In this final chapter we interrogate the notion – common to grand theory and government statements, to academic analysis of family change and policy research – that we are living in a period of transition. We show that although there are significant changes in household composition and in the proportion of the population that chooses to partner and parent, there are important continuities in the practices by means of which we ‘do’ family and that, given the overall rate of social change, family lives exhibit a quite surprising degree of continuity with the past. This is particularly unexpected given Rosser and Harris's predictions that increasing occupational and geographical differentiation would lead to a decrease in extended family cohesion and that the de-domestication of women would mean that the effectiveness of the extended family, ‘as a mechanism of support in need’, would depend on ‘the willingness of women to accept the burdens involved’ (Rosser and Harris, 1965, 290). The main finding of their study was that a modified extended family had emerged, which was

Our findings are surprisingly similar and show, above all, that extended family networks are resilient and are reproduced across time and space through family practices. Moreover, friends become ‘like family’ when they engage in similar practices, providing support and coming together for the major rituals of family life surrounding birth, death and the different occasions that mark progress through the life course. Contra those who argue that ‘the family’ has changed dramatically, we suggest that, while the extended family has been further modified in the years since 1960, the elementary family has not fragmented under pressure of time–space distanciation, nor been undermined by processes of individualisation, and is still crucially supported by a penumbra of wider kin.