteria, explaining the objectives of the research, and obtaining their explicit consent to participate voluntarily in the study, grandmothers were given a scheduled interview time. Interviews were carried out in a private room of the university by two different interviewers. They lasted from 20 to 40 minutes, were audiotaped, and transcribed verbatim.

Reasons for Caregiving: The Parental Responsibility of Helping in Case of Need
When asked about why they had ended up offering auxiliary and regular care for their grandchildren, most of the participants (18 out of 24) mentioned the parents' need for someone to look after their children while they were at work. This was particularly the case with respect to mothers who worked and spent many hours outside the home, a situation that led grandparents to fill that parental gap and assume regular care of their grandchildren. However, none of the participants mentioned the working role of fathers as a circumstance that prompted the need for auxiliary care, even when they were offering care to paternal grandchildren. At most, some of them referred to "working parents" as the circumstance that made their childcare necessary: I started looking after them when her maternity leave finished. As they need to work and she only has me, they have to bring them to me. (Aurelia, 60) It's been a necessity, because his parents work. So they needed me to pick him up from school and look after him for a couple of hours. (Eli,64) In most cases the care offered was related to everyday indoor activities, such as supervising grandchildren (while they watched TV, played, or did their school homework) or getting their food ready and feeding them. For the youngest grandchildren, care activities also included bathing or changing nappies. Among outdoor activities the most frequently mentioned was accompanying grandchildren to and from school or nursery. It should also be noted that these activities were carried out on a regular basis, often performing the same tasks every working day, and at about the same time and for the same duration.
In some cases, however, grandparents' participation was also required (and offered) in care tasks defined as "extraordinary." These were things such as unforeseen circumstances and emergencies, like helping in the event of a grandchild's illness, unexpected work-related trips, or working overtime. In these cases, childcare was offered more intensively, although grandmothers seemed to regard these sporadic demands as an extension of the parents' obligation (particularly that of mothers) not to miss working days or refuse extra work, the argument being that this was necessary in order to keep their job and maintain the family's income: Sometimes my grandchildren have spent the night with me, when their mother got home late from work and was too tired. . . . You have to make some sacrifices if you want to keep your job these days. (Conchita,71) Only three grandmothers mentioned the parents' leisure activities as circumstances that required them to care for their grandchildren. However, even these situations were defined as "necessities" that they responded to so that their children's family would function more smoothly: Sometimes they say to me: "Mum, we have a birthday," or "We'd like to go to the cinema," and they leave the kids with me. I understand. . . . A couple needs some time by themselves. (Juana,79) The grandparent role that consists of filling the gap left by working parents (and particularly by working mothers) is reinforced when the participants consider the alternatives were they not available as caregivers. In most cases the alternative would be some kind of formal care (nursery, babysitters, etc.), which for many families would imply a financial burden that could not be borne: If I weren't here they would have to find another solution . . . and paying someone to look after their child . . . well, they would find it really hard to make ends meet every month. (Maria,65) In three cases, participants even mentioned a more radical alternative: If they were not available their daughter (or son) would not have had children: Maybe Paula would not have been born. We wouldn't have her here. They would have resigned themselves to having just the one child, the older one. (Fina,71) As well as being expensive, some grandmothers (11 out of 24) mentioned that formal care alternatives to their caregiving would not be so good for the grandchildren. In their view, what they could guarantee was the full attention and loving care that was sometimes lacking in nurseries or with babysitters: Before she was born, the other grandmother . . . a nice woman . . . and I decided that we didn't want to . . . well, neither a nursery nor her being with someone you don't know. . . . After all, they're a business, and what babies need is as much family care as they can get. (Remei,63) Once these circumstances were present and the grandmothers perceived themselves as necessary for their children, being an auxiliary caregiver seemed to be seen as something natural and was taken for granted by most of them. Thus, they found it difficult to isolate a specific point at which they decided to take on this role, and their involvement in their grandchildren's care did not appear to be the result of any kind of discussion or negotiation with the grandchildren's parents. In many cases, not only had the participants not been asked explicitly to care for their grandchildren, but they could not even remember a specific time when they had offered to act as a substitute mother in the absence of their daughter (-in-law): I hadn't thought about it, it just came up. . . . She needed someone and I was there. . . . I saw she was very worried so I said to her: "Don't worry, I'll stay with him." They didn't ask me, but it was clear that I was going to be the one who would care for her baby when she had one. (Elena, 60) I didn't ask her and she didn't ask me. . . . (laughs). I suppose we both knew what was going to happen when her maternity leave ran out. (Hortensia,72) The transition to the auxiliary caregiver role is even smoother and more "natural" when certain factors are present. Among these facilitating factors the participants highlighted the following: the proximity of their own home to that of their children; having regular contact with their children, which even tended to increase when grandchildren were born; there being no other available grandmother who could assume or share the care tasks; and having acted as an auxiliary caregiver in the past with older grandchildren. In addition, all the widows who participated in the study mentioned that not having a husband made their caregiver role easier, as they had plenty of time without commitments and caring for grandchildren was a way to fill that time.
Most of our grandmothers mentioned a sense of duty toward their children as the main motivation for looking after their grandchildren. This duty is perceived as being part of their parental responsibility, i.e., they started caring for their grandchildren not so much out of love for them, but rather due to the responsibility they felt toward their own children. As parents they felt obliged to take care of their children and promote their well-being, particularly in difficult times or when help was needed. Thus, looking after their grandchildren was a way of helping their children, and they seemed to regard their involvement in childcare as a way of fulfilling their own parental role. Indeed, they often argued that their role as caregiver was a sign of their commitment to the family:

The Terms of the Contract: Nonintervention Plus Facilitation
Having talked about their care tasks and how they became auxiliary caregiving grandmothers, many of the participants went on to describe the conditions under which they perform their role, that is, the terms of the tacit agreement they had reached with their daughter or son. Daughters-in-law and, particularly, sons-in-law are rarely mentioned when those terms are discussed.
The first topic that emerges is the voluntary or somehow "forced" nature of their role. As was mentioned above, most of the grandmothers defined their role as a "duty," an expression of their parental commitment to their family (and, specifically, to their children). However, this does not prevent them from believing that they have freely accepted the role of looking after their grandchildren. Indeed, some grandmothers even said that they would have been disappointed were they not taking care of their grandchildren: Yes, one could say that it's an obligation . . . and it is, but it's a selfimposed obligation! Who made me do it? Nobody did . . . let's be clear about that. Bringing up children is a mother's duty, but I chose to do it with my grandchildren and I feel proud of being useful. (Aurelia, 60) I would have been very angry if my daughter hadn't thought that I wanted to care for my grandson. (Valentina,74) This complex balance between parental duty, as a motivation to offer childcare, and free will, as a basic condition of the tacit contract of care, is exemplified in opinions about grandmothers who do not want to take care of their grandchildren. Most participants understood that such women were entitled not to do it and recognized that caregiving is a difficult task that sometimes interferes with other leisure, social, or educational activities. However, at the same time, those grandmothers who didn't want to offer childcare were labelled by participants as self-centered, particularly if they were engaged in other kinds of discretional activities instead, that is, if they had chosen not to care for their grandchildren: I respect their opinion . . . my daughter-in-law's mother, for instance. . . . She says she can't look after our grandson because she goes to bellydance classes. OK. And she comes to see our grandson whenever she wants to, when she's bored . . . and that's OK too. . . . I'm not criticizing her, but I would never ever do it. For me, family always comes first. (Alicia,76) Maybe there're grandmothers who can't do it because of their health. But if you can, then you must do it. . . . If you don't do it for your children and grandchildren, then who are you going to do it for? It's the best inheritance you can leave them. (Fina,71) Another condition of the tacit contract of caregiving is that grandmothers assume caring tasks that would normally be associated with the parental role, but this by no means implies that grandmothers become substitute parents. For most decisions about their grandchildren, and particularly in regard to discipline and upbringing, grandmothers are subordinate to parents, who maintain control in these areas. Grandmothers are supposed simply to apply and support the criteria established by parents or, if possible, postpone decisions in these areas or in other complex situations until parents are present. Thus, authority over the children remains the parents' prerogative. Most of the participants agreed with this arrangement, which they perceived as a relief of responsibility. I always say that the parents are in charge of their upbringing, not me. So when my grandchildren ask me to do something, I tell them: "Ask your parents, if they allow you to, then you also have my permission." (Juana,79) Grandparents should enjoy their grandchildren without spoiling them. I can tell them off, but without contradicting their parents' rules. It is their job to bring them up the way they like. (Maria,65) A further condition of the caregiving contract is that the adult children should facilitate the caregiving task of grandmothers. Grandmothers depicted their own children as doing all they could to reduce to a minimum the burden and inconvenience attached to caregiving. Indeed, they felt their role was valued, thanks to the effort of their children to make things easier, and 21 out of 24 grandmothers mentioned some kind of facilitation strategy in this regard. The most often mentioned of these were avoiding giving them any additional tasks apart from childcare, picking grandchildren up as soon as possible, and giving the grandchildren food or pocket money if necessary.
And my daughter says to me: "Mum, you don't have to do this, his father should do it," and even my son-in-law says to her: "Honey, your mum is not your servant, she's only here to look after the kid, not to clean the house." (Ángeles, 76)

Consequences of Caregiving
The responses indicated that the caregiving experience was fully satisfactory for the grandmothers who participated in our study. None of them reported feeling regret, troubled, bothered, or resigned as a result of looking after their grandchildren, even though they sometimes acknowledged that it was hard work.
When asked about why it was such a satisfactory experience, the main reason given had to do with the intrinsic rewards derived from helping to bring up their grandchildren. Being able to offer love and help in relation to their grandchildren's learning and development was a great source of satisfaction for the participants in the study, who expressed a strong emotional attachment to their grandchildren. In their view, their role as auxiliary caregivers gave them an advantage over grandmothers who did not take care of their grandchildren regularly. As caregivers they could participate more actively in their grandchildren's upbringing, but at the same time, as grandmothers, they did not have the responsibility of making decisions in complex situations or need to get involved in negative discipline-related interactions, because such tasks remained the job of parents. Furthermore, the auxiliary caregiving role was being assumed at a time in life when grandmothers had more free time to devote to their grandchildren and have already had the experience of being a parent, thereby freeing them from the insecurities of youth. All in all, the participants portrayed their role as highly satisfactory, and sometimes said that it gave them the opportunity to enjoy a child's upbringing in a relaxed way, which they hadn't necessarily been able to do with their own children: Bringing up children is really lovely. . . . Caring for older people is important, I know, but it's harder because you see yourself reflected in them, and you know that in a few years' time you'll end up like them. But with grandchildren, you live a second youth . . . helping build a life and things like that, you know what I mean? I like to see them grow and progress. When I had my own children I was very busy and I didn't enjoy them properly, but now I'm getting the chance. I tell them stories, I buy them clothes, I spend all the time that is necessary so that they can grow up happy. (Nieves,65) In addition to the intrinsically rewarding task of offering care and love to their grandchildren, some grandmothers also said that feeling loved by their grandchildren and being a part of their happiness was a particularly strong source of satisfaction. Such love is depicted as being expressed spontaneously and sincerely by their grandchildren: She loves me so much and is always asking her parents: "Can I stay with grandma? Will grandma come with us?" She can't do anything without her grandma! (Valentina, 74) He appreciates any little thing I do for him, and he really means it. When I go to pick him up from school he comes running up to me and says "Grandma, what have you brought me today? Today I've been very good! Will you buy me an ice cream?," and I say: "The biggest one there is, my sweetheart." And he kisses me . . . and that's what makes me happy. (Angelina,73) Apart from these intrinsic rewards most grandmothers interviewed in the study also mentioned some extrinsic rewards. The most frequent of these was seeing the caring task as a way of avoiding what they regarded as the risks of older age: boredom, isolation, depression, or being left out. Thus, taking care of their grandchildren made them feel active, alive, and useful and was a creative way of occupying their time and renewing family ties: For me it's a kind of hobby. I've always been very active and I'm never happy when I'm just sitting around. So grandchildren are a real godsend for me, they keep me on my toes and make me feel that I'm still important in the family. (María O.,66) When I'm with them I really enjoy it. . . . Otherwise I'd be thinking, "I'm alone at home, I have nothing to do" and days can be very long with nothing to do. But with them around you don't have time to think about what's worrying you right now. (Valentina,74) This function of grandparenting was particularly relevant for the widow grandmothers who participated in the study. For them, grandchildren are a way of finding company and a purpose in life: I . . . when my husband was here . . . we always liked children, and I was happy, but now I thank God for having my grandchildren. They save me from loneliness. When they leave I miss them very much. (Valentina,74) This optimistic view would seem to challenge the concept of "slave grandmotherhood," which refers to burned-out grandmothers who experience their role as a burden that affects their health and well-being. Indeed, and as can be seen in some of the above excerpts, while many grandmothers acknowledge that the caregiving role is hard and sometimes tiring, the rewards always more than compensate for that negative side of caring, and none of the participants saw themselves as a "slave grandmother." Even when they were asked explicitly about the negative aspects of caregiving, they managed to frame those negative things, if they recognized that there are some ones, into a general positive portray.
However, although they never applied the concept of slave grandmotherhood to themselves, 14 out of 24 grandmothers did identify a grandmother stereotype that is similar to the slave grandmother. Specifically, they spontaneously spoke about grandmothers who they felt were taken advantage of by their children, either through being asked to take on too many hours of care or by being persuaded to assume caring tasks that were beyond their ability or means. These grandmothers were seen as not wanting to take on such a caregiver role, or were somehow forced into doing so. Thus, as well as the intensity or difficulty of the caring tasks that are assumed, the adult children's attitude and the grandmother's age and health would seem to be key factors in relation to the notion of slave grandmothers: Sometimes children take advantage of their parents a bit by making them spend too many hours caring for their grandchildren. Maybe this happens more with daughters . . . they don't take more advantage but they're more confident when it comes to asking . . . although my daughter-inlaw has never done that. (Elena,60) The older the grandparent, the more difficult it is. I see some grandparents who can hardly walk . . . they seem to use the pushchair as a walker . . . and I wonder: But how do they do it? They don't have the strength, the agility or the reflexes anymore . . . nothing. (Eli,64)

DISCUSSION
The first objective of the study was to explore the meaning that grandmothers gave to their experience as regular auxiliary caregivers. In this regard, the participants defined the situation that led them to provide care in terms of family need, i.e., the birth of grandchildren created a necessity that posed a risk to their own children's stability and well-being. This necessity usually had a financial basis, both directly, as their children couldn't afford formal care services, and indirectly, since by looking after their grandchildren they enabled the mother to continue working and contribute financially to the household. Thus, the grandmothers who participated in the study saw themselves as filling the gap for hard-working parents (and particularly mothers) in a context in which affordable formal care services were lacking (Hank & Buber, 2009;Tan, Buchanan, Flouri, Attar-Schwartz, & Griggs, 2010). Consequently, our findings suggest that the role performed by these grandmothers was what in the literature has been referred to as "family watchdogs" or "safety nets" (Troll, 1983), that is, grandmothers who monitor their family's situation and provide help in case of need.
Once the situation has been defined as a need, their participation in order to resolve it by offering care seems, to them, to be totally justified. Grandmothers also believed that such participation, although voluntary, is part of a sense of parental responsibility that continues even when their children are independent adults. Providing care to grandchildren is thus a way of being "a good mother," which for most of them meant being able to sacrifice time and effort for the good of the family. At the same time, the caregiver role reinforces their value and status within the family and gives them a sense of purpose and utility, which are some of the main sources of satisfaction associated with the role. Especially in the case of widows, caring for their grandchildren filled their time and emotional gaps in the same way as their own contribution to childcare filled a gap for working mothers. As such, it prevented them from feeling isolated.
Paradoxically, despite the active commitment of grandparents to family life, the participants continued to follow the so-called rule of nonintervention (Cherlin & Furstenberg, 1985), albeit in a modified version. According to this rule, grandparents neither want nor are expected to interfere with the parents' responsibilities toward their children. In the case of auxiliary caregiving grandmothers, they simply adapt this expectation by shifting the limits of their intervention. Thus, while they do intervene in small everyday aspects of caregiving, any major decisions that are perceived as more far-reaching, and which have to do with discipline and general upbringing, are left to the parents. Indeed, these matters are expected to be part of the parental role, and grandparents simply apply the parents' rules or, if in doubt, postpone decisions until the parents can decide what to do. This modified form of the nonintervention rule has two consequences: First, it helps grandmothers to perform their role with a sense of freedom, which is another source of satisfaction. They can thus enjoy their grandchildren's growth and have an influence on their development, but without the burden of responsibilities and discipline-related interactions, which are usually negatively tinged. Second, the rule modification highlights the commitment of their own children as (good) parents, despite their having delegated part of the childcare to grandmothers and missing out on many daily situations with their children because of work.
This positive attitude toward their own children's role as parents is another strong trend found in the responses of grandmothers. All the participants described their relationship with their children in terms of harmony and good coordination in the provision of care. Children did not force their mothers to offer care, and the grandmothers also highlighted how their children tried to facilitate caring tasks, this being mentioned as additional evidence of the parents' commitment to their child's upbringing.
This picture contrasts with the more mixed balance between benefits and costs that is reported in studies of custodial grandparents, who perform their caregiver role on a full-time basis and assume the whole range of parental responsibilities. Such a result is consistent with literature that found that being responsible for full-time care implies some kind of stress, but that looking after a grandchild part-time could be welcomed (Bowers & Myers, 1999). Providing auxiliary care for grandchildren could be seen as a continuation of the traditional role of family carer that women of older generations in Spain have developed throughout all their lives. It would be interesting to ascertain if this positive view appears also among grandfathers, traditionally more work focused, or among the next generations of grandmothers, who will have been fully involved in the labor market.
The second objective of the study was to explore whether the motivations and outcomes of the grandmothers' experience as auxiliary caregivers could be interpreted as an expression of generativity, where this is understood as a desire to nurture and guide younger generations, and which in turn implies personal growth. As for the outcomes of their experience as caregivers, the intrinsic rewards mentioned by participants seem clearly linked to generativity: They valued their participation and the influence they could have on their grandchildren's development and growth. Furthermore, this "involved" type of grandmotherhood gave them the opportunity to guide the youngest generations of their family and benefit them, which, in turn, enhanced their own feelings of well-being. Many grandmothers also mentioned the love they received from their grandchildren and the important role they had in the latter's lives as being a source of satisfaction. This could be related to the "need to be needed," which is often cited as being at the core of the generativity concept (Erikson, 1963;McAdams, 2001). By contrast, participants also described those grandmothers who could offer childcare but chose not to as being self-centered, this being representative of stagnation, or the opposite of generativity.
Generativity also seems to play a role in terms of what led the grandmothers to offer childcare in the first place. However, in this case generativity is not focused on grandchildren, but rather on their own children, the grandchildren's parents. In general, they got involved as caregivers because they wanted to contribute to their children's well-being and saw it as an expression of good parenting. Watching over and caring for their children is viewed as a lifelong task that does not end when they leave home to start their own family. There is a sense of duty here, and yet for nearly all the grandmothers interviewed in the study this was a duty that was freely assumed, one that brought them satisfaction, as well as being a way to enhance their importance and value within the family.
Thus, although Erikson's original proposal restricted generativity to middle adulthood, the present findings suggest that generativity may also account for certain experiences of older people. Parenting and grandparenting are ideal contexts in which to express generative concerns and behaviors, which may have both personal and social benefits (Villar, in press). On a personal level, performing generative acts can help to give meaning and value to older peoples' lives, foster growth and gains in late life, and, consequently, become a way of growing old successfully. On a social level, such expressions of generativity in older age challenge the stereotypical view that regards aging as a time of loss and decline, highlighting the contributions of older people to their family, the community, and society as a whole. Therefore, a focus on generativity in older age means emphasizing a view of older people as a resource and as providers of care, rather than as a burden and recipients of care.
Finally, there are several limitations to the present study that need to be addressed in future research. First, although the strategy to recruit participants was designed to obtain a diverse sample, we are aware that the study was conducted with a self-selected group of grandmothers who volunteered to share their experiences as auxiliary caregivers. In that sense the group could be biased because grandmothers with a positive experience of that task are more likely to share their feelings and opinions. By definition they were highly committed to their families, which could also make it more difficult for them to openly acknowledge the costs of that commitment and, particularly, to speak in negative terms about their children. That fact may account, at least in part, for the lack of negative caregiving experiences, even when the interviewers explicitly asked for them. These negative experiences only appeared indirectly when grandmothers recognize the presence of some kind of "slave grandmotherhood," which they do not experience at all.
Despite these limitations the study does have implications for policy and practice. Above all, it underlines how, particularly in the absence of affordable formal care options, grandparents are of key importance in terms of their children's attempts to reconcile family and professional responsibilities. Reinforcing the formal care network and implementing other measures so as not to overburden grandmothers (e.g., a better coordination between school and work schedules) could help make their role as caregivers more sustainable. Such a role is not only a potentially significant source of personal satisfaction but also one that merits greater recognition within social and policy agendas.