The domestic group is the unit of people who live together and carry on the everyday business of domestic life — cooking and eating from a common hearth, pooling and consuming resources, bearing and rearing children, caring for the sick and the old, and organising the labour that sustains all of this. It is defined by what it does and where its members are, not by how they are related. This is precisely why anthropologists distinguish it from the family, which is a category of kinship — a set of relationships of descent, marriage and filiation that may be distributed across several residences, or several of which may be crowded into one.
Keeping the two apart matters because they do not coincide. A widowed mother living alone is a domestic group but hardly a family; two married brothers who eat separately in adjoining rooms of an ancestral house are one family in the jural sense and two domestic groups in the practical one; a migrant labourer remitting wages from Surat belongs to a family whose economic functions cross two households. David Bender's classic refinement therefore separates three variables that ordinary usage fuses — kinship, co-residence and domestic functions — and insists that each be established empirically rather than assumed from the others.
Fortes and the developmental cycle
Meyer Fortes, introducing the Cambridge symposium on The Developmental Cycle in Domestic Groups (1958), made the decisive move. A domestic group, he argued, is not a fixed form but a process in time, passing through a regular cycle whose phases correspond to the reproductive career of its members. He distinguished three:
- Expansion, running from marriage until the last child is born — the group grows in numbers and its dependency ratio rises.
- Dispersion or fission, beginning with the marriage of the eldest child, as children establish domestic arrangements of their own and the property and authority of the parental generation begin to devolve.
- Replacement, in which the parental generation dies or retires and its place is taken by that of the children, so that the cycle begins again.
Fortes's point is that these are phases of one cycle, not different types of family, and that the cycle is generated by the interplay of the reproductive, economic and jural or political domains of the domestic group.
The census illusion
From this follows an argument of great methodological force. A household census is a synchronic slice through a population of domestic groups, each caught at a different point in its cycle. The resulting table of nuclear, extended and joint households therefore records phases, not types. A group counted as nuclear may be one in expansion whose grandparents have died, or one that has just completed fission and will be joint again when its sons marry. Statements such as the finding that a majority of Indian households are nuclear are, taken alone, close to meaningless — they describe the age structure of households as much as the norms of the society. To read family structure one needs either genealogies over time, or repeat enumeration of the same households, or the reconstruction of cycles from life histories.
The Indian joint-family debate
The point transformed Indian family studies. A. M. Shah, in The Household Dimension of the Family in India, argued that the long controversy over whether the joint family was disintegrating rested on a confusion between the household as a residential and commensal unit and the family as a field of kinship obligation. Working with nineteenth-century and modern census material from a Gujarat village, he showed that simple households have always been numerous in India, that they are generated by demography and the cycle rather than by modernity, and that the joint family is better described as a numerically weak but normatively dominant form. I. P. Desai's study of Mahuva made the complementary case: jointness is a matter of degree, measurable by the persistence of obligations in property, ritual and mutual aid, so that men who eat separately may remain functionally joint. Both conclusions follow from taking the domestic group as a process.
An illustration makes the trap concrete. Enumerate a lane of ten houses in a Gujarat village and find six simple households, three lineal joint and one collateral joint. It is tempting to conclude that nuclearisation has advanced. But if three of the six simple households contain young couples whose parents are alive and living next door, two contain elderly widows whose sons have married, and one is a migrant's wife with children, then not a single one represents a settled nuclear family type. The same ten houses re-enumerated in fifteen years would yield a quite different distribution with no change whatever in norms.
Household as a unit of enumeration
Because the household is administratively convenient, it has become the basic unit of the Census of India, the National Sample Survey and every poverty and welfare schedule — usually defined by a common kitchen together with common residence. The convenience carries costs. The criterion cannot easily handle joint families that have split kitchens while sharing land, seasonal and circular migrants, servants and attached labourers, women in polygynous unions maintaining separate hearths, or the so-called female-headed household created by male out-migration. Nor does a headcount reveal the intra-household distribution of food, work, schooling and health care, which is where much gender and age inequality actually lives. Feminist scholarship has pressed this hard: treating the household as a single unit with one welfare function conceals the bargaining and the deprivation inside it.
For the UPSC answer
Open by separating the domestic group as a functional and residential unit from the family as a kinship category, and cite Bender's three variables of kinship, co-residence and domestic function. Then give Fortes's cycle by its three phase names — expansion, dispersion or fission, replacement — and draw the methodological conclusion that a census records phases, not types. Carry this straight into the Indian debate: A. M. Shah on the household dimension and I. P. Desai on jointness as degree together dissolve the claim that the joint family is disappearing. Finish with the measurement problems of the census household, including intra-household inequality, to show command of both theory and method.
References & further reading
- Fortes, M. (1958). Introduction, in J. Goody (ed.), The Developmental Cycle in Domestic Groups. Cambridge University Press.
- Fortes, M. (1949). The Web of Kinship among the Tallensi. Oxford University Press.
- Shah, A. M. (1973). The Household Dimension of the Family in India. University of California Press.
- Desai, I. P. (1964). Some Aspects of Family in Mahuva. Asia Publishing House.
- Bender, D. R. (1967). A refinement of the concept of household: families, co-residence and domestic functions. American Anthropologist, 69(5).
- Laslett, P. and Wall, R. (eds.) (1972). Household and Family in Past Time. Cambridge University Press.