20 | Literacy and Education at the Turn of the Century, as seen by Census of India - 1901

Published on: 27 June 2026

Introduction: The Census and the Question of Knowledge

The Census of India, 1901 — conducted under the supervision of H. H. Risley, I.C.S., and E. A. Gait, I.C.S., and published in Calcutta in 1903 — stands as one of the most comprehensive demographic surveys ever undertaken in the subcontinent. Spanning the entirety of the Indian Empire from the Himalayas to the southern tip of the peninsula, and embracing Burma and Baluchistan, the census enumerated over 293 million persons and produced a statistical portrait of British India at the threshold of the twentieth century.

Chapter V of Part I (pp. 157–184) is devoted entirely to Education — a chapter of remarkable analytical depth that combines raw statistics, provincial narratives, and sociological commentary to illuminate one of the most persistent structural features of colonial society: the near-universal condition of illiteracy.

Alongside it, Chapter VIII of Part II (Tables, pp. 104–157) provides the empirical grounding— detailed tables of the literate population broken down by province, age, sex, religion, and caste — making the 1901 census an unparalleled source for historical analysis of educational attainment in South Asia at the dawn of the 20th Century.

This essay draws directly from these primary sources, supplemented by recent scholarship, to present a comprehensive account of literacy and education at the turn of the century. It examines the statistical landscape at the provincial level, probes the deep inequalities structured by gender, religion, and caste hierarchy, traces trends since the census of 1881, and analyses the political economy that shaped — and limited — educational provision under colonial rule.

-(Risley & Gait, Census of India, 1901, Chapter V, p. 158

Methodology and Definition of Literacy

The 1901 census represented a methodological departure from its predecessors. In 1881 and 1891, the population had been divided into three categories: literate, learning (i.e., under instruction), and illiterate. However, this tripartite framework had proved problematic. Students in advanced stages of education frequently caused themselves to be recorded as 'literate' rather than 'learning,' regarding the latter designation as derogatory. Conversely, children at the most rudimentary stages of instruction were often omitted entirely from the 'learning' category. A persistent discrepancy also emerged between census figures and those of the provincial Education Departments.

For 1901, the enumeration was simplified into a binary: individuals were classified as either literate or illiterate. The definition of literacy adopted was the ability to both read and write any language. Where a person was recorded as literate, the primary language they could read and write was noted, and English literacy was separately flagged.

This deliberately excluded two intermediate groups: those who could spell out words but could not write, and those whose sole calligraphic accomplishment was the signing of their own name — the latter comprising primarily messengers, menials, and certain persons of higher social rank.

The population covered by the literacy question fell short of the total enumerated population by approximately 946,150 persons in tracts where literacy was not recorded. The Central Provinces adopted the most stringent standard, requiring only persons who had passed the Upper Primary Examination or possessed equivalent qualifications to be treated as literate.

— Formal census instruction to enumerators, 1901

The State of Literacy in 1901

The headline finding of the 1901 census on education is stark. Of a total enumerated population exceeding 293 million across British India, only 53 persons per thousand (5.3%) could be said to be literate. The asymmetry between the sexes was dramatic: approximately 98 males per thousand were literate, compared with just 7 females per thousand. This implied a ratio of roughly 14 literate males for every 1 literate female across the empire.

The age-specific pattern of male literacy was also revealing. Among males under 10 years of age, only 13 per mille had been returned as literate; the proportion rose to 85 per mille between ages 10 and 15, to 132 between 15 and 20, and to 139 at ages 20 and above. The census report interpreted this steady rise with some skepticism, noting that enumerators tended to omit from the 'literate' category persons still under instruction, even where they had progressed beyond the most elementary stages.

The total number of literate males under 15 years of age was reported as 2,129,439 — compared with 675,357 literate and 2,518,240 'learning' at the same age-period in 1891, suggesting an apparent decrease of approximately 33% in recorded youth literacy. This was attributed to the methodological change rather than any real deterioration in educational attainment.

Provincial Analysis: Literacy Across the Empire

The variation in literacy across the provinces of British India was enormous, reflecting differences in the indigenous educational infrastructure, the reach of colonial schooling, caste composition, religious demography, and the degree of urbanisation. Burma, with its dense network of Buddhist monastic schools, stood dramatically ahead of all other large provinces.

Table: Literacy Rates by Province, 1901 (per 1,000 persons of each sex)

Source: Risley & Gait (1903), Census of India, 1901, Chapter V, §§281–302; Table VIII, Part II. '—' denotes sub-national state; figures approximate where not explicitly stated in chapter text.

The Burma Exception

Burma's extraordinary position at the top of the literacy hierarchy demands explanation. The census report attributed it to the ubiquitous institution of the Buddhist monastery (kyaung), where the presiding pôngyi (Buddhist monk) provided free instruction to the children of the surrounding village. The Burmese word for 'school' is identical to the word for 'monastery' — a linguistic fusion that speaks to the deep institutional embeddedness of education in religious life. In certain Upper Burma districts — Minbu, the Upper Chindwin, Shwebo, Magwe, and Mandalay — more than half of all males could read and write.

Among indigenous races, the Burman stood first (490 males and 55 females per mille literate), followed by the Talaings (357 and 62), the Karens and Shans (below half these proportions), the Chins (48 and 2), and last the Kachins (14 and 2). The census report cautioned, however, that monastic educational standards were very low — and that if a more rigorous test were applied, Burma's relative position would fall sharply (Risley & Gait, 1903, §288–289).

Urban–Rural Disparity

One of the most striking findings of the census was the vast gulf between urban and rural educational attainment. In large cities, 259 males and 49 females per mille could read and write — compared with only 98 males and 7 females per mille in the country as a whole. In Calcutta, 316 males per mille were literate; in Bombay City, 249 males and 95 females per mille. The contrast was sharpest in Bengal, where Calcutta's female literacy (115 per mille) was more than eight times the provincial average (Risley & Gait, 1903, §270).

Gender and Education: A Stark Divide

No aspect of the educational landscape of 1901 was more dramatically asymmetric than the gender divide. While the overall female literacy rate across the empire stood at just 7 per thousand, this aggregate conceals enormous variation across provinces. Burma, Cochin, and Travancore exhibited rates that were exceptional relative to the rest of India; large stretches of the subcontinent had female literacy effectively approaching zero.

Table: Male and Female Literacy Rates by Province, 1901 (per 1,000 persons of each sex)

Source: Risley & Gait (1903), Ch. V §§283–302. F:M Ratio = female literates per male literate — lower value indicates greater gender disparity.

Structural Causes of Female Illiteracy

The census report was candid about the social forces suppressing female education. In both Hindu and Muslim communities, it was customary to seclude young girls as they approached puberty, producing a strong prejudice not only against sending girls to school but also against permitting them to be taught at home. These social norms were 'slowly giving way amongst certain sections' of the population — but the pace of change was glacially slow.

The relative freedoms enjoyed by Burmese women — the complete absence of purdah and seclusion customs — were explicitly identified as the primary explanation for Burma's unusual female literacy rates. The anomalous position of Cochin and Travancore was attributed partly to the large Native Christian population and partly to matrilineal social customs that accorded women a more visible social role (Risley & Gait, 1903, p. 164).

Religion and Literacy: Community Differentials

The 1901 census provides detailed breakdowns of literacy by religious community — one of its most analytically rich features. Across the empire, there were dramatic differentials between communities, reflecting differences in formal schooling, in the connections between religious practice and literacy traditions, and in levels of social capital and political organisation.

Table: Literacy by Religious Community, All India, 1901 (approximate, per 1,000 of each sex)

Source: Risley & Gait (1903), Ch. V, §§271–302; Table VIII, Part II. Parsi figures from Berar provincial statistics (§285); Buddhist from Burma (§289). Figures approximate.

Muslim Literacy and the Quranic School

The census report offered a specific structural explanation for Muslim literacy lagging behind Hindu literacy. The learning by heart of long passages from the Quran was considered more important than the acquisition of secular knowledge in the village maktab or madrasah. This prioritisation of oral religious knowledge over read-and-write literacy as tested by the census meant that Muslim communities could sustain rich traditions of religious learning while recording low census rates.

There were significant regional variations, however. In Madras and a contiguous band including Mysore, Hyderabad, Berar, the Central Provinces, and Central India, Muslims showed a higher proportion of literate persons than their Hindu neighbours (Risley & Gait, 1903, §271). The general national position of Muslims was largely determined by the figures for Bengal and Punjab, where the bulk of the Muslim population resided.

Christians and Missionary Education

The Christian community's position in the literacy hierarchy was described in the census as a 'compound' of the very high literacy of Europeans and Eurasians, and the considerably above-average literacy of Native Christians produced by missionary activity. Wherever data were available for Native Christians separately, they were found to enjoy substantially higher literacy than other communities of similarly humble social origin. The dramatic contrast in Bengal between Santali converts to Christianity and those who retained Animist beliefs was cited as direct evidence of missionary impact (Risley & Gait, 1903, §283).

Caste and Education: Hierarchy Encoded in Learning

The relationship between caste and literacy was perhaps the most structurally significant finding of the 1901 census education chapter. The census was forthright in its conclusion: across all provinces, literacy correlated directly with social position in the Hindu caste hierarchy. This was not a coincidence but the product of centuries of exclusion — from both the right to teach and the right to learn — enforced through the social sanctions of caste.

— Risley & Gait (1903), Ch. V, §280, p. 166

The caste education table (Table IX, Part II) was optional and far from comprehensive — complete coverage was achieved only in the Central Provinces, Mysore, and Baroda; nearly three-quarters of the Hindu population in Bengal; and about two-fifths in Madras. The Punjab and several other provinces did not prepare the table at all (Risley & Gait, 1903, §280).

Table: Literacy by Caste: Bengal Province, 1901 (Males per 1,000)

Source: Risley & Gait (1903), Ch. V, §§283–284; Subsidiary Table VI. Figures approximate where not explicitly tabulated; '—' denotes not separately reported.

Table: Selected Castes: Cross-Provincial Comparison, 1901 (Males per 1,000)

Source: Risley & Gait (1903), Ch. V, §§285–302; Subsidiary Table VI. '—' denotes not separately reported. C.P. = Central Provinces.

The Depressed Castes and Structural Exclusion

For the communities at the lowest rungs of the Hindu hierarchy — Untouchables, tribals, and forest groups — literacy was effectively non-existent. In Berar, the Mahars, Dhangars, Wanjaris, Gonds, Kolis, Korkus, and Kolams were described as 'practically all illiterate.' In the Central Provinces, literacy among lower castes and forest tribes was in nearly all cases less than 1 per thousand males.

These findings were not merely a reflection of poverty but of active systemic exclusion. A strong prejudice existed on the part of higher castes against allowing their children to share a school building with children of low or 'unclean' origin; teachers often refused to admit them, or made them sit in the verandah (Risley & Gait, 1903, §274).

The institutional dimension was especially stark. In Bengal, of 137 Education Department officers (excluding 44 Europeans and Eurasians), no fewer than 111 were Brahmans, Baidyas, or Kayasths — 9 were Muslims, 5 were Native Christians, and 12 belonged to other castes. The lower grades of the community were entirely unrepresented in the inspecting staff who decided where schools would be opened and which institutions would receive grants-in-aid (Risley & Gait, 1903, §274).

English Education: The Language of Empire

In addition to vernacular literacy, the 1901 census separately tracked literacy in English — the language of administration, higher courts, and colonial employment. The proportion literate in English was minute even by the standards of overall literacy, yet its social and economic importance was enormous.

Table: Literacy in English by Province, 1901 (per 10,000 persons of each sex)

Source: Risley & Gait (1903), Ch. V, §§282–302. U.P. figure in §295 (per 10,000 for that province); National average from Chapter V introduction.

Within caste groups, English literacy was extraordinarily concentrated. In Bengal, the Baidya caste had 303 males per mille who were literate in English; the Subarnabaniks of Calcutta 268; the Gandhabaniks 176; and the Kayasths 132. The Brahmans held a relatively low position at 74 per mille overall (rising to 157 in Bengal proper). Among the great mass of lower castes, who formed the bulk of the population, there were practically none acquainted with English (Risley & Gait, 1903, §284).

Change Since 1881 and 1891

The census undertook a careful comparison with previous censuses, acknowledging that the change in methodology made direct comparison difficult. To enable meaningful comparison, the report adopted the convention of considering only persons over 15 years of age and treating all who would have been classified as 'learning' in 1891 as literate for comparative purposes.

Table: Trend in Male and Female Literacy, All India, 1881–1901 (per 1,000 population aged 15+)

Source: Risley & Gait (1903), Ch. V, §§272–273. '1901 literate' includes those who would have been 'learning' in 1891 methodology, for ages 15+.

The comparison reveals an apparent paradox: the absolute number of literate persons had grown substantially — literate males aged 15 and over rising from 11,357,996 in 1891 to 12,560,032 in 1901 — yet the male proportion appeared to fall when the full 1901 coverage was used. The census concluded that this apparent male decline was illusory, resulting from the inclusion in 1901 of approximately 32 million persons from backward tracts excluded in 1891. Female literacy showed a genuine proportional increase, rising from 6 to 9 per mille in British territory alone — a 50% proportional gain, though on an extremely small absolute base (Risley & Gait, 1903, §272).

Provincial Trends

At the provincial level, the picture was mixed. Bengal, Berar, and the Central Provinces showed improvement in male literacy since 1891. Bombay remained stationary for males. Burma, Madras, Punjab, and the United Provinces showed declines for males, attributed partly to stricter enumeration standards in 1901. Among Native States, progress was recorded only in Baroda, Mysore, and Travancore. Female literacy improved in all provinces of British territory except the Central Provinces (stationary), while Burma's female literacy showed the most dramatic rise — from 34 to 56 per mille.

Causes of Illiteracy: Colonial Policy, Caste, and Social Structure

Chapter V of the census report contains an extended analysis of the structural causes of illiteracy — an analysis that is remarkably candid about both the failures of colonial educational policy and the internal social dynamics that impeded the spread of education. The causes identified were multiple and interlocking.

Historical Legacy: Pre-British India

The census located the roots of mass illiteracy in the political history of pre-British India. For centuries, the subcontinent had been the 'cockpit of contending dynasties' — rulers whose primary concern was revenue extraction and military supremacy, not the welfare of subjects. The idea that it was the duty of a monarch to govern for the good of his subjects was 'alien alike to Pathan, Moghal and Maratha.' Under such conditions, literacy remained confined to those who depended on it for their livelihood: priests, traders, and accountants. The influence of caste reinforced this — with the imparting of knowledge to Shudras being strictly forbidden in many traditions (Risley & Gait, 1903, §273).

The Economics of Schooling and Peasant Poverty

The lower castes were identified as not only culturally indifferent to education but also economically unable to afford it. Even where fees were nominal, the opportunity cost of sending children to school rather than using their labour — herding cattle, collecting fuel — was prohibitive for the poorest families. Children's contribution to household subsistence was a real and immediate economic constraint that no amount of school-building could easily overcome without direct economic incentives or compulsion (Risley & Gait, 1903, §274).

Colonial Policy Failures: A Centralised, Secondary-Biased System

The total expenditure on education from all sources was Rs 91,29,887 in 1891 rising to Rs 3,16,07,277 in 1901. Yet the distribution of this expenditure was skewed toward secondary and higher education. Scholarship in economic history has confirmed the census's own diagnosis: human capital expenditures in British India averaged just 0.01 pounds per capita from 1860 to 1912, representing a mere 4% of the total colonial budget. Public expenditures on primary schools averaged only 34.3% of the education budget (Chaudhary, 2007).

The emphasis on secondary and higher education — driven by the colonial government's need for trained clerical workers and by the social demand of high-caste communities — meant that the vast majority of the population remained untouched by the formal school system. Chaudhary (2013) confirmed that low public spending, a strong emphasis on secondary education, and the presence of numerous and unequal caste and religious groups collectively constrained the development of human capital in India.

Caste Fragmentation and the Public Goods Problem

A further structural barrier was the effect of caste and religious fragmentation on the provision of public primary schooling. Using district-level data from the 1901 and 1911 censuses, Chaudhary (2007) found that higher caste and religious fragmentation was negatively correlated with the number of public primary schools. This was amplified by the power of high-caste groups to direct educational resources away from communities they perceived as competitors or subordinates — a process clearly visible in the composition of provincial Education Departments described above.

— Risley & Gait (1903), Census of India, 1901, Chapter V, §272, p. 162

Schooling Infrastructure: Distance, Expenditure, and Access

The 1901 census examined not only the results of education but also the physical infrastructure of schooling. Using a standard formula for mean distance between schools, the report calculated how far, on average, a child would need to travel to reach the nearest school if schools were evenly distributed — a significant overstatement of access in rural areas, where schools were highly clustered.

Table: Mean Distance Between Schools and Male Literacy Rates by Province, 1901

Source: Risley & Gait (1903), Ch. V, §278. Distance calculated using formula log d = 1.031235 + log √(1/n), where n = schools per 100 sq. miles. Figures approximate.

The comparison between provinces by school density and literacy rate was deliberately highlighted in the census to make a powerful point: the spread of education did not depend primarily on the multiplication of schools. Burma had more than twice the mean distance between schools as Bengal, yet its literacy rate was five times higher.

Conversely, the United Provinces had almost the same school density as Madras, yet far lower literacy. The census concluded that there were 'large sections of the population who will remain ignorant, however many schools there may be, unless something more is done to attract them than has hitherto been attempted' (Risley & Gait, 1903, §278).

Education Department Returns and the Gap with Census Figures

The census also analysed the relationship between pupils enrolled in Education Department schools and the number of literate persons returned by the census. Two broad groups emerged: Bombay, the Central Provinces, Bengal, and Berar averaged about 5 literate persons per 2 pupils enrolled; Madras, the United Provinces, and Punjab averaged about 7 literate persons per 2 pupils enrolled. The latter suggested that in those provinces, many persons — particularly from trading castes in Upper India — achieved literacy through private instruction or home teaching rather than through the formal school system (Risley & Gait, 1903, §279).

Conclusion

The picture of literacy and education that emerges from the Census of India, 1901 is one of extraordinary deprivation — a condition rooted in centuries of social exclusion, compounded by colonial policy failures, and shaped by the powerful institutional forces of caste hierarchy and religious tradition. At the turn of the twentieth century, only one adult male in ten across the Indian Empire could be said to be literate in the strict sense of being able both to read and write. For women, the figure was fourteen times lower.

The data reveal a society deeply stratified by literacy along every axis of difference. By province, Burma stood as a remarkable outlier, its Buddhist monastic schools creating a broadly literate male population that dwarfed any other region — while Kashmir, at the other extreme, was described as the most educationally backward tract in the entire empire.

By gender, the suppression of female education through purdah and social custom had left women of virtually every community in a condition of near-universal illiteracy, with partial exceptions in Burma, Cochin, and Travancore. By religion, a hierarchy from Parsi and Christian at the top through Jain, Buddhist, Sikh, and Hindu, to Muslim and Animist at the bottom reflected not innate capability but differential access to institutional knowledge. By caste, the correlation between social rank and literacy was near-perfect, with twice-born communities holding a near-monopoly on education.

The colonial state's record was one of structural inadequacy. Public investment in education was minimal as a share of revenue; what investment existed was heavily biased toward secondary and higher education; and the administrative apparatus of education was itself captured by high-caste communities whose interests lay in maintaining rather than disrupting the educational hierarchy.

The 1901 census thus stands as both a monument to the ambition of colonial knowledge-gathering and a record of its own subject's deepest failures. In enumerating illiteracy with such precision, Risley and Gait documented not merely a statistical condition but a profound social injustice — one whose reverberations extended far into India's independent history.

— Risley & Gait, Census of India, 1901, Chapter V, §266, p. 158

References

Primary Sources

1. Risley, H. H., & Gait, E. A. (1903). Census of India, 1901. Volume I — India. Part I: Report. Office of the Superintendent of Government Printing, India. https://archive.org/details/PARI.report-on-the-census-of-india-1901

   
   

2. Risley, H. H., & Gait, E. A. (1903). Census of India, 1901. Volume I-A — India. Part II: Tables. Office of the Superintendent of Government Printing, India. https://archive.org/details/censusindia00unkngoog

Secondary Sources

3. Chaudhary, L. (2007). An economic history of education in colonial India. University of California, Riverside. https://economics.ucr.edu/wp-content/uploads/2019/11/LatikaChaudhary5-6-07.pdf

   
   

4. Chaudhary, L. (2010). Land revenues, schools and literacy: A historical examination of public and private funding of education. Indian Economic and Social History Review, 47(2), 179–204. https://doi.org/10.1177/001946461004700202

   
   

5. Chaudhary, L. (2010). Taxation and educational development: Evidence from British India. Explorations in Economic History, 47(3), 279–293. https://doi.org/10.1016/j.eeh.2009.08.004

   
   

6. Chaudhary, L. (2012). Caste, colonialism and schooling: Education in British India [SSRN Working Paper No. 2087140]. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2087140

   
   

7. Chaudhary, L. (2013, May 1). Caste, religion and fragmented societies: Education in British India. LSE South Asia Blog. https://blogs.lse.ac.uk/southasia/2013/05/01/caste-religion-and-fragmented-societies-education-in-british-india/

   
   

8. Chaudhary, L., & Garg, M. (2015). Does history matter? Colonial education investments in India. Economic History Review, 68(3), 937–961. https://doi.org/10.1111/ehr.12078

   
   

9. Chaudhary, L., Musacchio, A., Nafziger, S., & Yan, S. E. (2012). Big BRICs, weak foundations: The beginning of public elementary education in Brazil, Russia, India, and China. Explorations in Economic History, 49(2), 221–240. https://doi.org/10.1016/j.eeh.2011.08.001

   
   

10. Chaudhary, L., & Rubin, J. (2011). Reading, writing and religion: Institutions and human capital formation. Journal of Comparative Economics, 39(1), 17–33. https://doi.org/10.1016/j.jce.2010.12.003

    
    

11. Chaudhary, L., et al. (2023). Railways, development, and literacy in India. Journal of Economic History. https://doi.org/10.1017/S0022050723000372

    
    

12. Dirks, N. B. (2001). Castes of mind: Colonialism and the making of modern India. Princeton University Press.

    
    

13. Fuller, C. J. (2017). Herbert Risley, William Crooke, and the study of tribes and castes in India. Journal of the Royal Asiatic Society, 27(4), 593–617. https://eprints.lse.ac.uk/84172/1/Fuller_Ethnographic%20inquiry_2017.pdf

    
    

14. Ghosh, S. C. (2000). The history of education in modern India 1757–1998. Orient Longman.

    
    

15. Than Nwe Sein. (n.d.). A brief study on the British education policy in Myanmar. University of Computer Studies, Lashio, Myanmar. https://www.uclmyanmar.org/wp-content/uploads/2018/05/Than-Nwe-Sein.pdf

    
    

16. Wood, C. (1854). Dispatch on General Education in India, 1854. Court of Directors of the East India Company. [Reproduced at Advocatetanmoy.com] https://advocatetanmoy.com/despatch-on-general-education-in-india-1854-education-in-colonial-india/