Book review: Tim Dyson,A Population History of India: From the First Modern People to the Present Day

The author of this book is Professor of Population Studies at the London School of Economics and so fully qualified to make the ambitious attempt to trace the history of Indian population from the arrival of Homo sapiens down to the last census (2011). For this purpose, Dyson has drawn on a mass of material comprising a large amount of secondary literature, notably for the pre-census period, where references to the demographic aspect may be only incidental. The long biblio- graphy (pp. 281–301) shows the vast range of printed material that has been used by Dyson to construct his narrative. In this review, we will mainly consider how he deals with population estimates before proper national censuses began around 1871.

When statistical data are not available, one has to use other methods to obtain quantitative results. Let us begin with the Indus Civilisation. If we can establish roughly the size of the urban population, we can estimate the total population by postulating roughly what percentage of the total population we can allow at the maximum to the urban population. If we assume that the urban population could then not have amounted to more than 10 per cent of the total, roughly the figure in the de-industrialised India of 1881, we can now go to archaeological sites and estimate the size of the used or built-up area of a town to deduce the size of population. Irfan Habib estimated the populations of Mohenjo Daro and Harappa at 85,000 and 65,000, respectively assuming a ratio of little above 400 persons per hectare of ‘occupied area’ as disclosed by archaeology. Assuming a total of 150,000 persons for the two cites he supposed the total urban population of the Indus basin to amount to 250,000. Assuming the rural population then to have been at least fifteen times the urban population in view of the primitive conditions then prevailing, he put the total population of the area of Indus Civilisation at about 4 million or nearly 6 persons per square kilometre (I. Habib, Indus Civilisation, New Delhi, 2002, pp. 22–24, 37). Dyson does not seem to take this estimate seriously, since on page 21, n, he refuses to recognise the rate of 400 persons per square kilometre as acceptable because ‘much of the city consisted of open places’, oblivious of the fact that one is considering only what is recognised by archaeologists as occupied or built-up area, from which open spaces are excluded. The figure of 400 per square kilometre of built-up space is not exceptionally high, since it implies an area (floor-space) of 5 × 5 metres per head (adult or child). The amount of 4 million for the Indus Civilisation is accepted by Dyson in a footnote (p. 6, f. n. 7), but he forgets it in his later discussions.

In fact, Dyson seems to remain afraid of what appear to him to be high figures for population before colonial times. Let us take the matter of population of India in Mughal times. W.H. Moreland after a very summary view of Abu’l Fazl’s area statistics put the population at the time of Akbar’s death (1605) at 100 million (India at the Death of Akbar, London, 1920, p. 22). Kingsley Davis, Population of India and Pakistan, Princeton, 1951, p. 24, by another impressionistic act, increased the estimated population to 125 million to cover areas supposedly left out by Moreland. In my detailed study, The Economy of the Mughal Empire, c. 1595: A Statistical Study, 2nd ed., New Delhi, 2015, pp. 405–16, I worked out two figures for the population of India in 1605, namely, (a) based on cultivated area, which yielded an estimated population of 136.3–149.9 million and (b) based on land-revenue, leading to an estimated 149.07 million. Dyson seems unduly influenced by Sumit Guha’s excessive underestimation of the population of eastern India (Bihar and Bengal), which then yields a population level for India, c. 1595, at 116 million only (p. 61). He also contests the possibility that urban population in Mughal India could be high as 15 per cent (pp. 62–65). While on page 62, Dyson refuses to take sides between Guha’s estimate and mine, in effect, he favours Guha’s, and follows the argument of Kingsley Davis, who rejected any idea of substantial growth of population in pre-colonial times and assigned the real period of population growth to India under colonial rule with all its presumed civilising benefits. In practically identical terms, Dyson assigns the main period of population growth to the period 1821–71 (Chapter 6, esp. p. 95).

What is significant, however, is that the supposed population growth stopped just when the regular censuses began, yielding a population rate of growth per annum of just 0.33 per cent (see table in Dyson, p. 125). Clearly, the presumed earlier growth in population was due to a steady improvement in enumeration rather than actual population growth. When, with regular censuses, the stage of enumeration improvement was over, the nominal high rate of population growth also disappeared. This was, indeed, accepted by Morris D. Morris, a fairly consistent defender of the Raj in his article ‘The Population of All-India, 1800–1951’, published in 1973 and included in Dyson’s bibliography.

This simple explanation is not enough for Dyson any more than it was for Kingsley Davis. In an entire chapter ‘Famines, Plague and Influenza, c. 1871 to c. 1921’, Dyson throws the responsibility upon natural disasters for the low rate of population growth, though these disasters were equally present in the previous period as well (see Irfan Habib, Indian Economy under Early British Rule, 1757–1857, pp. 105–7, for famines during 1813–57).

These criticisms notwithstanding, one is deeply grateful to Dyson for his Herculean effort. Whatever, his own conclusions (and preferences) he lays out all the relevant material before us, and it is for us, in many cases, to make our choice out of alternative interpretations.