Caste, class and family in India

In short

Caste, class and family are the three institutions through which Indian society organises status, resources and reproduction. Caste is a hereditary, endogamous, ritually ranked group; class is an economic position defined by property, occupation and income; family is the kin group that produces, socialises and inherits. Their interaction explains Indian realities that neither alone can — why a landowning farmer caste votes as a bloc, why a Dalit graduate still faces exclusion, why dowry rises with family income. Examined in BA/MA Sociology papers on Indian society, in UPSC Sociology and GS Paper I, in UGC NET Sociology Unit on Indian social structure, and used professionally in CSR and development work, HR diversity-and-inclusion roles, market research on household consumption, journalism, and public-policy and welfare administration.

Explained

Caste: definition, features and theories

• Caste (Portuguese *casta* = breed, lineage) is a hereditary, endogamous, usually localised group with a traditional occupation, ranked on a scale of ritual purity and pollution, membership in which is fixed at birth.
• Varna and jati must be separated. *Varna* is the four-fold textual scheme (Brahmin, Kshatriya, Vaishya, Shudra) plus those placed outside it; it is pan-Indian, scriptural and only four-fold. *Jati* is the actual endogamous group on the ground — several thousand of them, regionally named, internally segmented into sub-jatis (*upa-jati*) and marriage circles.
• G. S. Ghurye's six features (standard exam list): segmental division of society; hierarchy; restrictions on feeding and social intercourse; civil and religious disabilities and privileges; lack of unrestricted choice of occupation; restrictions on marriage (endogamy).
• Louis Dumont (*Homo Hierarchicus*, 1966): caste is an ideology of purity and pollution; the binary opposition pure/impure is the single principle generating the hierarchy, and status (Brahmin) is disjoined from power (king). Criticised for being too textual, too Brahmin-centred and for ignoring power and politics.
• Materialist critiques: Berreman, Gough and others see caste as a system of exploitation and surplus extraction, not a shared consensus on purity; Ambedkar called it "an ascending scale of reverence and a descending scale of contempt" and "graded inequality" that prevents the oppressed from uniting.
• Ambedkar's thesis (*Castes in India*, 1916; *Annihilation of Caste*, 1936): caste is not division of labour but division of labourers; endogamy, enforced by *sati*, enforced widowhood and girl-child marriage, is the mechanism that closes the group.
• Other classic accounts: Risley's racial theory (now rejected); Nesfield's occupational theory; Hocart and Hutton's ritual and magical theories; Senart's family-cult theory.
• Caste as a colonial product (Dirks, Cohn, *Castes of Mind*): the decennial census from 1871-72, the Ethnographic Survey and legal codification hardened fluid local statuses into enumerated, listable identities.

Caste dynamics: mobility, politics and change

• Sanskritisation (M. N. Srinivas, *Religion and Society among the Coorgs*, 1952): a low or middle caste imitates the customs, ritual, diet and ideology of a higher (often "dominant" rather than strictly Brahmin) caste to claim higher status. It is positional change, not structural change — the ladder stays, individuals move.
• Westernisation and secularisation: Srinivas's paired processes — adoption of Western technology, dress, institutions and values; and the shift of caste from a ritual idiom to a secular interest group.
• Dominant caste (Srinivas): a caste is dominant in a village when it has (i) numerical strength, (ii) substantial land ownership, (iii) political power, (iv) a tolerably high ritual position, and increasingly (v) education and urban jobs. Examples: Jats in western UP/Haryana, Marathas in Maharashtra, Reddys and Kammas in Andhra/Telangana, Vokkaligas and Lingayats in Karnataka, Patidars in Gujarat, Yadavs in Bihar/UP.
• Jajmani system (W. H. Wiser): hereditary, non-market exchange between patron landowning families (*jajman*) and service castes (*kamin/praja*) — barber, potter, washerman, priest — paid in grain and customary dues. Eroded by cash, markets, migration and mechanisation.
• Caste association and "caste in politics" (Rudolph and Rudolph; Rajni Kothari, *Caste in Indian Politics*, 1970): caste supplies ready-made networks for mobilisation; politics in turn secularises and horizontally federates caste. Three forms of mobilisation: vertical (patron-led), horizontal (caste-wide), differential (factional/intra-caste).
• Horizontal stretching vs vertical ties (Béteille): the old vertical dependence between castes in a village weakens, while horizontal solidarity across villages and districts strengthens.
• Dalit assertion: Ambedkarite movements, the Dalit Panthers (1972), Bahujan politics, Dalit literature; Ramaswami Naicker's Self-Respect Movement and Jotirao Phule's Satyashodhak Samaj are earlier anti-caste currents.
• Constitutional and legal position: Article 15 (no discrimination), Article 17 abolishes untouchability, Article 16(4) and 16(4A) enable reservation in services, Article 46 (promotion of interests of SC/ST), Articles 330/332 (reserved seats), Article 338/338A/338B (National Commissions for SC, ST, Backward Classes). Statutes: Protection of Civil Rights Act, 1955; SC and ST (Prevention of Atrocities) Act, 1989 (amended 2015, with special courts and a list of atrocities); Prohibition of Employment as Manual Scavengers and their Rehabilitation Act, 2013.
• Reservation landmarks: Mandal Commission (1980) recommended 27% for OBCs, implemented 1990-93; *Indra Sawhney* (1992) upheld it, capped total reservation around 50% and required exclusion of the creamy layer; the 103rd Amendment (2019) added 10% EWS quota, upheld in *Janhit Abhiyan* (2022); the 105th Amendment (2021) restored States' power to identify backward classes. The Supreme Court in *State of Punjab v. Davinder Singh* (2024) permitted sub-classification within SCs. Caste enumeration beyond SC/ST was last fully done in 1931; Bihar published a caste survey in 2023 and the Union Government announced that caste would be enumerated in the forthcoming Census — treat the schedule and the final design as subject to official notification.

Class in India and its interaction with caste

• Class is a category of people sharing a similar position in the system of production and distribution — in Marx, defined by ownership or non-ownership of the means of production (bourgeoisie/proletariat), with class consciousness and class struggle; in Weber, by market situation, so classes (property/acquisition), status groups (honour, lifestyle) and parties are analytically distinct.
• Caste and class compared: caste is ascribed, endogamous, ritually ranked, closed and legitimised by religion; class is largely achieved, open to mobility, economically defined, with no ritual ranking and no commensal rules.
• Correlation, not identity: caste and class overlap substantially (most manual scavengers and landless labourers are SC; most large landowners are from dominant or upper castes), yet the fit is imperfect — there are poor Brahmins, rich Dalit entrepreneurs (the Dalit Indian Chamber of Commerce and Industry), and sharp intra-caste class differences.
• Agrarian class structure (Daniel Thorner's categories): *malik* (landlord/proprietor), *kisan* (working peasant), *mazdur* (landless labourer). Utsa Patnaik used labour-exploitation ratios; Ashok Rudra and André Béteille debated whether classes in Indian agriculture are "classes" in the Marxist sense.
• Béteille's *Caste, Class and Power* (1965, Sripuram, Tanjore): historically caste, class and power converged in the same families; with land reform, education and electoral politics they are increasingly dissociated — a key exam proposition.
• Urban class categories: industrial working class (largely informal — a very large majority of Indian workers are informal), the old and new middle classes (Pavan Varma, Leela Fernandes on "new middle class" politics), and the capitalist class (business communities: Marwari, Parsi, Gujarati Bania, Chettiar, Nadar).
• Mobility routes: education and reservation, migration and urban anonymity, non-farm enterprise, remittances, and sanskritisation for ritual status. Barriers: credit and network exclusion, discrimination in hiring and housing (documented in audit studies), poor quality schooling.

Family: forms, functions and change

• Family is a group based on marriage, descent and common residence that performs reproduction, socialisation, economic cooperation and care.
• Classification:
  1. By size/structure — nuclear (couple + unmarried children), joint/extended (two or more generations or married siblings sharing residence, hearth, property and worship), supplemented nuclear, stem family.
  2. By authority — patriarchal, matriarchal (rare; matrilineal is not matriarchal).
  3. By descent — patrilineal, matrilineal (Nayar and Tiyyar of Kerala, Khasi and Garo of Meghalaya), bilateral.
  4. By residence — patrilocal/virilocal, matrilocal/uxorilocal, neolocal, avunculocal (Nayar *taravad*; Khasi youngest daughter inherits).
  5. By marriage form — monogamy, polygyny, polyandry (Toda; Khasa of Jaunsar-Bawar — fraternal polyandry).
• Irawati Karve (*Kinship Organisation in India*, 1953): four kinship zones — Northern (clan exogamy, village exogamy, no cross-cousin marriage, bride a stranger, *gotra* and *sapinda* rules), Southern (preference for cross-cousin and maternal-uncle/niece marriage, kin married within known circles), Central (mixed), Eastern (tribal, diverse). Terminology differs accordingly — Dravidian terminologies distinguish cross from parallel kin.
• Joint family's "jointness": measured by co-residence, common kitchen, common property, common worship and functional obligation. I. P. Desai showed that families nuclear in residence often remain joint in function and obligation — so the "decline" is overstated. A. M. Shah distinguished the household (residence unit) from the family (kinship and property unit).
• Forces of change: industrialisation and urban employment, migration, education and women's paid work, housing costs, legal individualisation of property, fertility decline (NFHS-5, 2019-21, TFR 2.0 — just below the replacement level of 2.1), later marriage, and new forms: dual-career couples, single-parent and female-headed households, live-in relationships (recognised for protection under the Protection of Women from Domestic Violence Act, 2005, as "relationship in the nature of marriage"), transnational and "split" families of migrants.
• Problems: dowry and dowry deaths, domestic violence, son preference and sex-selective abortion (child sex ratio concerns; Sex Ratio at Birth improved to 929 in NFHS-5), elder neglect as co-residence falls, intra-household inequality in food, health and education.

Marriage, kinship rules and family law

• Rules of marriage: endogamy (within caste/jati — the core caste rule), exogamy (outside gotra, pravara, village, or *sapinda* range), hypergamy (*anuloma*: woman marries up — associated with dowry inflation and the Kulin Brahmin case in Bengal), hypogamy (*pratiloma*: woman marries down — traditionally disapproved), isogamy, levirate and sororate.
• Descent, lineage, clan: descent traces group membership; lineage is a demonstrable descent group; clan (*gotra*) a stipulated one; *kula*/*kutumba*, *taravad* (Nayar matrilineage) and *coparcenary* (Hindu joint family's property body) are the Indian property-bearing units.
• Current family law — essentials:
  • Hindu Marriage Act, 1955 (monogamy, degrees of prohibited relationship, sapinda rule subject to custom, divorce grounds including mutual consent under s.13-B); Special Marriage Act, 1954 for inter-faith/inter-caste civil marriage; Muslim Personal Law (Shariat) Application Act, 1937 and the Muslim Women (Protection of Rights on Marriage) Act, 2019 (instant triple talaq void and penalised, following *Shayara Bano*, 2017).
  • Hindu Succession (Amendment) Act, 2005 made daughters coparceners by birth with equal rights; *Vineeta Sharma v. Rakesh Sharma* (2020) held this applies whether or not the father was alive on 9 September 2005.
  • Prohibition of Child Marriage Act, 2006 — minimum age 18 (women) and 21 (men); a Bill to raise women's age to 21 has been under consideration and is not law as of now.
  • Dowry Prohibition Act, 1961; IPC provisions on cruelty and dowry death now re-enacted in the Bharatiya Nyaya Sanhita, 2023 (in force 1 July 2024) — cite them as "the cruelty by husband or relatives provision" and "the dowry death provision" rather than by old IPC numbers.
  • Protection of Women from Domestic Violence Act, 2005 (civil remedies: protection, residence, monetary and custody orders); Maintenance and Welfare of Parents and Senior Citizens Act, 2007; Pre-Conception and Pre-Natal Diagnostic Techniques Act, 1994.
  • Inter-caste couples and honour-based violence: the Supreme Court in *Shakti Vahini* (2018) directed police and administrative measures against khap-led honour crimes; *Lata Singh* (2006) upheld an adult's right to marry outside caste.
• Uniform Civil Code: Article 44 is a Directive Principle; Uttarakhand enacted a State UCC (notified 2025), while Goa has long had a common civil code — a live and contested policy question.

Key points

• Caste - hereditary, endogamous, ritually ranked birth group with traditional occupation; jati is the real unit, varna the textual four-fold scheme.
• Ghurye's features - segmental division, hierarchy, commensal restrictions, disabilities/privileges, occupational closure, endogamy.
• Dumont - purity/pollution is the single structuring principle; status is separated from power; criticised as textual and power-blind.
• Ambedkar - caste is division of labourers, not labour; endogamy is its mechanism; "graded inequality" blocks united resistance.
• Sanskritisation - imitation of a higher caste's ritual and lifestyle; positional mobility without structural change.
• Dominant caste - numbers + land + political power + fair ritual rank + education; the real local power holder.
• Jajmani - hereditary patron-client service exchange paid in kind; eroded by cash, markets and migration.
• Caste in politics - caste supplies networks; politics federates and secularises caste (Kothari); vertical, horizontal and differential mobilisation.
• Class - economic position; Marx defines it by ownership, Weber by market situation and separates class from status group.
• Béteille's thesis - caste, class and power are progressively dissociated in modern India, having once converged.
• Thorner's agrarian classes - malik, kisan, mazdur.
• Joint family - common residence, kitchen, property, worship and obligation; nuclear in residence can still be joint in function (I. P. Desai).
• Karve's kinship zones - Northern (exogamy, no cross-cousin), Southern (cross-cousin preferred), Central, Eastern.
• Marriage rules - caste endogamy, gotra/sapinda exogamy, hypergamy (anuloma) and hypogamy (pratiloma).
• Matrilineal ≠ matriarchal - Nayar, Khasi and Garo trace descent through women but authority usually rests with male kin.
• Key law - Article 17 (untouchability abolished), SC/ST (PoA) Act 1989, Hindu Succession Amendment 2005 (daughters as coparceners), DV Act 2005, Dowry Prohibition Act 1961.
• Reservation - Mandal 27% OBC; *Indra Sawhney* (1992) ~50% ceiling and creamy layer; EWS 10% (2019, upheld 2022); SC sub-classification allowed (2024).

Formulas and examples

Sociology has few algebraic formulas; the measures below are the ones examined and used in survey and policy work.

Example 1 — Household size and family type in a village survey. A survey covers 400 households and 2,180 persons; 96 households have two or more married couples or three generations co-resident.

• Mean household size = 2,180 / 400 = 5.45 persons per household.
• Index of jointness = (96 / 400) × 100 = 24%; so 76% are nuclear or single-member in residence.
• Reading: a mean size of 5.45 with only 24% joint households shows large nuclear families, not widespread jointness — a common misreading in exams.

Example 2 — Endogamy and inter-caste marriage. Of 1,250 marriages recorded in a block, 1,163 are within the same jati.

• Endogamy rate = (1,163 / 1,250) × 100 = 93.04%; inter-caste = 6.96% (87 marriages).
• If 20 years earlier the figure was 97.2% endogamous, the fall is 4.16 percentage points — a relative rise in inter-caste marriage of (6.96 − 2.80)/2.80 = 148.6%. Both statements are true; always say which one is being used.

Example 3 — Caste-class correspondence from a land-holding table.

• SC households are 120/400 = 30% of all households but 84/156 = 53.8% of the landless.
• Upper-caste households are 20% of households but 32/52 = 61.5% of large holders.
• Representation ratio for SC among the landless = 53.8 / 30.0 = 1.79 (over-represented); for upper castes among large holders = 61.5 / 20.0 = 3.08.
• Conclusion: strong but incomplete correspondence — 8 upper-caste households are landless and 2 SC households hold large farms, which is exactly Béteille's dissociation argument.

Example 4 — Reservation arithmetic in a recruitment panel. A State advertises 240 posts with 15% SC, 7.5% ST, 27% OBC and 10% EWS.

• SC = 36, ST = 18, OBC = 64.8 → 65 after rounding, EWS = 24; reserved total = 36 + 18 + 65 + 24 = 143 (59.6%).
• Unreserved = 240 − 143 = 97 (40.4%). Note that *Indra Sawhney*'s ~50% ceiling was applied to caste-based reservation; EWS was upheld (2022) as outside that computation, which is why totals above 50% now appear. Open-category seats remain open to all, including reserved-category candidates on merit.

Example 5 — Comparison table for direct exam use.

Where it is used

• Public administration and welfare delivery: identifying SC/ST/OBC/EWS beneficiaries, scholarships, SC/ST sub-plans, atrocity case monitoring, and Socio-Economic and Caste data for targeting — work of IAS/State service officers, District Social Welfare Officers and NCSC/NCBC staff.
• Development and CSR: NGOs and foundations (Pratham, SEWA, Azim Premji Foundation, PRADAN) design interventions around caste exclusion, women's property rights and household structure; monitoring-and-evaluation and field-research roles.
• HR and diversity: workforce composition reporting, PSU and university recruitment rosters, anti-discrimination and POSH policy, campus hiring design.
• Market research and consumer insight: NCCS/SEC household classification by education and asset ownership, joint-versus-nuclear household size driving FMCG pack sizes, white goods and housing demand — used by Nielsen, Kantar, IMRB-type agencies and banks' credit scoring of household income.
• Survey research and official statistics: NSSO/NSO, NFHS, India Human Development Survey, Lokniti-CSDS election studies; field investigator, data analyst and research associate roles.
• Law, journalism and policy advocacy: matrimonial and succession litigation, atrocity and honour-crime reporting, legislative research in think tanks (PRS, CPR, IDFC Institute).

Interview questions

• Q: What is the difference between varna and jati? A: Varna is the scriptural four-fold classification applying across India; jati is the actual endogamous, localised, named group of which there are several thousand, and it is the jati that governs marriage, commensality and occupation in practice.
• Q: Is sanskritisation structural change? A: No. It moves a group's position on an unchanged hierarchy — positional change — because the ladder of purity and the principle of ranking remain intact.
• Q: What makes a caste "dominant" in Srinivas's sense? A: Numerical preponderance, substantial land ownership, political power and a tolerably high ritual rank, now reinforced by education and urban employment; ritual rank alone is not enough.
• Q: Does class replace caste as India urbanises? A: No; the two coexist and interact. Caste persists in marriage, networks and discrimination while class governs market position — Béteille's point is that caste, class and power become dissociated rather than that caste disappears.
• Q: How do Marx and Weber differ on class? A: Marx defines class by ownership of the means of production and expects polarised class struggle; Weber defines it by market situation and separates class from status groups and parties, which fits caste-ridden India better.
• Q: Is the joint family really declining in India? A: Co-residential jointness has fallen with migration, housing costs and smaller families, but obligations over property, ritual and care often continue — so the household has nuclearised more than the family has.
• Q: Are matrilineal societies matriarchal? A: No. Among the Nayar, Khasi and Garo descent and often property pass through women, but authority typically rests with the mother's brother or husband.
• Q: What did the Hindu Succession (Amendment) Act, 2005 change? A: It made daughters coparceners in joint family property by birth, with the same rights and liabilities as sons; *Vineeta Sharma* (2020) confirmed this applies irrespective of whether the father was alive in 2005.
• Q: What is the creamy layer and why does it exist? A: It is the socially advanced and economically better-off section within OBCs excluded from reservation under *Indra Sawhney* (1992), so benefits reach the genuinely backward; the income-and-status criteria are revised periodically by notification.
• Q: Why does dowry persist despite being illegal since 1961? A: It is sustained by hypergamous marriage preference, caste endogamy narrowing the marriage market, daughters' historically weak claim on productive property, and its reinterpretation as "gifts" — enforcement reaches only the extreme cases.
• Q: How is caste relevant to a corporate job? A: It shapes who gets access to networks, internships, references and urban housing; HR roles deal with reservation rosters in PSUs, diversity reporting and complaints of discrimination.
• Q: What was the significance of the Mandal Commission? A: Its 1980 report recommended 27% reservation for OBCs in central services; implementation from 1990 restructured Indian politics around backward-class assertion and was upheld with a ceiling and creamy-layer condition in 1992.

Common mistakes

• Using varna and jati interchangeably, or claiming there are "four castes" in India.
• Treating sanskritisation as structural change, or assuming it is always imitation of Brahmins — it is often of the locally dominant caste.
• Equating caste exactly with class, or the reverse extreme of denying any correlation; the examiner wants "overlapping but not identical", with evidence.
• Calling matrilineal systems matriarchal, and assuming women hold authority there.
• Asserting that the joint family has "disappeared" without distinguishing residence from function, or quoting household size as proof of jointness.
• Citing repealed or renumbered provisions — quoting IPC 498A/304B rather than the corresponding Bharatiya Nyaya Sanhita, 2023 provisions, or treating 21 as the legal marriage age for women (it is still 18; the amendment is pending).
• Giving reservation percentages without the conditions attached — the ~50% ceiling, the creamy-layer exclusion, EWS being treated separately, and the 2024 permission to sub-classify SCs.

Related topics

• Social stratification · Religion and society · Rural and urban sociology · Gender and society · Population and society · Social change and modernisation in India