The Statistics of P C Mahalanobis

Modern government depends on numbers that most citizens rarely see being produced. How much food is being grown? How many people are working? What do households consume? Where is poverty concentrated? A census can answer some questions, but it is too expensive and infrequent to measure everything. Administrative records are incomplete and often shaped by the purposes for which they were collected. Statistical sampling offers another route: observe a carefully designed fraction of a population and infer properties of the whole.

Prasanta Chandra Mahalanobis became one of the central figures in bringing that logic into Indian public life. His name is familiar to statisticians because of the Mahalanobis distance, a measure for comparing observations when several correlated characteristics matter at once. In India his institutional legacy is even broader. He founded the Indian Statistical Institute, helped establish the journal Sankhya, advanced large-scale sample-survey methods and played a major role in building the National Sample Survey. He also became influential in post-independence economic planning.

The unifying theme was measurement. Mahalanobis believed that public decisions should be informed by systematically collected evidence rather than intuition alone. That belief seems obvious today, yet creating credible national statistics in a vast, poor and administratively uneven country required new methods, trained people and organizations. His biography is therefore not only about equations. It is about the construction of a statistical state.

From Calcutta to Cambridge — with physics before statistics

Prasanta Chandra Mahalanobis was born in Calcutta on 29 June 1893 into a prominent Bengali family associated with education and reform. He studied at Presidency College, where physics attracted him, and later went to King's College, Cambridge. His formal training was not initially that of a professional statistician.

A frequently repeated episode in accounts of his life describes his encounter in Cambridge with Biometrika, the journal associated with the development of modern mathematical statistics. The important point is less the anecdote than the direction it indicates. Mahalanobis entered statistics through scientific problems rather than through a conventional statistics curriculum.

On returning to India, he joined Presidency College and remained connected to physics, but statistical questions increasingly occupied him. This interdisciplinary origin shaped his later work. He treated statistics as a practical scientific language that could move between anthropology, meteorology, agriculture, economics and administration.

That breadth was characteristic of an era when statistical methods were still being standardized across disciplines. It also helped Mahalanobis see something institutionally important: India did not merely need isolated applications of statistics. It needed a community of researchers who could develop methods, train practitioners and work with real data from Indian conditions.

Anthropometry and the origin of the Mahalanobis distance

One of Mahalanobis's best-known mathematical contributions emerged from anthropometric data. Researchers wanted to compare groups using multiple body measurements at the same time. A simple difference in height or arm length was inadequate because measurements can be correlated. A person who is tall may also tend to have longer limbs; counting each difference independently can exaggerate how unusual an observation is.

The Mahalanobis distance addresses this problem by measuring multivariate separation while taking the covariance structure of the variables into account. In intuitive terms, it asks not simply how far a point lies from a centre, but how surprising that displacement is given the typical directions and scales of variation in the data.

The idea became foundational in multivariate statistics and later found applications far beyond anthropology: classification, pattern recognition, quality control, machine learning, anomaly detection and other fields. Its modern ubiquity can obscure its historical context. Mahalanobis was trying to solve concrete comparative problems using imperfect observational data.

The contribution also illustrates an important feature of his style. He was interested in mathematics that improved empirical inference. The distance measure was not an isolated abstract result; it belonged to a broader effort to make complex observations comparable in a disciplined way.

A laboratory becomes the Indian Statistical Institute

Mahalanobis began gathering a small group of collaborators around statistical work in Calcutta. What started as a statistical laboratory at Presidency College developed into the Indian Statistical Institute, formally established in 1931.

ISI was unusual because it combined research, training and applied projects. Statistics was not treated as a service department that simply calculated results for other disciplines. It was a field with its own theoretical questions, yet it remained connected to agriculture, industry, demography, economics and public administration.

The institute helped create a professional identity for statisticians in India. It trained researchers, undertook field studies and built relationships with leading international statisticians. Over time it became one of India's most important centres for statistics and related mathematical sciences.

Institution building was central to Mahalanobis's influence. A theorem can circulate through papers, but national survey systems require organizations. They need sampling frames, field investigators, supervisors, data processing, quality checks and a culture that understands error. ISI provided a base from which such methods could be developed and taught.

Sankhya and the creation of a statistical community

In 1933 Mahalanobis founded Sankhya, the Indian Journal of Statistics. The title evoked an Indian intellectual tradition while positioning the journal inside modern international statistical science.

A scholarly journal may seem less dramatic than a national survey, but journals perform institutional work. They create a forum in which methods can be criticized, replicated and extended. They help define professional standards and connect researchers who might otherwise remain isolated.

For a colonized country, the ability to produce rather than merely import scientific literature also carried symbolic weight. Mahalanobis wanted Indian statistical research to participate in global debates on its own terms.

The international orientation of ISI reinforced that ambition. Distinguished statisticians visited, collaborated and interacted with the institute. Mahalanobis himself maintained extensive relationships abroad. His nationalism was therefore not intellectual isolation. Like several major Indian institution builders of his generation, he pursued autonomy through participation in international science rather than withdrawal from it.

Why sample surveys mattered to a poor and enormous country

The practical problem that increasingly occupied Mahalanobis was how to obtain reliable information about large populations without measuring every unit. India made the question unusually difficult. Villages were numerous, infrastructure was limited, records were uneven and social conditions varied enormously across regions.

Sample surveys offered a way to estimate national and regional conditions at lower cost and greater frequency than a full census. But sampling was not simply a matter of choosing a few convenient places. Poor design could produce precise-looking nonsense. Samples had to represent relevant variation, field procedures had to limit bias and statistical formulas had to describe uncertainty.

Mahalanobis and his colleagues experimented with survey designs suited to Indian conditions. Work on crop estimation, acreage and socioeconomic data helped develop techniques of large-scale sampling. He advocated methods in which survey design and field organization were treated as part of the statistical problem rather than as clerical details.

This was a significant intellectual shift. Measurement error, interviewer behaviour and faulty reporting could be as consequential as sampling formulas. A good statistical system therefore required continuous checks on how information was generated.

Interpenetrating samples and the problem of field error

Among the ideas associated with Mahalanobis's survey work was the use of interpenetrating networks of subsamples. In simplified terms, independent subsamples could be assigned to different teams or investigators so that variation in field performance became measurable rather than invisible.

The method addressed a central difficulty of large surveys: non-sampling errors. Even a mathematically perfect sample can fail if investigators misunderstand instructions, respondents give inconsistent answers, measurements are taken differently or supervisors do not detect systematic mistakes.

By designing field operations so that independent work overlapped in controlled ways, survey organizers could estimate components of error and identify problematic procedures. The principle was broader than any one technique. Data quality had to be engineered.

That lesson remains highly modern. Administrative databases, digital platforms and massive datasets can create an illusion that scale guarantees truth. Mahalanobis's survey practice suggests the opposite. The more a society relies on data, the more important it becomes to understand the process that produced them.

The National Sample Survey and a new source of public knowledge

After independence, the need for timely socioeconomic data became urgent. The new government wanted to plan development, allocate resources and measure change, but many basic quantities were poorly known. In 1950, with support from Prime Minister Jawaharlal Nehru, the National Sample Survey was launched.

Mahalanobis played a central role in its creation and methodology. The survey programme was designed to generate recurring information on consumption, employment, agriculture, industry and other aspects of Indian life. Over decades it became one of the major pillars of India's official statistical system.

The significance was constitutional in a broad sense, even though it did not arise from a constitutional clause. A democratic state that claims to improve welfare must possess some credible way of knowing whether living conditions are changing. Surveys make populations statistically visible.

Yet visibility is never neutral. What a survey defines, asks and classifies affects what policymakers can see. Categories of employment, consumption, household structure and poverty embed choices. Mahalanobis helped build the machinery, but later generations would debate how categories should evolve as the economy changed.

Statistics enters the planning process

Mahalanobis's influence expanded from measurement to economic planning. Independent India adopted Five-Year Plans as instruments for directing public investment and accelerating development. The Second Five-Year Plan, covering 1956–61, became especially associated with a strategy that emphasized heavy industry and the domestic capacity to produce capital goods.

Mahalanobis developed planning models that explored how investment allocated between sectors could affect long-run growth. In the simplified two-sector version, the key distinction was between a capital-goods sector and a consumer-goods sector. Expanding the capacity to produce machines could constrain consumption in the short run but raise the economy's future investment capacity.

Later versions added sectors and greater detail. The models influenced the intellectual framework of the Second Plan, though the plan itself was the product of a much wider political and administrative process involving economists, ministries, the Planning Commission and competing interests.

The episode illustrates how Mahalanobis's career crossed a boundary that many statisticians avoid. He moved from estimating what existed to modelling what policy should try to create.

The Mahalanobis strategy: ambition and criticism

The heavy-industry strategy addressed a real postcolonial concern. India imported many capital goods and lacked industrial depth. Dependence on foreign machinery could constrain development and expose the economy to external shortages. Building domestic steel, machine tools, engineering and infrastructure promised greater autonomy.

But the strategy also attracted powerful criticism. Economists questioned whether it underweighted agriculture, employment, exports and the efficiency of investment. A capital-intensive path could generate industrial capacity without absorbing labour fast enough. Foreign-exchange constraints could become severe if industrial projects required imports before they produced substitutes.

Later assessments of India's early planning era remain contested. Industrial foundations created in the period mattered to subsequent development, but the broader system of licensing, controls and public-sector dominance accumulated inefficiencies that cannot be attributed to one model or one person.

Mahalanobis should therefore neither be praised as the sole architect of India's industrialization nor blamed for every weakness of the later "licence raj." His influence was substantial but bounded. The historical value of his models lies partly in showing how newly independent states tried to convert scarce data and developmental priorities into an explicit strategy.

Nehru, planning and the authority of expertise

Mahalanobis's relationship with Jawaharlal Nehru gave him unusual influence. Nehru valued scientific expertise and saw planning as a way to accelerate structural transformation. Mahalanobis offered both quantitative tools and an institutional apparatus capable of producing data.

This closeness between expert and political leadership produced results, but it also raises questions that remain current. How much authority should technical experts have in democratic policymaking? Models simplify reality by design. Their assumptions can privilege some objectives over others. Statistical indicators can make value judgments appear technical.

Mahalanobis himself did not invent technocracy, but his career became one of its most important Indian examples. He represented the belief that disciplined measurement and mathematical reasoning could improve state decisions.

The strongest version of that belief includes humility. Statistics can estimate uncertainty; it cannot decide what a society values. A plan can model investment; it cannot derive distributive justice from an equation. The productive relationship between expertise and democracy therefore depends on making assumptions visible rather than hiding political choices behind numerical authority.

Computing, data processing and an expanding technical culture

Large surveys produce enormous quantities of information. In the mid-twentieth century, processing those data was itself a technological challenge. ISI became involved in mechanical and electronic data-processing methods and later in the development and use of computers.

This connection was natural. Statistical ambition creates computational demand. If a survey collects millions of observations but results arrive years late, the information may lose policy value. Data systems therefore need processing capacity as well as field methods.

Mahalanobis understood statistics as an infrastructure that extended from questionnaire design to computation and analysis. His institute's work crossed disciplinary lines that would later become separate fields: statistics, operations research, computer science and quantitative economics.

That breadth is another reason his legacy cannot be reduced to a single formula. The Mahalanobis distance may be the easiest contribution to name, but the more consequential national achievement was helping build an ecosystem in which numerical evidence could be collected, processed, debated and used.

International standing and a global view of statistics

Mahalanobis participated actively in international statistical organizations and scientific networks. His work on sample surveys was recognized beyond India, and ISI became a destination for researchers from other countries.

This mattered because many newly independent states faced similar problems: limited administrative records, scarce resources and urgent demand for development statistics. Survey sampling offered a way to build knowledge without waiting for perfect bureaucratic systems.

The international dimension also complicates any simple nationalist narrative. Indian statistics grew through exchange with British, European, American and other researchers even as Indian institutions developed distinctive methods. Scientific autonomy was built through networks.

Mahalanobis's reputation helped make statistics part of India's diplomatic identity as well. The country could present itself not only as a recipient of technical assistance but as a producer of methods and expertise relevant to development.

What the numbers could not solve

The statistical state Mahalanobis helped create was powerful, but numbers cannot eliminate politics, poor administration or bad incentives. Surveys are only as useful as the questions they ask, the fieldwork they sustain and the willingness of institutions to respond to inconvenient results.

Official statistics also face a perennial tension between independence and government control. Policymakers need data, yet governments may dislike findings that reveal unemployment, poverty, inflation or programme failure. The credibility of a statistical system therefore depends on professional norms strong enough to survive political pressure.

Another limitation is lag. Economic structures change faster than classifications. Informal work, services, migration, new forms of household production and digital activity can escape categories designed for an earlier economy.

Mahalanobis's legacy should not be invoked as proof that India's statistical problems were solved in the 1950s. It is better understood as a standard: serious public policy requires sustained investment in methods, institutions and transparency, and those systems must be continuously rebuilt.

A legacy in equations, institutions and public reason

P. C. Mahalanobis died on 28 June 1972, one day before his seventy-ninth birthday. By then his influence had spread across several distinct domains. Statisticians knew the distance measure bearing his name. Researchers knew ISI and Sankhya. Government officials worked with survey systems that had become part of routine administration. Economists continued to debate the planning strategy with which he had been associated.

These legacies differ in durability and judgment. The Mahalanobis distance remains a standard mathematical tool. The Indian Statistical Institute remains a major academic institution. National sample surveys evolved through reorganizations and methodological changes. The planning model belongs to a specific historical era and is assessed more critically.

Together, however, they reveal a coherent life project. Mahalanobis believed that societies could reason better if they measured better. He also understood that measurement required institutions, not just formulas.

That is why his biography still matters in the age of big data. Governments now possess information on a scale he could not have imagined, but quantity does not guarantee representativeness, accuracy or public trust. The central question he confronted remains unresolved: how can a country convert observations about millions of different lives into evidence reliable enough to guide collective decisions?

His answer was to combine mathematics with fieldwork, institutions with training, and statistical inference with an ambition to make public policy more empirical. The answer was never perfect. It was foundational.

Sources / Further Reading

Indian Statistical Institute, founder profile of P. C. Mahalanobis: https://dean.isical.ac.in/static/about/our_founder

Indian Statistical Institute, history of ISI and Sankhya: https://dean.isical.ac.in/static/about/publications/sankhya

ISI Digital Commons, historical scholarship on Mahalanobis distance and statistical work: https://digitalcommons.isical.ac.in/

Ministry of Statistics and Programme Implementation, Government of India, material on P. C. Mahalanobis and the National Sample Survey: https://www.mospi.gov.in/

Government of India / Planning Commission historical documents on the Second Five-Year Plan: https://www.niti.gov.in/planning-commission-archive

Scholarly histories of Indian planning and the Mahalanobis growth models, for editorial cross-checking of interpretation.

Suggested Internal Links

The Economics of Amartya Sen

The Engineering of M Visvesvaraya

Planned: How Sample Surveys Changed Indian Policymaking

Planned: The Mahalanobis Model and the Second Five-Year Plan

Approximate article body word count: 2,779