Arvindbhai M. Shah is the sociologist who taught Indian scholarship to be precise about the family. His decisive move was conceptual: he separated the household, a residential and commensal group that can be counted, from the family, a set of kinship relationships that cannot be reduced to a residential unit. Once the two were distinguished, the confident modernisation claim that India's joint family was disintegrating under industrialisation and urbanisation lost its evidential base.
Shah's second major contribution lies in the study of caste. Working on Gujarat with I. P. Desai and later writing on caste in the twenty-first century, he argued that what has changed and what has survived must be stated separately: the caste system — the interlinked local order of ranked jatis with its ritual and economic interdependence — has substantially weakened, while individual elements of caste such as endogamy and caste-based association have proved remarkably durable.
Life and intellectual formation
Shah was trained in the Baroda and Delhi tradition of Indian sociology and belongs to the cohort formed around M. N. Srinivas at the Delhi School of Economics, where he spent the greater part of his teaching career. His fieldwork village, Radhvanaj in the Kheda district of Gujarat, became one of the best documented settlements in Indian sociology, because he returned to it repeatedly and — unusually — combined observation with archival work on nineteenth-century revenue and census records.
That combination of ethnography and historical demography is the signature of his method. Where Srinivas established the field-view against the book-view, Shah added a third source: the record. If one wants to know whether households were larger in the past, he insisted, one must count past households rather than infer them from the normative statements of texts or from the nostalgia of informants.
Household and family
The Household Dimension of the Family in India (1973) opens with a critique of terminology. Writers had used household, family, joint family and extended family interchangeably, and had then read census data on household size as if it measured family structure. Shah's distinctions cleared the confusion.
The household is the group that lives together, shares a hearth and holds property in common — the unit that a census or survey actually enumerates. The family is a field of kinship relationships and obligations that extends beyond any single household: brothers who have separated their kitchens remain kin with claims on one another, and their obligations do not appear in any enumeration. Jointness, therefore, is a property that has to be located: one can speak of a joint household, and separately of the persistence of joint family relations among households that no longer share a roof.
Shah then classified households by structure rather than by size. A simple household contains one married couple with unmarried children; a complex one contains more. Within the complex type he distinguished the lineal joint household, where a couple lives with one married son, from the collateral joint household, where two or more married brothers live together, and from the lineal-collateral combination of both. This gave a usable grammar for the comparative study of household form.
Numerical versus structural jointness
The most consequential analytical distinction is between two senses of jointness. Numerical jointness asks how many people, and how many married couples, live in a household — an arithmetical question. Structural jointness asks what kind of relationships obtain within it, and in particular whether the household includes a married couple in a filial or collateral relation to another married couple. A household of ten containing one married couple and eight unmarried children is numerically large but structurally simple. A household of five containing a widowed mother, a married son and his wife is numerically small but structurally complex.
This distinction destroys the disintegration thesis at a stroke, because the statistical evidence marshalled by earlier writers measured numbers, not structure. Average household size in India has been modest for as long as records exist; small households are not a modern product.
Shah's second instrument is the developmental cycle. A household passes through phases: a couple's establishment, the growth of children, the marriage of sons — which makes it structurally complex — the death of the parents, and division among the brothers, each of whom begins a simple household again. A census taken at any moment catches households at different points in this cycle, and the resulting mix of simple and complex forms is the normal state of affairs, not evidence of decay. To call a snapshot of many simple households the disintegration of the joint family is to mistake a cycle for a trend.
Exploring India's Rural Past (2002) supplies the historical proof. Working from early nineteenth-century records for Radhvanaj, Shah showed that households in a pre-industrial Gujarat village were largely small and simple, with complex households concentrated among the propertied. Large joint households were an achievement of wealth and demographic luck, not a universal traditional norm that modernity dissolved.
Division and coexistence of households
Shah also examined what happens after a household divides. Division of the hearth and the partition of property do not sever the relationship: separated brothers commonly continue to occupy adjacent quarters of the same house or compound, share a well or a courtyard, cooperate in cultivation, contribute to weddings and funerals, and present a united front in village disputes. He described this as the coexistence of households descended from one — a condition in which residential separation and kinship jointness are simultaneously true.
This has two implications. First, the family in India is best analysed as a network in which jointness is a matter of degree, distributed across residence, property, ritual and mutual obligation, rather than as a binary of joint or nuclear. Second, urbanisation and migration do not straightforwardly produce nuclear families: the migrant's household in the city may be simple while his family remains joint across a considerable distance, sustained by remittances, marriage arrangement and return visits. Shah's later essays in The Family in India (1998) extend the argument to demographic change, ageing and the effects of smaller sibling sets on the very possibility of collateral jointness.
Caste in modern India
In Division and Hierarchy (1988), written with I. P. Desai, Shah argued that Gujarat's caste order was organised as much by division — the segmentation of jatis into endogamous units of varying scale, with hypergamous ranking among them — as by the single hierarchical principle that Dumont had emphasised. The Gujarat material, with its wide-ranging Patidar and Rajput marriage circles, showed regional caste structures that a purity-based model handles poorly.
His late essay Caste in the 21st Century: From System to Elements (2007) states the argument that has proved most useful for examinations. The caste system — the village-level interdependence of jatis through jajmani-type services, the enforcement of ritual rank and pollution avoidance in public life, occupational closure, and the caste panchayat's coercive authority — has decayed substantially under law, market, education, urban anonymity and democratic politics. But caste persists as separable elements: endogamy, which remains the most tenacious; caste identity and sentiment; caste associations and welfare bodies; and caste as a resource in electoral mobilisation and in claims on reservation. The mistake, Shah argued, is to treat the survival of the elements as proof that the system endures, or the decay of the system as proof that caste has gone.
Criticisms
Shah's work has been faulted on several counts. His concentration on structure and demography leaves relatively little room for power within the household — feminist scholars note that a classification of household types says nothing about who controls property, who eats last and who is beaten, and that jointness is experienced very differently by a daughter-in-law than by a son. Critics also observe that his corrective can be pushed too far: to show that small households have always been common is not to show that nothing has changed, and the meaning of household division may differ profoundly between a nineteenth-century village and a contemporary metropolis. His caste argument has been criticised by scholars of untouchability for understating the persistence of hierarchy and violence in the countryside, where the weakening of the system has not ended stigma. Finally, the Gujarat base of much of his material raises the usual question about how far it generalises.
Legacy
The distinctions Shah drew are now part of the discipline's working equipment. No competent discussion of the Indian family proceeds without separating household from family, or without invoking the developmental cycle to interpret cross-sectional data. His insistence on historical records alongside fieldwork shaped later historical demography of the Indian household, and his system-versus-elements formulation has become one of the standard ways of answering the question of whether caste is disappearing.
For the UPSC answer
Shah is the indispensable citation on the household dimension of the family, a syllabus topic named in his terms. Begin by distinguishing household from family and numerical from structural jointness, then use the developmental cycle to explain why census evidence of small households cannot demonstrate the disintegration of the joint family — this single argument answers most questions on the subject. Add the Radhvanaj historical material to show that small households predate industrialisation, and the coexistence of divided households to argue that jointness survives as a network of obligation rather than as a residential form. For caste questions, deploy the system-to-elements formulation as your framework for change and continuity, and balance it with the criticism that hierarchy and untouchability persist more stubbornly than the decay of the system implies.
References & further reading
- Shah, A. M. (1973). The Household Dimension of the Family in India. Berkeley: University of California Press.
- Shah, A. M. and I. P. Desai (1988). Division and Hierarchy: An Overview of Caste in Gujarat. Delhi: Hindustan Publishing Corporation.
- Shah, A. M. (1998). The Family in India: Critical Essays. New Delhi: Orient Longman.
- Shah, A. M. (2002). Exploring India's Rural Past: A Gujarat Village in the Early Nineteenth Century. New Delhi: Oxford University Press.
- Shah, A. M. (2007). Caste in the 21st Century: From System to Elements. Economic and Political Weekly, 42(44).
- Shah, A. M. (2010). The Structure of Indian Society: Then and Now. New Delhi: Routledge.