6 | India through 1871-72 Census: Findings from the census report

By Gaurav Kalyani and Shivakumar Jolad (author details below)

In the previous article we discussed the process of making the 1871-72 Census. Despite the acknowledged flaws and significant exclusions, the 1871-72 Census  produced a wealth of statistical data that revealed the extraordinary diversity and complexity of Indian society. This article will look closely at its findings, and explore them through a thematic account of different categories.

Population 

The data on area and population (c. 1865–72) reveals that there are directly administered provinces with a substantial share of territory under feudatory (princely) states. Of the total 1.45 million square miles and nearly 239 million people, about four-fifths of the population and three-fifths of the area were under British administration, while the rest lay in indirectly governed regions such as Hyderabad, Rajputana, and Central India.  As noted in the previous article, many feudatory states, Kashmir , districts under British India, and frontier areas (NEFA/ current Arunachal Pradesh), Lushai Hills (Mizoram– not yet annexed) were excluded. 

Large provinces like Bengal, Madras, and the North-Western Provinces dominated demographically, forming the core of British rule, whereas major princely states contributed significantly to the overall population but remained outside direct administrative control. Bengal alone had over 62.6 million people, far exceeding provinces like Bombay (25.6 million) or Punjab (22.9 million), while smaller regions like Coorg had barely 0.17 million. Feudatory states were also uneven—Hyderabad ( 10.7 million, in 1867) and Rajputana (9 million) were large demographic units, whereas others like Manipur had just 0.13 million, highlighting the fragmented and unequal structure of colonial India.

Population Density

British India averaged 211 persons per square mile (165 including feudatory states) however, provincial densities varied greatly with Oudh at highest (468 persons per sq. mile), followed by Bengal (397) and North West Provinces (378). British Burma recorded the lowest density of 31 persons per sq mile. The population density in the Ganges Valley (current Uttar Pradesh and Bihar) and the neighbouring districts had an average density of 480 persons per sq mile. This itself exceeded Belgium by 7% and England by nearly 14%. In the North‑West Provinces, 12 districts exceeded 500 per sq. mile, and in Oude, 7 of 12 districts did so. It was also noted that despite the famines and epidemics, there was a notable trend in population growth, which was attributed to increase in the area of cultivation, improved hygienic conditions and introduction of public health measures. 

In Appendix, we provide tables for the largest districts in Bengal and North West Provinces and Oudh.  In Bengal, districts like Hooghly (1,045 persons/sq mile) and 24 Parganas (951) were extremely dense, showing high urbanization and commercial concentration in and around Calcutta, while larger districts like Midnapore (500) and Birbhum (518) were relatively sparse. In contrast, districts in the North-Western Provinces and Oudh show moderate but more even densities—for instance, Benares (797) and Lucknow (697) are dense, but most others cluster between 500–650, indicating less extreme variation. Overall, Bengal appears as a region of high-density pockets alongside large agrarian districts, whereas the northern plains show a more uniform settlement pattern.

Houses and Urbanization

Dwellings were categorized by their structure (masonry/pucca vs. mud/kutcha*) to serve as a measure of the prosperity of the people. The census recorded 37,041,468 inhabited houses, averaging 41 houses per square mile and 5.14 persons per house. Notably, Indian towns had fewer persons per house than villages because many town dwellers were traders whose families remained in their home villages. There were 493,444 villages or townships, averaging 75 houses and 386 persons each.

Table 1: Top ten cities with the largest population.

Age and Sex

The total population by sex was 98,054,403 males and 92,501,565 females, giving the sex ratio of 94.34 females per 100 males and revealing a significant gender disparity. The census also categorized the population by age, primarily distinguishing between children (under 12) and adults (above 12). Children under 12 numbered 66,844,343 (35.1% of the population), adults over 12 numbered 122,929,112 (64.5%), with 782,513 age unspecified and 7,080 sex and age unspecified.

Table 2: Summary of population by Age and Sex in the 1872 census

The sex ratio varied enormously by province. The proportion of children to adults was much higher than in England. The report also indicates a very low rate of life and high mortality among adults. The lowest was Ajmer and Coorg at about 48.75 children per 100 adults; the highest was the Central Provinces with 61.41, attributed to greater adult mortality and early marriage practices.

Table 3: Provincial breakdown of Age and Sex ratios

** Ajmer anomaly was noted and a re-Census was ordered

Female Infantcide: 

The girls-to-boys ratio (under 12) provides one of the clearest signals of female infanticide and neglect, since distortions at young ages cannot be explained by mortality differences alone. 

The census also provided a first statistical confirmation of practice of female infanticide, practiced particularly among certain high‑caste communities in northern and western India. It was attributed to the ‘heavy burden’ felt by fathers/family, where concerns over marriage alliances and dowry created strong disincentives to raising daughters. The evidence was drawn from the striking imbalance in sex ratios, especially among children, and the correlation between the proportion of higher castes and the deficit of females. This is reflected in highly skewed sex ratios in Punjab (1.20), NWP (1.14), and Bombay (1.10), compared to near-balanced ratios in Madras, Mysore, and Bengal (around 1.00–1.04).

In regions like the North-Western Provinces, a large proportion of female births did not survive infancy. Across British India, the census recorded 98 million males and 92.5 million females – a surplus of 5.5 million men. Boys under 12 outnumbered girls under 12 by over 4.5 million across India. In some districts the imbalance was stark: one village in Meerut had only 8 girls for every 80 boys. As Dyson (2015) notes and reports corroborates, the imbalance was greatest “where the higher castes are in the greatest proportion” (Dyson, 2015; Waterfield 1875).

Efforts to curb the practice had been made since the early nineteenth century, but in 1870 the British government passed an Act empowering officials to apply special regulations to districts or villages suspected of female infanticide. The census data gave statistical weight to this policy and identified the areas where enforcement was most needed.

Religion

The British started classifying the Indian population into religious groups early on. The 1872 census recorded that approximately 19 out of every 20 people in British India were either Hindu (Hindoo) or Muslim (Mahomedan). The total population was divided into approximately 140.5 million Hindus (including Sikhs), 40.75 million Muslims, and 9.25 million belonging to other groups, including Buddhists, Jains, and Christians. Hindus were most prominent in the south, making up 95% of Mysore and 92% of Madras and Coorg. Muslims were most numerous in the north, forming 53% of the Punjab population and 49% of the population in Eastern Bengal. The Sikhs were almost entirely concentrated in the Punjab, specifically in the region between the Ravee and Sutlej rivers. 

The Buddhists were largely concentrated in the region of British Burma, which accounted for 86% of the total Buddhist population. In the rest of India, the population of Buddhists is unclear as it was counted along with the Jains. Christians represented less than one in 200 people. About five-sixths of the total Christian population resided in the Madras Presidency, although the report questioned how many of the native Christians can be genuinely considered Christians. Other religious groups included hill tribes, aborigines, Parsees and Jews (most of whom were residing in Bombay).

Caste

The census attempted to classify by caste but encountered immense difficulties. The report explicitly acknowledges this failure of systematically classifying caste, due to lack of uniform classification plan across different provinces. Each provincial report writer adopted a method suited for their specific region, resulting in a vast but fragmented collection of data. Yet the data that were collected offer a remarkable snapshot of India’s complex social hierarchy, revealing the numerical strength of different castes, their geographical distribution, and the vast number of groups that defied easy categorization.

Table 4: Classification of Hindus according to castes noted in 1872 census

The table reveals that the vast majority of people of Hindoo origin were grouped under “other castes”, a catch-all term that included hundreds of distinct communities. The “aboriginal tribes” and out‑castes together accounted for nearly 18% of the total, reflecting the large population that stood outside the traditional Varna system classification. This was a precursor to the refinement of caste and tribe groupings, which ultimately led to Scheduled Caste and Scheduled Tribe classification in 1931.

Brahmins, the priestly caste, were the most thoroughly enumerated. Across British India they numbered 10,131,541, or about 5.3% of the total population of Hindu origin. Their distribution was uneven, with the highest concentrations in the Gangetic plain and the south. The census noted that Brahmins were not confined to religious duties and in Benares alone, 107 distinct Brahmin sub-castes were recorded. The Kshatriyas/Rajpoots were concentrated in northern India with many recorded as soldiers, landowners and cultivators. They also formed a significant part of the armies in Bengal and Bombay.

The provincial census reports contained detailed lists of castes, often running into hundreds of names. It is not feasible to reproduce those tables here. 

Although Islam doesn’t recognize caste, the census found that Muslims in India had developed social divisions similar to castes, which was viewed more as a tribal classification in the 1872 census report. The census recorded four chief branches: Syuds (791,000), Sheikhs (4.7 million), Pathans (1.8 million), and Moghuls (219,000).

Nationality and Language

The 1872 Census report admits that although people can broadly be grouped into two religions, ‘nationality’ is far more diverse. Although it is important to note that ‘Nationality’ was not considered in the modern sense of political nationality, i.e. belonging to a nation-state, but rather used loosely to denote ethnic-cultural groups tied to a specific region, descent or way of life. The census categorized the population into four broad groups based on this idea of nationality and origin: 

• Natives of India, 

• Asiatic Non-Indians, 

• Mixed Races, and 

• Non-Asiatics. 

Interestingly, the report states that the enumeration of the European population (Non Asiatics) was the "least accurate portion of the whole inquiry,” with significant errors found in municipal returns for cities like Calcutta.

In the category Natives (Hindoo), the title of ‘Hindoo’ was used broadly to include all persons of Hindu origin, even if they were now Native Christians, Buddhists, or Aboriginal tribes who did not strictly follow the Hindu religion. For individuals of Muslim origin, the category Natives (Mohamedan) was employed. 

The second category of ‘Asiatics Non-Indians’ meant Asiatics comprised groups like Parsees, Jews, Persians, immigrants from border states like Nepalese, Afghans, Arab, Armenians and Chinese.  There was some administrative debate regarding certain groups, specifically the Beloochees, as to whether they should be considered Asiatic foreigners or part of the native Muslim population.

The category of Mixed Races was specifically defined to include Eurasians and Indo-Portuguese. Whereas, Non- Asiatics were defined as persons other than those of Asiatic birth. This broad category included British subjects (English, Scotch, Irish, and Welsh), individuals from continental Europe as well as Americans, Africans, and Australasians.

 Table 5: Population distribution across Nationalities as per the 1872 census

Although the 1872 census collected information of Language, it was not tabulated uniformly. Nevertheless, the census highlights that language often defined regional identity and influenced the social character of the people. While Hindi (Hindee) and Bengali were found to be among the most widely spoken, the census report describes a linguistic patchwork in the southern regions and a remarkable variety of languages in frontier regions and Burma. Persian and Arabic were noted as languages of the higher classes or those learned in literature, particularly in cities like Peshawur (Peshawar) or among scholars (Pundits and Moulvees). The following table summarizes the language diversity recorded across the major provinces:

Table 6: Language distribution across provinces

Occupation

The 1871–72 Census of British India represents the first systematic attempt to classify the population by occupation at a subcontinental scale. However, this occupational data must be interpreted with caution. The census itself acknowledges serious methodological inconsistencies such as - in some regions only adult males were classified, in others entire households were assigned a single occupation. Moreover, women were frequently recorded under the occupation of male heads and enumerators struggled with ambiguous categories such as ‘service’, which could refer to either government employment or domestic work. Nevertheless, the census offers a valuable snapshot of the economic and social structure of late 19th-century India revealing a predominantly agrarian society with limited industrialization and a socially constrained occupational order.

The census provides an approximate distribution of occupations among roughly 62 million adult males, which could be interpreted as a proxy for the economic structure. This distribution shows that economic activity was overwhelmingly concentrated in the agricultural sector, over 56% of adult male employment. However, agriculture was far from homogeneous; it encompassed a wide range of social and economic positions. These distinctions highlight a layered agrarian hierarchy, including landlords, tenants, and cultivators with varying degrees of rights over land. The category ‘cultivator’ itself obscures significant variation in tenure, control, and economic security. 

The census also identified Domestic servants (6.2%), reflecting hierarchical social relations and Independent and non-productive classes (3.4%), including pensioners, rentiers, and others not engaged in productive work. The domestic servants included over 594,000 barbers and 467,000 washermen. These categories underscore the social stratification embedded in occupational structures, where economic roles were closely tied to status and hierarchy.

The census also provides limited data regarding female labour participation. In most provinces, women were either omitted from occupational returns entirely or included with men in a way that made them inseparable. Detailed information was only recorded and analyzed for Bengal, Assam, and Bombay. 

For the roughly 2.86 million women the occupations specified for them were in the following sectors: Agriculture (407,000 proprietors and 421,000 cultivators). In Bombay specifically, 128,000 recorded as farm laborers), Industrial and Artisan (weaving, spinning, food), domestic and professional. Apart from it, the vast majority were recorded as non-productive or unspecified, under which 140,000 were recorded as beggars and paupers and 59,000 prostitutes.

Table 7: Distribution of Female labor participation in the provinces of Assam, Bengal and Bombay

This occupational data from the 1872 Census offers more than just a statistical profile, it provides a window into the structure of colonial Indian society, which was shaped as much by social institutions and historical hierarchies as by markets or commerce.

Education and Literacy

The 1872 census intended to record the number of persons ‘able to read and write’ or ‘under instruction’, but the effort was hindered by significant administrative gaps and regional inconsistencies. As a result, the educational statistics provided in the census are considered imperfect. 

In Bengal, Education returns were not sought, except for a few municipalities. In North West Province, the data was flawed because the forms failed to include a column for females, resulting in the near-total exclusion of women and girls from educational records. While in Oude (Oudh), census only recorded boys and girls currently in school and excluded counting literate adults. 

The census also tracked education as a profession, recording 189,000 persons engaged in education, literature, and science. Literacy was overwhelmingly concentrated among adult males. In the nine provinces with data, there were approximately 3.87 million educated males compared to just 93,276 educated females.

Land under cultivation and Land Revenue

The 1872 census also sought to measure the extent of cultivated land and the revenue it generated for the state. It recorded approximately 130,720 square miles (30.6%) was under active cultivation. Another 103,486 square miles (24.3%) was identified as capable of being tilled but remained uncultivated. Roughly 190,842 square miles was classified as barren or otherwise unsuitable for agriculture. The North-West Provinces showed the highest intensity of land use, with 77.7% of all cultivable land under tillage. In the eight detailed provinces, 53% of the land was subject to government revenue, quit-rent, or tribute. 

In addition to ordinary land revenue, the census recorded local rates and cesses levied for the maintenance of roads, schools, and village watchmen. These rates were relatively highest in Madras and the North-West Provinces (13.25% to 14.25% of the revenue) and lowest in Bengal and Assam, where they amounted to only 1.75%.

Health, Infirmities

The 1872 census recorded the prevalence of specific infirmities across the population in British India. The categories used during that time period: insane, “idiotic”, deaf and dumb, blind, or suffering from leprosy, might appear insensitive and outdated. Approximately 67,000 persons were recorded in the category of Insanity and Idiocy, a rate of roughly 1 in 2,700. By comparison, this was only one-eighth of the proportion found in England and Wales at the time. The deaf and dump numbered 134,000. The high prevalence was attributed to the commonality of zymotic diseases and a general neglect of sanitation. Over 354,000 were recorded under the category of Blindness, the rate of which was nearly double than that of England at the time. 

Nearly 96,000 lepers were recorded, a proportion about half that of Norway. The census also revealed that the Indian population was remarkably youthful but suffered from a very low rate of life and high adult mortality. Surgeon-Major Lumsdaine noted that the average age in the Bombay Presidency was 11 or 12 years lower, which was lower than in England.

Conclusion

The findings of the 1871–72 census offer a look into a society of extraordinary complexity, serving as the first systematic effort to provide an inventory of the nation for the British administration. Despite its flaws, it established a vital statistical base. It revealed some stark contradictions such as high population density in the Ganges Valley, yet relatively minimal rate of urbanization, as compared to England and Belgium. Equally revealing was the census's failure to systematically classify caste, testifying its complexity that is difficult to capture. Ultimately, the 1872 census did more than just count heads (or in this case households), it laid bare the demographic profile of a complex society and profound regional diversities that continue to remain a challenge even today.

*Precursor to modern classification- Pucca, Semi Pucca, and Kutcha houses.

(Authors: Gaurav Kalyani works as Research Associate at the Center for Legislative Education and Research, FLAME University, Pune; 

Dr. Shivakumar Jolad works as Associate Professor (Public Policy), and is the Chair of Center for Legislative Education and Research and Director India State Stories, FLAME University, Pune.

Gaurav contributed to conceptualization, research and primary writing;  Shivakumar contributed to conceptualization, research and editing)

References and Sources:

Natarajan, D. (1971). Indian Census Through A Hundred Years—Part I. Census of India.

Waterfield, H. (1875). Memorandum on the Census of British India 1871-1872.

The Census of British India of 1871-72. (1876). Journal of the Statistical Society of London, 39(2), 411–416.

(1875). General Report on the Organization, Method, Agency & c. Employed for Enumeration and Compilation: Part I (Census of the Bombay Presidency).

Appendix

1. Population of top districts in Bengal and North West Provinces and Oudh

2. Largest Cities and Towns in British India (c. 1872)

3. Castes and sub castes in Bengal and Assam , and Madras

4. Caste Grouping in Madras (17 Functional Sets)