ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Ecological Construction of an Epidemic in Colonial Bengal

Crop, Climate and Malaria

The decade of the 1870s was marked across Bengal by drastic rainfall variations, changing crop patterns and a devastating malarial epidemic that depopulated many villages. Though hampered by a paucity of data, this paper relying on contemporary records establishes the link between widespread incidence of malaria to not merely the declining water supply, but the increasing inaccessibility to existing water sources, thanks to the large-scale construction of embankments and the colonial reclamation of land.

Until the 18th century, argues Ladurie, when societies were still essentially agricultural and dominated by the chronic problem of food supply, there was an intimate link between history and climate.1 But unfortunately traditional societies have left us almost no sustained quantitative records of temperature and rainfall. Because of this lack of any firsthand written records, any discussion of meteorological fluctuations and their effects on the course of history has frequently been ill-founded. This situation is however now a thing of the past, and many of the little – known themes of European history have been readdressed in the light of climatological assumptions. It was in this spirit that Bruckner boldly explained the fall of the Roman empire by a deviation in the cyclone route and the drying-up of the Mediterranean region.

But this is not so with the history of India, and far less with Bengal. This paper is a humble attempt, a meteorologic-historical approach to explain the proliferation of malarial fever in Bengal during the colonial period. Hampered by the shortage of data, the study dealing with the question may quite often allude to collating events which had for various reasons caught the imagination of contemporaries.’Terrible’ drought, period of ‘dreadful’ aridity, ‘torrents’ of rain, ‘devastating’ flood, for instance, are subjective imagination, often heterogeneous and piecemeal. But since certain meteorological observations retained in collective memory seem to us to be incontestable, one can reconstruct the whole story on the basis of meteorological changes along with other changes associated with it.



The 19th century malaria epidemic in Bengal has generally been ascribed to the colonial interference with its physical topography.2 Nineteenth century British commercial and political penetration in Bengal and the subsequent creation of colonial infrastructure, it is argued, broke through the geographical barriers and spatial distance. Roads, railways, systems of labour migration, military recruitment and civilian administration enhanced internal mobility of people. It stirred up a hitherto endemic property of a disease and disseminated it in such a scale that the endemic property turned out to be an epidemic reality. The expansion of irrigation canals and the construction of railway embankment created favourable habitats for malaria-carrying mosquitoes, and no wonder Bengal experienced a long-spell of devastating epidemic fever.

These admittedly are some of the basic factors which lurked in the immediate proximity of an epidemic. But such a position ignores the agenda of long-term ecological changes generally associated with devastating depopulation. One can argue that when a population ceases to expand rapidly, it may be merely because local circumstances such as food supply are not specially favourable to expansion. But an actual loss of population, as was the case in Bengal, meant very much more than a falling rate of growth. It indicates to some definite change in the climate or other ecologial conditions vitally affecting the life of the people.

When, however, a change of climate or of local conditions results in a decline of population, it seems to be reflected in the flora and fauna of the locality. The best recognised change in relation to fauna was to be seen in the great increase of malaria. If one compares the area suffering from a decline of population with the area showing relative prevalence of malaria, one will find that there was a rough correspondence between the two.3 Depopulation was therefore generally associated with intense malaria. Bengal was no exception to this overall pattern. Evidences available indicate that in the area in which the population was declining, malaria had increased alarmingly during the preceding 60 years.4 This admitted increase of malaria can be explained by the assumption that there had been a corresponding increase in the number and distribution of anopheles mosquitoes and malaria parasites in the affected areas. But unfortunately contemporary literature is rather reticent on the subject. No one mentioned that mosquitoes were exceptionally prevalent in the affected tracts. Of course, since the mosquito theory of malaria was then unheard of, it is plausible, observers would have thought of mentioning any such exceptional prevalence even if they had noticed it. Nevertheless, the assumption is not logically improbable that there was a great increase of mosquitoes at any particular place or time. Obstruction to drainage and the conversion of running streams into dead rivers would no doubt have largely increased the number of suitable breeding places of anopheles. Besides, in all the rice-growing districts of Bengal, we are told, the paddy was grown right up to the doors of the village house.5 In such rice-growing districts where for several months in the year, the whole surface of the country was a swamp, there were breeding places already available for the mosquitoes.

Other important changes in the fauna of the areas undergoing depopulation were the notable decrease in the fish population, and the great increase in the prevalence of certain mammals, wild hog for example.6 Both these changes were important from an economic point of view. The reduction of fish meant a decrease in food supply with adequate protein content, and the increase in wild hog meant a greater damage to crops during periods of scarcity.

The most noticeable change related directly to precipitation and meteorological fluctuation was however the lapsing of some of the cultivated or inhabited land into jungle. It was of course confined entirely to the more elevated and drier tracts, for instance, high river banks, or certain village sites. It was particularly noticeable in places where population had declined. Besides, contrary to what is often stated, this growth of jungle was of a kind that preferred a dry to a damp habitat. It was not swamp vegetation that increased, but trees and shrubs whose roots required soil in which aeration occurred. This fact had a very important bearing on the nature of ecological changes that had occurred in the areas undergoing depopulation.

There is however little statistical information about changes in the proportion of cultivated and uncultivated land, respectively. We have of course a few general statements relating to particular areas. Bentley reported that prior to the epidemic fever, seven-eighth of the land in Burdwan was under plough, whereas the returns in 1922 showed only 47 per cent of the cultivable areas as being cropped. This latter figures compares very unfavourably with Dhaka, where during the same period 92 per cent of the cultivable areas was brought under plough.7 A progressive contraction of the cultivated area in Nadia too had been reported by successive collectors. A loss of fertility and consequent agricultural deterioration had been noted in many other districts of central and western Bengal. Reports from Faridpur show that it was impossible for landlords to enhance rent and to evict tenants when the fertility of the soil was declining and land was gradually going out of cultivation.8 Similar reports came in abundance from Hooghly and Howrah districts as well in 1873.9

Thus taking into consideration the broad fact that only 58 and 60 per cent of the cultivable areas were brought under cultivation in central and western Bengal, respectively, it would appear, in view of the remarkable decay of population that had taken place in the former areas, that some climatic changes had occurred which affected the flora and fauna of these areas to such an extent as to diminish very seriously the people’s food supply.



When one examines in detail the crops grown and recent changes in crop pattern, certain very significant facts are brought to light. First, let us consider the character of rice sown in different parts of the country. There were in fact three main varieties of rice – ‘aman’ or winter rice, ‘aus’ or autumn rice, and ‘boro’ or summer rice. The crops, of course, varied both in kind and season according to the fertility of the soil, the general meteorological conditions and the amount of water supply.10 Aus paddy grew on relatively high land and required least amount of water. Aman required abundant water and grew best along river edges and permanent swamps. Boro grew always in low and marshy lands. Boro paddy gave the most abundant yield, aman paddy the finest grain and aus paddy the smallest out-turn and coarsest rice.11 In spite of these facts, aus paddy was gradually replacing aman in certain areas.

The Gazetteer of Jessore district, for instance, stated that the area under aman rice was contracting due to deficient floods. A comparison of the agricultural statistics of 1920-21 with those of 1906 shows that in Jessore the normal area under aus had increased by 10 per cent, and in Nadia district by 50 per cent.12 The figures for individual thanas in Faridpur district indicate that the extra-ordinary variation in the growth of aus paddy in different parts of the district might have had some relation to water supply. ‘In the north of the district’, the Settlement Report states, ‘water for drinking and other purposes have become scarce’. The north of the district was malarious, and in the malarious northern thanas of Pangsa, Bhushna, Baliakandi, Goalkunda and Faridpur, the proportion of aus paddy grown was from 34 per cent to 46 per cent of the total, whereas in the non-malarious more southern thanas, it was very small, ranging from 6.3 per cent in Kotwalpara to 7.5 per cent, 7.9 per cent and 9.6 per cent in Madaripur, Gopalganj and Muksudpur, respectively.13 ‘The yield of aman rice grown in the south of Faridpur was very heavy’, recorded the Settlement Report, ‘but in the north of the district it is ordinarily much poorer and only becomes good when inundation comes early and are widespread’.

Though an almost similar phenomenon was noticeable in lower Bengal, there was however some additional features here. Before the advent of the British rule in Bengal, cotton and sugarcane were produced over a large tract of land. Except for certain parts of Midnapur and Bankura, almost every household in lower Bengal used to employ its members to cultivate some amount of cotton and sugarcane not for market but for its own domestic consumption. These two crops took almost a year to ripen for harvest, and required intensive irrigation for nearly eight to nine months. This meant that sugarcane and cotton cultivation demanded plenty of water and that large beels and ponds were exhausted of their water resources in the process. Fallen leaves in the ponds from the surrounding trees found no time to rot, and the proliferation of mosquitoes was therefore non-existent.14

The introduction of British cotton goods in Bengal and its penetration into remote village markets, a process largely facilitated by the expansion of railway network, and the establishment of sugar mills, changed the situation altogether. Cotton cultivators virtually lost interest in the production of cotton since cheap and durable imported cotton clothes were made available in plenty to the villagers. And introduction of sugar mills discouraged individual initiative for sugarcane cultivation meant for private consumption. These two factors thus combined together to remove or reduce the cultivation of the two traditional water-consuming crops.15

Instead of cotton and sugarcane, the cultivators now took to the production of potatoes. It required less water, and the village ponds as a result were seldom depleted, thereby providing ample scope for mosquito breeding.16 A good number of vernacular literary tracts of the late 19th and early 20th centuries refer to the existence of numerous irrigation ponds, at least 10 to 20 in each tiny hamlet in lower Bengal.17 The numbers seemed to increase with the size of village, sometimes 150 in case of larger ones. According to Census Reports, there were as many as 1,23,245 villages in Bengal in 1911.18 If one takes on an average 15 irrigation tanks per village, one can argue that no less than a million and a half vegetated and stinking ponds were created during the period under review.

The situation further worsened with the expansion of jute cultivation in lower Bengal. The manufacture of jute was then the most important industry in the province, and the export of raw and manufactured jute was her most valuable commercial asset. By 1925, 60 per cent of the jute grown in Bengal came from the relatively less malarious eastern districts, and the supply of Burdwan and the Presidency Division constituted barely 10 per cent.19 In terms of acreage, normal area under jute in eastern Bengal in 1925 was 13,73,300 acres, compared to 3,53,000 acres in the Burdwan and Presidency divisions.20 But available figures of 1908 show that cultivation of jute in lower Bengal had expanded to an area of 7,45,600 acres of land. In 1907, it was no less than 6,23,000 acres or almost double the figure of 1925.21

Jute was harvested over the three months of September, October and November, and these three months were notoriously malarious. Jute plants were cut and submerged in ponds or in beels, and kept there for a month to rot. When the plants became quite rotten, they were taken up from water, stocks were broken to reveal the fibre. They were, then washed, cleaned and dried in the sun. This process of ‘retting’ the stalks turned the ponds filthy and obnoxious, a veritable breeding place for mosquitoes. Thus expansion of jute cultivation brought in its trail the proliferation of malarial fever. Contemporary vernacular literary tracts more often than not lamented over the extension of jute cultivation and its associated evils.22



One important aspect of the malarial epidemic in Bengal was that more often than not it was preceded or followed by flood or famine, a phenomenon which again was closely linked to the fluctuation in rainfall. Peterson rightly observed in 1910 that the deltaic portion of the district of Burdwan was more liable to famine as the people depended almost entirely on the winter crop for their means of subsistence. Failure of this crop or a diminished out-turn could hardly be compensated for by the autumn rice crop which was cultivated on a comparatively small scale.23 During the period when epidemic fever was at its worst, the district of Burdwan experienced two devastating famines, one in 1866 and another in 1874. The early autumn rice crops of 1865 was a full one, but the winter rice crop did not probably average less than two-thirds of a full out-turn. By March 1866, distress began to make itself felt, and coarse rice which formerly sold at from Re 1-4 to Rs 2 per maund at that time of the year was selling at Rs 4-8.24 In the beginning of July, we are told, there was a sudden influx of paupers into the town of Burdwan, principally from the weaving towns of Chandrakona in Midnapur, Bishnupur in Bankura, and from Birbhum district. Outside the town the distress was not very serious, but the country on the opposite side of the river Damodar was flooded, and 3,065 houses were reportedly destroyed. The homeless multitudes left for the towns in search of food.25

Again, the year 1872 was unfavourable in Burdwan. The rains commenced very late, and when it did, it was not equally distributed throughout the district.26 As a result, the out-turn of aman rice crop was very small. Several plots of land, even at places entire villages, remained uncultivated. Only low lands, and those which had the advantage of irrigation, yielded a fair crop, the average out-turn of the district being about 50 per cent.27 Closely associated with the phenomenon of unfavourable harvests was an onslaught of epidemic fever which raged violently all over the districts of Burdwan, Birbhum and Midnapur.28

The rains in 1873 in Burdwan were not so greatly deficient in quantity as they were unseasonable in distribution. In the first six months of a year normal fall was 17 3/ 4 inches of which 13 1/5 inches fell in May and June. But in the first half of the year 1873, only 11 1/7 inches fell; of this only 7 1/4 inches fell in May and June. During July and August the normal fall was 23 1/2 inches. But in 1873, during these months 30 1/5 inches fell. Finally instead of a normal fall of 13 inches in September and October, there was only 4 3/4 inches in the former and only 3/4 of an inch in the latter month of 1873.29 Rainfall in Burdwan in 1873 was deficient when copious amount were required and excess when moderate moisture was required. The effects of such unseasonal rainfall on the harvest varied with the crops. The aus or the early rice crop escaped its fury except in places where it was damaged by floods consequent upon the abnormally heavy rains in July and August. The late crops however suffered severely. The year 1874 thus found the district prostrated from the effects of two successive years of markedly adverse harvest. Not only were the material resources of the people depleted, their physical capacity to bear a strain was also seriously impaired.



Putting aside for a moment the phenomena of increased malaria, the loss of population, changes occurring among flora and fauna and climatic changes bringing in deficient harvests, one can view the decay from a different standpoint. Perusing the various relevant records of the period, Bentley had concluded that ‘a reduction in water supply is the origin of the trouble’.30 Bentley’s observation can hardly be contradicted. For, first of all, there was the receding margin of cultivation, invariably associated with a deficient supply of water. It was always the drier and more elevated lands that went out of cultivation. Then, there was the increase of jungles largely confined to the relatively high banks of rivers and old village sites, showing that the type of vegetation was one which preferred a dry situation rather than a damp water-logged one. The tendency for aus paddy to displace the aman variety was also an indication of diminished moisture. The decreasing supply of fish also afforded further evidence of a lessened supply of water, as did the increase of wild hog and other animals which followed the spread of jungle.31 The only fact which appears to militate against this view was the very great increase of malaria which had taken place along with these other changes and which at first sight appears to preclude the possibility of a diminished supply of water being the cause of the trouble. But one has only to recall certain well known facts with regard to malaria in order to realise that this contradiction is only an apparent one.

It is well known that swampy low lying locality ceases to be unhealthy when temporarily submerged. In the olden days, flooding was often deliberately applied both in Italy and Holland for ameliorating malaria with most excellent results.32 But while the flooding of swamps had been resorted to for checking malaria, the partial removal of water from a marshy area had always been regarded as a dangerous exercise, calculated to produce a serious intensification of the disease. This belief is strengthened by the numerous instances in India in which the reclamation of swamps had led to severe outbreak of malaria.33 Bentley recorded the occurrence of such outbreaks in connection with the construction of the Colaba Causeway (1824-41), various reclamation schemes carried out in 1861-66, and in later times the construction of water works at Malabar Hills and Bhandarware Hill and of the Alexander Docks and Hughes Dry Dock.34 Three more well known instances of this kind were the case of the Grand Chartreuse Swamp in France, the low-lands of Briensten and de Wonner in Holland, and the Whittlesea Mere in England. Such outbreaks were formerly ascribed to miasma given off from the exposed bottom of the marsh. But the modern explanation is that a great increase of anopheles mosquitoes takes place when by partial drying up, a large expanse of water is converted into many small shallow pools.35

One can therefore see that all the important changes which coincided with the epidemic fever in Bengal were consistent with a reduction in the water supply of the affected areas. One can produce abundant records of complaints that the tracts concerned witnessed diminished water supply. In 1867, Reverend G Schurr of Kaspadanga in Nadia reported in a letter to the magistrate: ‘During the 20 years I have been in this district, I have observed a gradual diminution of water supply in old tanks and khals as well as in the rivers’.36 The following year Sutherland observed of the same district: ‘The district is becoming more and more arid; tanks and other reservoirs of water dry up even before the hot weather; tanks full to overflowing in the rains rapidly dry up – facts which prove that there has been of late no increase or excess of moisture in the pool’.37 As late as 1912, Bholanath Banerjee, executive engineer in charge of the special drainage division, Jessore, reported: ‘If we draw a line from Pultia on the Ichamati to Jhikergacha on the Kanadak, the country lying to the north of this line may be generally taken as suffering from a scarcity of good drinking water’. It was reported, Banerjee added, that tanks in those parts ‘do not hold water but dry up with the subsidence of the rivers’.38

One can produce many more evidences to explain the above phenomenon. The Nadia Fever Commission, for instance, stated that ‘observation in December of the water-level in the well showed it to be from 15 to 24 feet from the general soil surface level in the west and north-west of the district’.39 In the 1920s observation in Burdwan, Nadia, Jessore, Pabna, Murshidabad had shown very similar results, and had proved, contrary to the commonly accepted belief, that areas in which level of the subsoil water was especially low were far more malarious and unhealthy than those in which it approached within three to five feet of the ground level. In Burdwan, the worst affected district, the mean level of the subsoil water taken in 28 wells in different parts of the district was 26 feet in the dry weather and nine feet in the rains. By contrast, in the healthy parts of Howrah, Dhaka and Mymensingh it varied from three to five feet in the dry season and was level with the ground surface in the rains.40

True that rainfall fluctuates from year to year, but there is no reason to believe that the average amount precipitated had at that times undergone any marked diminution. On the contrary, meteorological evidence of as late as 1933-34 shows that it was heavier than normal.41 There is again plenty of evidence to show that embankments which had been constructed along the margin of the rivers and for the purpose of roads and railways had the effect of shutting out from the surface of the country a large amount of water which found access to it in earlier times. Thousands of miles of embankments which covered the country with a network extending in almost every direction had deprived the country of much moisture it enjoyed earlier.

It has often been suggested by scholars that these embankments had obstructed drainage which led to water-logging and the consequent malaria. This may now sound paradoxical. But a critical observation will show that the lands which were water-logged had actually been flooded before the bunding operations. When the flood subsided the flooded tracts turned into large lagoons. But now, due to embankments, flood ceased to occur, and instead of lagoons there existed only pools. So except for a very few places, the country within the influences of the embankments, instead of being wetter than earlier was actually more drier. In the light of the facts brought forward so far, the explanation is a simple one. It was mainly in the construction of embankments that one must look for the cause of increase of malaria in Bengal and the depopulation that accompanied the increase; but one should not lose sight of the associated phenomena such as climatic changes bringing in deficient harvests, changes occurring in flora and fauna, as well as changes in crop pattern all indicative of a over-all decline in the general health condition of the people of Bengal.



 1 Ladurie, Emmanuel Le Roy (1972): ‘History and Climate’ in Economy and Society in Early Modern Europe by Peter Burke (ed), London, p 134.
 2 Mitra, Raja Digambar (1873): The Epidemic Fever in Bengal, Calcutta. Klein, Ira (1972): ‘Malaria and Mortality in Bengal, 1840-1921’ in Indian Economic and Social History Review, Vol IX, No 2, June. Arnold, David (1989) (ed): Imperial Medicine and Indigenous Societies, Oxford. Kumar, Anil (1998): Medicine and the Raj: British Medical Policy in India, 1835-1911, New Delhi.
 3 Bentley, C A (1922): ‘Some Economic Aspects of Bengal Malaria’ in Indian Medical Gazette, Vol LVII, September, p 323.
 4 Krishak, Vol 21, Nos 6 and 7.
 5 Crawford, D G (1903): Hughli Medical Gazetteer, Calcutta, p 171.
 6 Bentley, C A, op cit.
 7 Ibid.
 8 Jack, J C (1916): Final Report on the Survey and Settlement Operations in the Faridpur District, 1904-1914, Calcutta, p 50.
 9 Report of R C Mukherjee, summarised by E H Pellow in a short report dated March 24, 1874, reprinted with Pellow’s report of 1878.
10 Loveday, A (1914): The History and Economics of Indian Famines, London, pp 3-4.
11 Imperial Gazetteer, Vol III, 1907, pp 5-7.
12 Bentley C A, op cit, p 324.
13 Ibid.
14 Mandal, Rajkrishna (1908): Malariar Karan O Pratikar, Calcutta, p 24.
15 Ibid, pp 24-25.
16 Ibid.
17 One may cite, for instance, Ray, Karali Charan, Bange Malaria, Basantapur, 1917, Bandyopadhyay, Haridhan, Banglar Shatru, Sodpur, 1924.
18 Census of India, 1911, Vol V, Part-II, p 2.
19 Bentley, C A (1925): Malaria and Agriculture in Bengal, Calcutta, p 57.
20 Ibid, p 59.
21 Mandal, Rajkrishna, op cit, p 27.
22 Ray, Karali Charan, op cit, p 19.
23 Peterson, J C K (1910): Bengal District Gazetteers, Burdwan, Calcutta, pp 99-100.
24 Ibid, p 100.
25 Ibid, p 101.
26 Macdonnell, A P (1874): Foodgrains Supply and Famine Relief in Bihar and Bengal, p 335.
27 Ibid.
28 Samanta, Arabinda: Anatomy of an Epidemic: The Malarial Fever in Colonial Bengal, In Press, Firma K L M, Calcutta.
29 Macdonnell, A P, op cit, pp 336-37.
30 Indian Medical Gazette, Vol LVII, 1922, p 325.
31 Ibid.
32 Celli M (1912): Malaria in Italy during 1910, Cited in the Proceedings of the General Malaria Committee held at Madras, November 18, Simla, 1913, p 119.
33 Covell, G (1928): Malaria in Bombay, 1928, Bombay, p 13.
34 Indian Medical Gazettee, Vol LVII, September 1922, p 323.
35 Jaramillo-Arango, Jaine (1950): The Conquest of Malaria, London, p 77.
36 Bentley, C A (1922): Indian Medical Gazette, Vol LVII, p 325.
37 ‘Report on Epidemic Fever’ by J Sutherland, No 51, dated April 13, 1868. Proceedings of the government of Bengal, General Department, Sanitation Branch, June 1868, p 15, West Bengal State Archives.
38 Bentley, C A, op cit, p 325.
39 Ibid.
40 Ibid.
41 Cited in Report on the Working of the Anti-Malaria Campaign in Rural Areas of Bengal with Quinine and Plasmachin, Calcutta, 1934, pp 7-8.

Dear reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top