The statistician who "planned" India
I was talking the other day," said William Rogers to the other villagers gathered around the inn fire, 'to a gentleman about the place called Louvain, what the Germans have burnt down. He said he knowed it well - used to Wait a Belgian friend there. He said the house of his P*nd was in a long street, numbered on this side, one, hpo, three, and so on, and that all the numbers on one side of him added up exactly the same as all the numbers an the other side of him. Funny thing that! He said he Awtv there was more than 50 houses on that side of the street, but not so many as 500. I made mention. of the matter to our parson, and he took a pencil and worked ~W the number of the house where the Belgian lived. I dont know how he done it.
Perhaps the reader may like to discover the number of that house.
THIS PUZZLE, written in the working-class dialect in the Britain of the time, appeared in the December 1914 issue of Strand magazine. As you will discover if you try to solve the problem, the solution is not so easy. The answer is House No 204 in a street of 288 houses.
But finding out this answer took only a few minutes for P C Mahalanobis, then studying physics at King's College, Cambridge, after switching from mathematics. Maha'lanobis then took the problem to his mathematician friend Ramanujam, who noted that if the 50-500 limit on houses is removed, there can be several answers.
Ramanujam. then came up with a single continued fraction that provided all the answers.
The anecdote speaks volumes for both men, who were close friends at Cambridge. .Prasanta Chandra Mahalanobis was born in Calcutta on June 29, 1893, into a family that hailed from Panchasar village, now in Bangladesh. The family's original surname was Bandyopadhyaya but switched to Mahalanobis, which was conferred on the family by the Muslim nawab. In 1854, Prasanta's grandfather left Panchasar for Calcutta in search of a livelihood.
Prasanta studied at the Brahmo Boys' School and Presidency College in Calcutta, graduating with honours in physics in 1912. A year later, he went to Britain to study in London, but was so impressed during a visit to Cambridge by King's College Chapel that he decided to study in Cambridge instead.
That was the first of two chance elements that shaped his future. The second was a vacancy for a physics teacher at Presidency College in Calcutta, which came up when Mahalanobis returned to India for a vacation. He had intended to go back to Britain and do research at Cavendish Laboratory, but he tookup the Presidency College post instead as a temporary assignment. He became so engrossed in his work in India, however, this never happened.
Mahalanovis' interest in statistics was fostered by 1@is reading of the journal Biometrika aid Karl Pearson's Biometrik Tables. Wherib he was criticised for abandoning physics for statistics, Mahalanobis explained it.Atistics would be a "key technology" in plaianing the economic develop- ment of independent India, because statistics represents the "arithmetic of human welfare".
The Indian tradition does not stress accuracy in figures, which explains to some extent why the number of soldiers who fought at Kurukshetra or the age of important characters in the Hindu epics are generally exaggerated. Historically, the Indian belief,is that precision in figures has no significance and that the larger the number the greater its respectability.
Mahalanobis' contribution to the development of statistics in India needs to be viewed against this background. He propounded a new approach to figures through his writing on statistics and his responsibilm ities -at the Indian Statistical Institute (11SU and the Central Statistical Organisatioa (CSO), both of which he helped to set up. ISI was launched in 1931; the Indian jour- nal of statistics, Sankhya, in 1933; the National Sample Survey (NSS) in 1950, and in 1941, Mahalanobis was instrumental in forming a statistics section at the Indian Science Congress.
The statistical laboratory that would later evolve into ISI was initially part of the physics department at Presidency College, which Mahalanobis headed. He was also honorary secretary of the laboratory from its formation in 1931 until his death on June 28, 1972. While a professor at Presidency College, Mahalanobis pioneered certain statistical techniques that were used for a jute sample survey in Bengal. This led eventually to the formation of the National Sample Suivey.
The launching of CSO was extremely important in the development of a statistical system in independent India. Then honorary statistical adviser to the cabinet, a post specially created for him in 1949 by Prime Minister Jawaharlal Nehru, Mahalanobis was responsible for India becoming an elected member of the UN Statistical Commission.
Mahalanobis wa's the dominant figure at ISL and referred to always as "Professor". His word was law within ISI and in his later years, some of the younger faculty members began to reAnt his authority and especially his insistence that faculty members had to obtain his permission to attend outside seminars.
Mahalanobis became embroiled in a controversy with the Indian Council of Agricultural Research when he demanded all statistical work in India had to be centralised at ISI in Calcutta, regardless of the subject.
In his later years at ISI, Mahalanobis' interest shifted from pure statistics to economic planninSkand the Second Fiveear Plan is associated with him. The two-sector growth model developed by him and published in Sankhya in 1953, argued for a shift in industrial investments to capital. goods industries. A similar model was developed in the Soviet Union at about the same time.
The question of determining the precise choice of investments that would go into the capital goods industries, Mahalanobis answered by developing a four-sector model, which he published in Sankhya. The Mahalanobis model was perfunctory and did not employ the linear programming techniques that were fairly wellknown at ISI. This drew a lot of criticism.
In hindsight, however, certain characteristics of the Mahalanobis model seem quite remarkable. First, it treated savings as an exogenous variable, which means they were not explained in terms of the behavioural characteristics of households. Perhaps this reflected his training as a physicist, as economists building growth models start with a savings function postulate. Second, the model was autarkic: (selfsufficient) and set in a closed economy. And, third, the model assumed the government could control consumption completely. The second and third characteristics are understandable, given Mahalanobis' socialist background.
In 1955, Mahalanobis became a. fullfledged member of the P.1anning Commission and retained his membership until 1967, serving occasionally as chairperson.
In a paper published in Sankhya in 1957, Mahalanobis traced the origin of qualitative statistical thinking to the system of logic developed in the 4th century by the Jain philosopher Bhadrabahu. Mahalanobis was particularly interested in the multi-valued logic evolved by the Jains. In a 1933 editorial in Sankhya, he asserted evidence existed in Kautilya's Arthashastra showing administrative statistics had reached a high level of organisation in ancient India. He noted also that although written much later, Abul Fazl's Ain-i- Akbari is fundamentally a statistical compilation.
The three over-riding interests in Mahalanobis' life were ISI and statistics, his close relationship with Rabindranath Tagore and the Brahmo Samaj, a reformist @.Hindu group that evolved a dogma in which Western intellectual rationalisation and Christian philosophical and theological strands were interwoven with Hindu tenets.
Mahalanobis attended Brahmo meetings regularly. He garnered funds industriously on the group's behalf, organised its board of trustees in Delhi and was a liberal patron of promising Brahmo youngsters.
Mahalanobis was devoted to his wife, Nirmal Kumarl, daughter of a noted Brahmo Samaj member named Heramba Maitra, and though they had no children, whenever they were apart, they would write to each other every day.
The Mahalanobis home was named Amrapali by, Tagore and though Mahalanobis was neither student nor teacher at Santiniketan, he was a frequent visitor to Tagore's famed institution near Calcutta. When Visva Bharati was set up in 1921. Mahalanobis became one of its secretaries.
Mahalanobis went to the US in 1971 on a lecture tour. On his return, he fell ill. Early next year his condition deteriorated and he underwent abdominal surgery in May. He died on June 28.
In his obituary on Mahalanobis, the founder of the statistics department at the University of California in Berkeley, Jerzy Neyman wrote, "P C Mahalanobis was not just a professor of statistics, as there are thousands of them all over the world. Mahalanobis did more. He tried to do something effective about the problems discussed."
What better tribute can there be?
---Bibek Debroy is a professor at the Indian Institute of Foreign Trade, New Delhi.