Divorce leaves its mark on the entire kinship system as relatives, particularly grandparents, adjust to the changes incurred by parents leading separate lives. The effects of the upheaval can be quite profound: for some grandparents, it means a closer relationship with their grandchildren as parents seek support and comfort. For others, often on the non-resident parent's side, divorce means losing contact. The situation can become even more fragmented and complex when parents repartner and introduce step-relations.

The reorganisation is a dynamic process, of course, rather than a one-off event, as parents deal with the crisis of marriage breakdown, and make progressive adjustments (Cherlin and Furstenberg 1986; Johnson 1988). Nevertheless, such authors describe the subject of the effects of divorce on intergenerational relationships as largely unexplored, although intriguing.

It is worthwhile considering what constitutes a family in Western societies in the late twentieth century. The nuclear family, comprising a couple and children living separately from kin, predominates today. The Institute's 1987 follow- up study, Parents and Children After Marriage Breakdown, showed that even with a high divorce rate, separated parents still tended to gravitate towards forming nuclear families again.

Of all family types established after separation, this was the most stable and the only one that became more common with time, although there remained a large group of sole- parent families headed by women (Weston 1992). Despite their poverty, few of these mothers and children were living with relatives when the Institute conducted its initial Economic Consequences of Marriage Breakdown study in 1984 (some two to five years after the respondents' separation), or when the follow-up study was conducted in 1987, although at the time of separation, around 30 per cent of mothers who left the marital home went to live with their parents for a period.

Although longer lives have resulted in an increasing number of surviving grandparents in the post-Second World War era, the large extended family is still often looked upon as a treasure now lost. The popular image of this so-called 'traditional' family has members valuing loyalty towards each other, sharing in the same interests and commitments, and working together to promote family stability and continuity (Johnson 1988). Yet as Cherlin and Furstenberg (1986) have pointed out, the shorter lives of a few generations ago meant that many children did not have any grandparents, and most did not live with their grandparents.

The nuclear family is often viewed as a small, isolated entity, more fragile than it was in the past, in which members' behaviour remains private from outsiders (Johnson 1988). There are several concerns about very private, isolated families: they are less open to outside beliefs, attitudes and values; they pose a barrier to external sources of support; and, as George (1988) noted, they can allow patterns of child abuse to be established and maintained while preventing children from receiving compensatory acceptance, nurture and support from outsiders.

But just how isolated and private is the nuclear family? Although separated from relatives in terms of residential location, there is a great deal of evidence in Australia and overseas that ties with relatives remain strong (d'Abbs 1992; Cherlin and Furstenberg 1986; Lentjes and Jonker 1985). Research suggests that family members, including grandparents, maintain regular contact with each other (largely owing to the initiative of female members) and are important sources of mutual support. Migrant families, however, lose the opportunity to draw on such sources of support if they leave extended family behind in the home country.

THE MEANING OF GRANDPARENTS

When looking at the effects of divorce on intergenerational relationships, it is also worthwhile to consider the nature of the grandparent-grandchild relationship today.

The meaning of grandparenthood has changed (Cherlin and Furstenberg 1986; Denham and Smith 1989; Funder 1989; Peterson 1989). Whereas children and grandchildren used to defer to the elderly, relationships now tend to be characterised more by affection than power. Although grandparents assume greatly varying roles, the predominant role in Western society seems to be that of an easy- going companion who interacts with the grandchildren in a light- hearted manner.

Other potential grandparent roles include acting as historians, mentors, models for ageing, caregivers or surrogate parents, and as sources of other forms of support. Whatever the roles, they remain voluntary and negotiable.

Grandparenting, of course, is not always viewed positively. While grandparents may offset overly critical parents by their affirmation of the grandchildren, they may also create tensions within the nuclear family. Indeed, problematic relationships between mothers-in-law and daughters-in-law are the grain of truth in many jokes. It appears that disagreements between grandparents and parents typically centre around child rearing (Peterson 1989).

FACTORS AFFECTING ROLES

The roles and importance of grandparents in the lives of their grandchildren are affected by several factors, including the parents' marital status and their activities (parents often control the levels of contact), the quality of the relationship between grandparents and parents, cultural differences, the age and sex of the grandchildren and grandparents, distance between residences, and the precise link (maternal or paternal) of grandparents to grandchildren (Peterson 1989).

Given the greater propensity for women to initiate contact with their extended family, several studies (some others disagree) have found that maternal grandparents, and especially maternal grandmothers, are more important in the lives of grandchildren than any surviving paternal grandparents (Eisenberg 1988; Peterson 1989).

Eisenberg (1988) found that young adult grandchildren saw their maternal grandparents more frequently, felt closer to them, and liked them more than their paternal grandparents. Further, these grandchildren preferred maternal grandmothers in particular, and they engaged in a wider range of activities, especially enjoyable ones, with maternal grandmothers than with other grandparents. Although distance influences contact with grandparents, the grandchildren in this study did not live significantly closer to one set of grandparents than the other.

DIVORCE

Divorce has traditionally been regarded as divisive in terms of grandparent-grandchild contact, with contact between them declining (Peterson 1989). Indeed, some parents have denied grandparents' access to their grandchildren, and this has led to the formation of lobby groups to challenge such moves (Kaganas and Piper 1990). Grandparents, along with their grandchildren, have thus often been portrayed as losers when parents separate (Friedman 1990; George 1988).

However, some grandparents have greater contact with their grandchildren when parents separate. They may assume greater caretaking or supportive roles, possibly reducing the disruption of the parents' separation (Denham and Smith 1989; Peterson 1989). Grandparents may provide stability, support, and a sense of identity and 'roots' to grandchildren when parents separate, although the assumption that grandparents act in the best interests of the grandchildren has also been questioned. George (1988) argues that a child may feel torn between a resident parent and a number of others (including grandparents) pressing for visitation rights.

Maternal grandparents seem particularly likely to maintain contact with their grandchildren; first, because the grandchildren usually live with their mothers, and, second, because fathers are less likely than mothers to initiate contact with relatives. Further, the Institute's follow-up study of divorced parents indicated that children tend to become progressively less important to fathers who no longer live with them, with many mothers lamenting that their former husbands rarely, if ever, see their children (Weston 1990). It seems likely, then, that for many children in this situation, contact with paternal grandparents will decline.

What happens, however, to intergenerational relationships for the minority of children who live with their fathers after divorce? Does contact with grandparents largely switch to those on the paternal side?

The Institute's study provides an excellent opportunity to examine intergenerational relationships from the parents' point of view. (Funder (1989) has already described these relationships from the children's point of view). At the time of the second survey (in 1987), there were 56 fathers in this sample who were living with at least one of the two children of their former marriage (here called 'resident fathers'). The views of these men were compared with those of the 277 resident mothers and 178 non-resident fathers in the sample. (There were only 12 non-resident mothers too few to make generalisations.)

Parents were asked about how often the children had contact with each set of grandparents (via letter, telephone or visits), whether the divorce affected relationships between children and grandparents, and the importance of the grandparents in the children's lives and in providing support to the respondents in their role as parents. At this stage, the average ages of the two children of this first marriage were 13 and 15 years, and the parents had been separated for around five to eight years.

CONTACT WITH GRANDPARENTS

More than 80 per cent of both resident and non-resident parents (measured separately) reported that their children had contact with at least one set of grandparents at least weekly or monthly. Further, around one-third of resident mothers and non-resident fathers, and 27 per cent of resident fathers said that their children had contact with both sets of grandparents weekly or monthly. In this sample, then, the extended family was very much 'alive'.

Of the remainder, around 10 per cent of parents in each of the three groups indicated that either the maternal or paternal sets of grandparents had died, and 10 per cent of men (mainly non-resident fathers) did not know how much contact their children had with their maternal grandparents. Only three women felt unable to report how often their children saw their paternal grandparents.

As Figure 1 shows, the amount of contact with maternal or paternal grandparents is shaped by the living arrangements of the children. Frequency of contact with each set of grandparents, as reported by resident fathers, is almost the exact opposite of the frequency of contact reported by resident mothers.

Where children were living with their mother, the majority of parents reported that the children had contact with maternal grandparents at least once a month (79 per cent of resident mothers and 68 per cent of non-resident fathers), with around half stating that contact occurred at least once a week. Only 27 per cent of resident fathers said their children had contact with their maternal grandparents at least monthly, with only 15 per cent reporting weekly contact.

As expected, frequent contact with paternal grandparents was more likely when the children were living with the father rather than mother. Eighty-four per cent of resident fathers reported at least monthly contact (with half indicating at least weekly contact), while 35 per cent of resident mothers and 44 per cent of non-resident fathers reported at least monthly contact with paternal grandparents (with only 12 and 14 per cent respectively reporting at least weekly contact).

Although frequency of contact was higher for grandparents on the resident parent's side, as previous figures indicate, at least monthly contact was reported for many grandparents on the non-resident parent's side. Parental separation, then, does not necessarily imply loss of ties with grandparents on the non-resident parent's side.

Some children rarely or never had contact with one set of grandparents, but only around 5 five per cent of parents in each group indicated that their children rarely or never had contact with both sets of grandparents. Forty-six per cent of resident fathers indicated that their children had little if any contact with their maternal grandparents a situation that was more likely when the children no longer saw their mothers.

Where the children were living with their mothers, 39 per cent of women and 29 per cent of men said that the children rarely or never had contact with their paternal grandparents. Once again, frequency of contact was influenced by whether or not the children saw their non- resident parent. Where the non-resident father no longer visited his children, most parents indicated that the children rarely or never had contact with their paternal grandparents, while such low contact was reported by less than one- quarter of parents who indicated that the non-resident father continued to visit the children. Continued contact between children and grandparents on the non-resident parent's side, then, often seems to depend on this parent.

QUALITY OF RELATIONSHIPS

Figure 2 shows that, regardless of with whom the children lived, the most common response was that the divorce had no effect on the relationship between the grandparents and children. This was particularly the case with maternal grandparents when the children were living with the mother, and with paternal grandparents when the children were living with the father.

Around 40 per cent of resident mothers and non-resident fathers felt that the divorce had a damaging effect on the children's relationship with their paternal grandparents, while the same proportion of resident fathers said this of the children's relationship with maternal grandparents. Improved relationships were most commonly reported for maternal grandparents where the child was living with the mother (15 and 22 per cent) and for paternal grandparents and children where the child was living with the father (24 per cent).

Damaging effects, then, were more likely to be perceived for relationships with grandparents on the non-resident parent's side. These trends are consistent with those concerning frequency of contact, and support the view that some grandparents (usually paternal grandparents) are 'victims' of divorce. It should be noted, though, that reduced contact may be a welcome relief for grandparents and grandchildren if they already had a poor relationship.

IMPORTANCE OF EXTENDED FAMILY

Parents rated the importance of each set of grandparents and other relatives in the lives of their children on a scale of 1 ('not at all important') to 5 ('very important'). The average (mean) ratings of resident and non-resident parents are shown in Figure 3. Once again, the patterns of answers given by resident mothers and fathers were mirror images. Resident mothers considered maternal grandparents to be the most important followed by other relatives on their side, then paternal grandparents with other paternal relatives being the least important. By contrast, resident fathers considered paternal grandparents to be the most important, followed by other paternal relatives, then maternal grandparents, and finally other maternal relatives.

Compared with resident parents, differences in ratings given by non-resident fathers concerning the importance of each set of relatives were small, although like the resident mothers, non-resident fathers considered maternal grandparents to be the most important and other paternal relatives to be the least important in their children's lives.

As for the other issues discussed, these trends suggest that living arrangements strongly influence which set of extended family is important. When children live with their mothers - by far the most common arrangement - the pre- existing bias towards maternal relatives, as suggested in past research, continues and may well be strengthened.

For the minority who live with their fathers, links with their maternal family are typically weaker than links with paternal relatives. This may derive from an upheaval in relationships between children and grandparents since separation - changes rarely mentioned in discussions of the effects of divorce on children. Alternatively, relationships with extended family may have been atypical for these unusual families before separation.

There were important exceptions to these trends, however. For example, around 20 per cent of resident mothers and non-resident fathers considered the paternal grandparents to be 'very important' in the children's lives, while 17 per cent of resident fathers said this about maternal grandparents.

While the trends so far discussed were shaped by residential arrangements, the importance of a family's support for parents was shaped by blood ties (Figure 4). That is, fathers (whether resident or non-resident) and resident mothers typically indicated that their own parents were the most important providers of support, followed by other relations on their side, with parents and other relations on their former spouse's side reported as providing little support.

However, of all the ratings of support shown in Figure 4, the highest was that received by resident mothers from their own parents. In statistical terms, their ratings were significantly higher than those given by fathers in relation to their own parents.

IMPLICATIONS

The extended family is important for parents and children alike. The Institute's divorce study shows that the bias towards maternal relatives continues and may well be strengthened when the children live with their mothers, while ties with paternal kin are comparatively weak. In fact, 40 per cent of parents of these children claimed the divorce had damaged relationships with paternal grandparents.

The pattern is different when children live with their fathers. The bias towards paternal relatives may derive from a weakening of ties with maternal kin since separation, and a strengthening of ties with the wider paternal family. However, resident-father households are unusual, and relationships with the father's family may also have been atypical before separation.

It appears that loss of contact with grandparents on the non- resident parent's side occurred not so much through the resident parent denying access, as through the resident parent leaving the responsibility for maintaining contact with their former spouse. Where their former spouse failed to visit the children, the children were also unlikely to have much to do with the grandparents on that side.

Parents typically saw their own parents and other relatives as important sources of support, with resident mothers in particular emphasising the importance of their own parents.

These results are consistent with those of Johnson (1988), based on interviews with grandmothers and their divorcing children. As mentioned above, a number of mothers who left the matrimonial home upon separation went to live with their parents, but by 1984 most had moved out.

Johnson found that the dependency of parents, where it existed, often created tensions between parent and grandparent, but the strain was temporary. When the crisis was over, parents and grandparents distanced themselves somewhat, in a manner that improved relationships. Given the heavy reliance on relatives during the crisis, an important research question is: how do families who do not have extended family within easy reach manage at this time? Who helps out?

Johnson's study shows the importance of examining changes in relationships with the extended family, as parents make progressive adjustments after divorce. One common adjustment is repartnering, whereby children inherit step-relatives. As Sanders and Trygstad (1989) noted, roles of step-grandparents in the family are unclear and families need to deal with these issues.

Sanders and Trygstad conducted a survey of young adult college students in the United States who were either grandchildren or step- grandchildren, and found that relationships with grandparents were more central in the lives of grandchildren than were relationships with step- grandparents. Grandchildren had more contact with their grandparents and described the relationship as more important and stronger, than was the case for step- grandchildren in relation to their step-grandparents. Further, more grandchildren than step- grandchildren reported that the existing level of contact fell short of that desired. The shorter history of the step-relationship was not solely responsible for these trends.

This Institute report is based on the views of parents rather than grandparents and grandchildren. Comparisons between the views of the three generations remain an important area of research, along with longitudinal studies that identify changes in relationships between grandparents and grandchildren as each grows older, and as circumstances (such as financial resources, marital status of parents) change.

As Peterson (1989) noted, future research is also required into the possible influences of ethnic background and socio- economic status, being an 'only' grandchild, as opposed to one of many grandchildren, the effects of being the first grandchild or a later grandchild, the effect of grandparents' psychological wellbeing on quality of grandparenting, and the dimensions along which grandparent - grandchildren relationships vary.

This discussion has focused on the relationship of grandparents and grandchildren after the parents divorce. The issues raised, however, are relevant to other situations, including the death of a parent, and adoption or foster care of children.

REFERENCES

• Cherlin, A. and Furstenberg, F.F., Jr. (1986), The New American Grandparent: A place in the Family, A Life Apart, Basic Books, New York.
• d'Abbs, P. (1991), Who Helps? Support Networks and Social Policy in Australia, Mongraph No.12, Australian Institute of Family Studies, Melbourne.
• Denham, T.E. and Smith, C.W. (1989), 'The influence of grandparents on grandchildren: a review of the literature and resources', Family Relations, Vol.38, pp.345-350.
• Eisenberg, A.R. (1988), 'Grandchildren's perspectives on relationships with grandparents: the influence of gender across generations', Sex Roles, Vol.19, pp.205-218.
• Funder, K. (1989), 'Grandparents in children's post-divorce families', Family Matters, No.25, December.
• Friedman, L. (1990), Why Can't I Sleep at Nana's Any More?, Matchbooks, South Melbourne.
• George, J. (1988) 'Children and grandparents: the right to visit', Children Today, November-December, pp.14-18.
• Johnson, C.L. (1988), Ex Familia: Grandparents, Parents and Children Adjust to Divorce, Rutgers University Press, New Brunswick.
• Kaganas, F. and Piper, C. (1990), 'Grandparents and the limits of the law', International Journal of Law and the Family, Vol.V, pp.27-51.
• Lentjes, S. and Jonker, J.M.L. (1985), 'Social support network: a literature study', in J.A. Yoder (ed), Support Networks in a Caring Community: Research and Policy, Fact and Fiction, Martinus Nijhoff Publishers, Dordrecht.
• Peterson, E.T. (1989), 'Grandparenting', in S.J. Bahr and E.T. Peterson (eds), Aging and the Family, Lexington Books, Lexington, Massachusetts.
• Sanders G.F. and Trygstad, D.W. (1989), 'Step-grandparents and grandparents: the view from young adults', Family Relations, Vol.38, pp.71-75.
• Weston, R. (1990), 'How is it going to affect the kids? Parents' views of their children's wellbeing after marriage breakdown', Family Matters, No.27, November.
• Weston, R. (1992), 'New families, new finances', Family Matters, No.31, April.