habib studying colonial economy without perceiving colinialism modern asian studies critique of...

7/29/2019 Habib Studying Colonial Economy Without Perceiving Colinialism Modern Asian Studies Critique of Cambridge Eco

1/28

Studying a Colonial EconomyWithout Perceiving ColonialismAuthor(s): Irfan HabibReviewed work(s):Source: Modern Asian Studies, Vol. 19, No. 3, Special Issue: Papers Presented at the Conferenceon Indian Economic and Social History, Cambridge University, April 1984 (1985), pp. 355-381Published by: Cambridge University PressStable URL: http://www.jstor.org/stable/312444 .

Accessed: 29/08/2012 03:55

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access toModern

Asian Studies.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=cuphttp://www.jstor.org/stable/312444?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/312444?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=cup


2/28

Modern Asian Studies, 19, 3 (I985), pp. 355-38I. Printed in Great Britain.

Studyinga ColonialEconomy-WithoutPerceivingColonialismIRFAN HABIB

Aligarh Muslim UniversityFrom the size of India's population alone the economic history of Indiaconstitutes an important segment of the economic history of mankind.But with the middle of the eighteenth century, it assumed a further,special significance: subjugated by the first industrial nation of theworld, it offered the classic case of the colonial remoulding of apre-modern economy. Not surprisingly, the changing nature andconsequences of this process and all its surrounding conditions haveformed the constant theme of a long and continuing debate.To the editors of the CambridgeEconomicHistory of India, Vol. II(henceforth referred to as CEHI), the debate has apparently been anexercise in irrelevance. They refer condescendingly to 'the sturdyclassics-Dutt, Gadgil, Anstey, Naoroji'-their worth being not evensufficient to justify the editors' taking the trouble of placing them in acorrect order. Dutt and the others have been rendered 'out of date,' theeditors say, because 'over the previous decades'-'scholars haveenhanced our understanding of the historical experience of specificregions and communities and illuminated aspects of economic activitybut lightly touched upon previously'. As far as increase in information isconcerned, the editors are on safe ground enough. But going by theirdescription, the field for which new information has been gained is adifferent one from that with which Naoroji and Dutt were concerned. Inother words, the editors set out to dispense not only with the 'classics',but also with the questions raised there. They seem determined to readmodern Indian history without looking at colonialism.1This becomes obvious from the plan of the volume. Part I, comprisingmore than half the book, is entitled 'The Land and the People'. In thisPart we first have a chapter on India in mid-eighteenth century,

1 Characteristic of the editors' outlook is their avoidance in their preface of anyreference to colonialism or Britain's exploitative relationship with India. The closestthey come to it is in their allusion to literature on 'underdevelopment and developmentpolicies'.oo26-749X/85/o8o9-o204$05.oo ? I985 Cambridge University Press

355


3/28

whereafter the country is chopped up into four regions. The agrarianrelations of these regions are studied separately up to I947 in foursections; and then there are four separate studies of their 'economies', to1857 only. With a characteristic defiance of chronology (the regionsubjugated earlier by Britain should surely have come first) the study ineach chapter begins with the northwestern and central regions. Theseregional studies are followed by three chapters on National Income,Population and Occupational Structure, all the three basically con-cerned with the post-I857 period. From these we are led to Part II,styled 'The Beginning of the Modern Economy'. This begins with achapter on the growth of large-scale industry. There is next a singlechapter on Irrigation and Railways (a pairing for which it is difficult tofind a reason, unless it lies in both of these being treated as 'public-works'). Chapters follow on Foreign Trade, Price Movements and theFiscal System. Part III consists of two chapters on 'Post-IndependenceDevelopments', one on India, the other on Pakistan.I beg to be forgiven for setting out the list of contents like this, but thisis necessary to show that such a plan of chapters practically precludes awhole series of the most crucial questions from being taken up. Wherecan one study the Tribute or Drain of Wealth, let us say in the'mercantilist' phase, I757-1813? Or the process of 'de-industrializa-tion'? The editors say disarmingly that there ought to have been achapter on handicrafts, though curiously enough they allow ProfessorMorris, who writes on large-scale industry, to have a full say onsmall-scale industry as well.2 The editors themselves hasten to add thatthey wished there were a separate chapter on the theme simply becauseof'the continuing importance of handicrafts', lest some one might thinkthat they have had 'de-industrialization' in mind. There is similarly nospace provided to discuss the process of the late nineteenth-centurycommercialization of agriculture; and the question of growth of landlesslabour is suitably tucked away among the 'regions'. The whole issue ofthe 'Imperialism of Free Trade', which since the Gallagher-Robinsonessay should have gained a place in 'current literature',3 eludes both theeditors and the contributors.If the straitjacket of the editors' scheme was not sufficient, thecontributors themselves, with a few exceptions, have succeded in settinga course which suggests that to them economic theory came to a close

2 CEHI, II, 668-76.3John Gallagher and Ronald Robinson, 'The Imperialismof Free Trade', EconomicHistoryReview, 2nd series, VI (1953), I- 5, and the large body of subsequent writing onthe theme.

356 IRFAN HABIB


4/28

STUDYING A COLONIAL ECONOMYwith Free Trade and Comparative Advantage. Explicitly, and mostoften implicitly, they are out to demolish the nationalist criticisms ofexploitation and impoverishment under colonial rule. It is worth somereflection that there is little in this volume at which official annoyancecould have been aroused even before I9I4.In this paper, I take up what I think have been important aspects ofthe colonial impact on Indian economy (up to the beginning of thetwentieth century) that are either overlooked in the CEHI or receivetreatment on which one may have considerable reservations. It shouldbe understood that I do not pretend to be more knowledgeable than theeditors and authors of the CEHI, and that I note and raise particularquestions simply because the CEHI's way of dealing (or not dealing)with them has proved so disappointing to me.

I. The Tribute, 1757-x8I3In so far as the East India Company and its servants saw their conquestsin India as commercial acquisitions, the seized treasuries and tax-income appeared to them nothing but gross profits. The 'PlasseyPlunder', where enormous individual fortunes were made, was followedby a continuous extraction of wealth, through taxation, monopoly andcorruption, generating a stream of exports from India, without anycorresponding imports. Those who decry 'scholarship-by-quotation'and crave for statistical data have here quantitative evidence of greatvalue to estimate the 'drain' in terms of both prime costs in India andsale proceeds in overseas markets. They can compare the size of the totalwealth transferred from India with the estimated gross national productof the Company's dominions (for which they can use estimates ofproduction by Grant and Shore),4 and with estimates of British nationalincome. With Furber's work already decades old,5 and the convenientpresentation of British historical statistics by Mitchell and Deane,6 somuch quantitative evidence could have been laid out, and the groundcleared for enquiring as to how the transfer affected India and benefited

4James Grant in The Fifth Report, &c., ed. W. K. Firminger (Calcutta, 1917), II, p.276, and Shore in ibid., pp. 27-8.5 Holden Furber, John Companyat Work (Cambridge, Mass., 195I), esp. pp. I 12-I6.One searches in vain for any reference to this work in CEHI.6 B. R. Mitchell and Phyllis Deane, Abstractof British Historical Statistics (Cambridge,I962), esp. pp. 309ff. (col. for trade with Asia). The statistics have been used forcalculating the drain by Sayera I. Habib, Proceedings f theIndianHistoryCongress,AligarhSession, 1975, section IV, pp. xxii-xxiv.

357


5/28

Britain. This was, after all, the main economic aspect of the colonialrelationship in its initial phase; and the drive for revenue which it setafoot is basic to any understanding of the Permanent Settlement. Yet theCEHI has no room for all this; and the only place where an allusion to itoccurs is in K. N. Chaudhuri's chapter on Foreign Trade. He begins bytelling us that... the far more important characteristic of the half a century following therevolutionof 1757 [a euphemismforPlassey]was the fact that the Indian tradestill continued to flow along the traditional hannels and its composition wasbased on an exchangef fine textiles, foodstuffs and other raw materials forpreciousmetals and certain manufacturedproducts.7

The description is surprisingly inaccurate: the end and purpose of the'revolution of I757' was to disrupt altogether the 'traditional channels'and to enable Britain to obtain Indian goods without any export oftreasure in return.8 Indeed, Chaudhuri himself, while speaking ofBengal before the Company seized its revenues, shows how the influx ofbullion had led to 'an expansion of the economy of Bengal' and how adifferent situation prevailed when this stream dried up after 1765.9 Buthe holds back from drawing a converse picture of what must now havehappened to the commerce and crafts of Bengal when it ceased receivingany recompense for its exports. His entire assessment of the consequencesof the new phase suddenly narrows down to two rather secondaryphenomena: the inflationary pressure of demand for textiles and themonetary disturbance created by export of Bengal silver to China.10The effects on the crafts and internal trade of Bengal that the author ofthe Siyaru-l Mutdkhir7n wrote about and Cornwallis referred to injustification of his scheme of Permanent Settlement1" are all silently

7 CEHI, II, 806 (emphasisours).8 See what S. Bhattacharyasays,CEHI,II, 289. FromChaudhuri'sown Table 10.2 Bon p. 8I9 it can be seen that the Company's export of silver to India ceased after1757-58.9 CEHI,II, 8I4. One recallshow in his TradingWorld fAsia and heEnglishEastIndiaCompany,660-1760(Cambridge, 1978), p. 462, Chaudhuri attributed 'industrializing'qualitiesto the influxofbullion underthe aegisof the Dutch and EnglishCompanies;the'coastal provincesof India' were turned into 'majorindustrialregions'.10Even here Chaudhuri'sargumentneedsqualification.Since the EnglishCompanypurchasedcotton textiles out of the revenues of Bengal, Bihar and Orissa,it did not putmorecurrencyinto circulation; t only diverted it. Thus the 'inflation' wasnothing but arise n pricesofgoodsin which it 'invested';ofothergoodsno longerin demand thepricesshould have fallen. Moreover, export of silver to China should have had a deflationaryeffect.1l Ghulam Husain Tabataba , Siyaru-lMutakhirzn,ub. Nawal Kishore, II, 836-7,840-I; Cornwallisin FifthReport, d. Firminger, II, 542.

358 IRFAN HABIB


6/28

STUDYING A COLONIAL ECONOMYpassed over, as if these are not even hypotheses that should be tested. S.Bhattacharya does cite Steuart for the statement that Bengal's exporttrade with countries other than Britain suffered as a result of the EnglishCompany's claims on export goods;12 but, unfortunately, he too quicklypasses on to other matters, and the implications of this enormouscommercial diversion for both Bengal and the regions previously tradingwith it are left unexplored.

II. 'De-industrialization'The Charter Act of 1813, in so far as it dismantled the East IndiaCompany's monopoly of Indian trade, opened the country to theindustrial products of England, notably the textiles of Lancashire. Areader of writers as diverse as Marx and R. C. Dutt can see how thisadded a new and major element in the economic disruption of India.13Previously, the 'drain' of revenues by diverting a large portion to exportgoods had destroyed such avenues of employment as were sustained bythe demand of the deposed ruling classes and their dependants. After1800 the textile exports, the mainstay of the eighteenth-centurytribute-realization, could not withstand the competition of Englishfactory-produced cottons in the world-market. After 1813 Lancashireinvaded India as well. Tribute-realization could now only take place inthe form of export of raw materials and agricultural products, like indigoand raw cotton, but the ultimate answer, found by the second quarter ofthe nineteenth century, lay in opium. The entire process constitutes partof what has come to be called 'de-industrialization'.

This 'received' view obtains partial support here and there from someindividual contributions in the CEHI. But the general tone of the work isin sharp opposition to it. The spirit is manifest in Tom Kessinger, whenhe discounts the critical view held by Asiya Siddiqi and C. A. Baylytowards the tribute-induced 'growth' before 1857. Kessinger assures usthat 'all of the evidence points to expanded employment in agriculture,processing and transportation in a period of little or very gradualpopulation growth'.14 The evidence that Kessinger himself presents isnot of much weight, and, in respect of towns, at least, his pronounce-ments must be treated with the greatest caution, as we shall see. We may12CEHI, II, 289.13 Marx's articles bearing on India are best collected in Karl Marx, On ColonialismandModernization,ed. Shlomo Avineri (Anchor Books, New York, 1969). I have summarizedhis views in The Marxist, I(I) (Delhi, i983), pp. 16-33.14 CEHI, II, 269 (emphasis ours).

359


7/28

also note here the resort to a useful device, viz., speculative demography.Kingsley Davis postulates a very rapid population growth for thenineteenth century by assuming conditions of prosperity brought aboutby that noble phenomenon, 'the European influence',14a and now hiscompatriot, Professor Kessinger, seeks to establish the very sameeconomic prosperity by assuming a very slow population growth.These comments are made in passing. The main theoretician against'de-industrialization' is Morris D. Morris, and it is his propositions thatneed to be scrutinized the most closely. 5 On this theme we have alreadythe benefit of reading his two previous essays;16 but it will be best toconfine attention to what he now says in the CEHI.He begins by deprecating the emphasis usually placed on the textileindustry as the index of the fortunes of Indian handicrafts: 'Despite theimportance of the enormously varied handicraft activities and theconsiderable-sometimes overwhelming-detail that is available aboutthem, the only one to which scholars have paid much attention is thetextile industry'.17 But Amya Kumar Bagchi met this objection longago, through a notable paper, 'De-industrialization in Gangetic Bihar,

I809-190I', in which he compared the data on craft employment incertain Bihar districts given in Buchanan's surveys and the figures of theI90I Census. Bagchi found that the 'industrial' segment of thepopulation in these districts contracted from about I8.6% of the total toabout 8.5% in the course of the ninety years following 1809.18 It isdifficult to understand Morris's silence over this paper, since hisfellow-contributor, Professor S. Bhattacharya, duly notices it andaccords it much importance.19

Coming to the textile industry, to which Morris himself devotes hismain attention, he argues that the imports of Lancashire goods initiallyaffected the spinners only, and not the weavers, since imported yarnenabled weavers to compete with the imported piece-goods. He ignoreshere the pertinent criticism levelled by Meghnad Desai, himself aneditor of the CEHI, that even after shifting to cheaper British yarn,Indian weavers would have had to increase their productivity by 43%or accept a corresponding diminution of their income, between I81 8-2 Iand I829-31, in order to compete with British cloth.20 Morris also14a Populationof India and Pakistan (Princeton, 1951), pp. 25-6.15 CEHI, II, 668-76.16 IESHR, V (I), I-I6, and V (4), 319-88. 17 CEHI, II, 668.18 Essays in Honourof Prof. S. C. Sarkar(People's Publishing House, New Delhi, 1976),pp. 439-522. A subsequent controversy between him and Marika Vicziany in IESHR,XVI, 2 (April-June, I979), 107-62, enabled Bagchi further to defend his evidence.19 CEHI, II, 290-I. 20 IESHR, VIII (4), 354.

360 IRFAN HABIB


8/28

STUDYING A COLONIAL ECONOMYadmits that contemporary witnesses spoke of the distress of the weavers;but this was because, unlike Morris, they did not know that the distresswas really being 'caused by weather instabilities which periodicallycaused crop failure, by declines in agricultural incomes, and by steepfalls in local demand for cloth.'21 Such ignorance on the part ofcontemporaries should surely excite wonder. It is true, however, that theinitial assault of the powerloom was on the Indian weaver manufactur-ing for export, as can be seen from William Bentinck's minute of 30 MayI829.22 Yet in I837 Crawfurd (a defender of the Lancashire imports aswell as the opium trade) estimated that English manufacturers had bythen replaced about 6% of the Indian production of cloth (not yarn) forthe domestic market, besides eliminating Indian textile exports.23 Thedestruction of exports alone must have hurt a large number of weavers:At the close of the eighteenth century 40,000 looms worked for theCompany in the Madras Presidency; and these amounted to no less thanone-sixth of the number of looms working in that Presidency about sixtyyears later.24

Though Morris overlooks these pre-i835 misfortunes of the weaver,he concedes that between I 835 and 1870, 'we would expect to encounterthe great competitive squeeze on handloom weavers'.25 Apparently,this modifies his previous rejection of such a squeeze even whenconfronted with a decrease in the price of British woven cloth.26 But theeffect on the weavers is still doubted by Morris though for a new reasonaltogether, namely, the lack of'public attention' paid to it 'during theseyears'.27 With easy neglect of chronology, he places Marx among thepre-I835 critics of Lancashire, while in actual fact Marx first referred tothe sufferings of the Indian weavers only in 1847; he thereafter returnedto the theme repeatedly in 1853, and the subsequent years of thatdecade; and references to it appear in Capital, Vol. I, published inI867.28 He thus incessantly called 'public attention' to the matter21CEHI, II, 669.22 Report of Select Committee on East India Company's Affairs, I831-2, BritishParliamentaryapers I.U.P.): Colonies,East India, 5, GeneralAppendix, pp. 275-6. It isespecially worth referringto, because Morris and a 'distinguishedBentinck scholar'could not tracea quotation given by Marx from Bentinck(IESHR,V (4), 383 n.). HereBentincksays the same thing in the same vigorouslanguage.23 Economic evelopmentf India underTheEast IndiaCompany,d. K. N. Chaudhuri(Cambridge, I971), pp. 239-42.24 This is calculated on the basis of data provided by Dharma Kumar in CEHI, II, 369and n. 25 Ibid., 669.26 Bipan Chandra in IESHR, V (I), 55-7; Morris's reply, IESHR, 380 n.27 CEHI, II, 669.28 The I847 reference is in ThePoverty f Philosophy,nglish tr. (Foreign LanguagesPublishing House, Moscow, n.d.), p. I 13; and the 1853 statements in articles in New YorkDaily Tribune,for which see Avineri (see note I3).

36I


9/28

throughout the period defined by Morris. If there was less open officialconcern with the sufferings of the weavers, this was undoubtedly due tothe fact that with the export of opium to China the problem oftribute-realization had been solved; and there was little ground anylonger for that anxiety on this score which had been manifested byBentinck while writing in I829 on the destruction of Indian textileexports.29 Besides, Free Trade was now an article of faith in Britain, andofficials could hardly have expected to get a hearing if they contested itswisdom. Thus there is sufficient reason for any official reticence whichmight now have developed about the fate of the Indian weaver: it is noproof that what the officials were reticent about did not exist.30

Morris goes on to restate his dictum about the 'shift to the right of thedemand curve' for cloth,31 but he does so now in more measured terms.The dramatic decline in cloth prices... should have stimulated a substantialrise in cloth demand. We do not know whether demand elasticity andpopulation growth after I835-40 increased the market rapidly enough toenable Indian producersto retain their absolute shareof the market,but thesefactorscertainly seem to have cushioned some of the impact.32

Let us put the matter in simple terms: If the Indians spent the sameamount in value per capita on cloth as they did before the Lancashireimports, then, at lower prices prevailing now, they would be able toconsume larger quantities of cloth. Ellison thus erred in supposing thatthe Indian consumption per capita had all along remained the same (setby him at 2.5 lbs) as it was in I880-8I (when British imports hadcaptured much of the Indian market); fifty years earlier the consump-tion should have been much smaller. Thus if we take Ellison's table,33and assume a per capita consumption of cloth in 1831-35 at 1 5 lbs only,and the population then at 222 million (as estimated for 1833 byMorris,34 as against I50 million accepted by Ellison), the per capitadomestic production would have fallen only from 1.44 lbs (and not 2.4lbs) in I831-35 to a little above I.oo lb. in I880-8I. If the per capitaconsumption in 1831-35 had been smaller than even 1.5 lbs thedomestic production would have been still smaller at that time. And yetthat domestic production must have accounted for about 97% of thetotal value of cloth consumed per capita in early I83os, while the29 Report of the Select Committee (see note 22), 275-6.30 In New YorkDaily Tribune,2 June i853, Marx said of a speech by Bright that his'pictureof India ruinedby the fiscalexertionsof the Companyand Governmentdid not,of course, receive the supplement of India ruined by Manchester and Free Trade.'(Avineri, p. 79). 31 IESHR, V (i), 9. 32 CEHI, II, 669.33 Thomas Ellison, The CottonTradeof GreatBritain(orig. published, I886; London,1968), pp. 62-3. 34 IESHR, XI (2-3), 3 I1.

362 IRFAN HABIB


10/28

STUDYING A COLONIAL ECONOMYdomestic production of I880-8I obtained for only 41.6% of the sameamount assuming that the value of cloth consumed per capita hadremained constant. Thus the spinners and weavers together would havehad not only to face a contraction of production, but also to suffer a verylarge fall in real earnings per unit of product.The situation would have been different only if there was a shift 'to theright of the demand curve' and if the total value laid out on per capitacloth consumption had increased. Of this there is not the slightest proof.The increase in cloth consumption supposedly due to moral admoni-tions of missionaries,35 and changes in fashion, can all be accommo-dated within the expansion of cloth consumption owing to the loweringof prices along the same old demand curve. Indeed, one maytheoretically argue that if cloth-prices fell, wages might also fall andrents increase, to cover the reduction in the costs of subsistence: in thatcase, the demand curve might well have shifted to the left!It is worth noting that some of the specific evidence presented by twoother contributors to the CEHI does not accord with Morris's line ofinterpretation. Professor Dharma Kumar finds for Southern India that'the unit value of cloth output certainly declined sharply in manydistricts and probably also the incomes of certain classes of weavers.'36She is less sure about a decline in the number of looms, but there is novery plausible ground for rejecting Dodwell's estimate of 4oo,ooo loomsin the Madras Presidency in 800, when it is rather reasonably workedout on the basis of the number of looms working for the Company,assumed to be a tenth of the total. About 1857 there were 225,000 loomsin that Presidency, and there was a noticeable shift to inferior cloth.37For Western India, Divekar's finding is the same as that of DharmaKumar: the weavers' real earnings fell 'considerably' between I82os andI840S. He also finds a decline in the number of looms though he givesnumbers for only one place in the Bijapur district.38A more definitive result comes from Amalendu Guha's calculations.He made an attempt to estimate handloom production by estimatingraw-cotton production, reduced by the volume of exports and consump-tion in weaving factories. He found that the net availability of cotton

35 CEHI, II, 369. 36 Ibid., 370. 37 Ibid., 369-7o(and 369 n).38 Ibid., 349. Unluckily, in his survey of Eastern India S. Bhattacharya does not givedata for either real earnings of weavers or numbers of looms, though he says that weaving'showed surprising survival capacity' (ibid., 292). Kessinger does not seem to have cometo grips with the question in his survey of Northern India, though he does give interestingdata on the disappearance of the supply of piece-goods from the Upper Provinces toCalcutta between I812- 3, when they composed 41% of the total trade, and I835-36,when their share had dwindled to 2% (ibid., 253-4).

363


11/28

yarn (including imported and machine-spun yarn) for handloomindustry declined from 419 million lbs in I850 to 240 million lbs in I870and 184 or 221 million lbs in I9oo.39In the twentieth century there happens to have come about a stabilityand some revival of the handloom; and much of Morris's owntheorization seems to be a projection of this phenomenon back into thenineteenth century.40 This apparent reversal in the fortunes of thehandloom is noticed by Gadgil, who says reasonably enough that 'at acertain stage' in the decline of handicrafts under the pressure of modernindustry 'a point of equilibrium was reached in the competition betweenthe two industries'; and that 'this point was [only] reached towards theend of the last century'.41 Bagchi, too, has examined the reasons behindthis revival of handloom, and he notes the lag in the development of theweaving departments in Indian factories, the application of fly-shuttleto the loom, and, not the least, the Swadeshi movement.42 Atwentieth-century recovery (ultimately to be a passing phase) owing tothese and other factors, thus cannot by itself persuade us to disbelieve inthe eclipse of the handloom during the preceding century.

III. De-urbanizationAs the Indian ruling classes were deposed and the revenues which theyappropriated were transferred to the Company, the cities and townswhich depended on the supply of articles of consumption and services tothe older ruling classes and their households, courtiers, and retainersrapidly declined. The assault of Free Trade after 1813 devastated

39Paper (cyclostyled), presented at a seminar on the Transformation from Medievalto Colonial Economy, Aligarh, 1972. Amalendu Guha's conclusion can also besupported from the figures of India's raw cotton exports to Britain. From 145 million lbsin 1855 these climbed to 443 million lbs in I872, constituting I6% of total British rawcotton imports in i855, and 31 % in 1872. (Peter Harnetty, Imperialismand Free Trade;Lancashireand India in theMid-igth Century Manchester, 1972), p. 49). Unless the Indianper capita cotton production also increased on such a dramatically high scale, the percapita availability of cotton within the country must have undergone an enormousdecline. The force of M. B. McAlpin's studies has been to deny such an increase in percapita cotton acreage during this period (JEH, XXXIV, 1974, pp. 662-84; and IESHR,XII (I), pp. 43-60).40 See IESHR, V (4), 378, where Morris takes as his starting point the existence of3. Imillion handlooms in India and Pakistan in 1950-51.

41 D. R. Gadgil, The IndustrialEvolution of India in Recent Times, i86o-i939 (5th edn,Delhi, 197 ), p. i80.42 Amiya K. Bagchi, Private Investment n India, Igoo-ig39 (Cambridge, 1972), pp.220-8.

364 IRFAN HABIB


12/28

STUDYING A COLONIAL ECONOMYcentres of handicrafts, notably textiles, and a fresh process of urbandecay began. Gadgil summed up the argument in his cautious mannerand postulated a 'slight' relative fall in the urban population in thecourse of the nineteenth century.43One would have thought that this was a fairly important matter to begone into at some length in a I,o74-page textbook of modern Indianeconomic history. L. and P. Visaria in their chapter on Populationbriefly refer to Gadgil's conclusion, but the only comment they makethemselves is that 'the level of urbanization [in India] has always beenvery low'.44 The Visarias' implicit scepticism about a nineteenth-cen-tury decline in the relative size of the urban population is shared by S.Bhattacharya, who, dealing with Eastern India in the nineteenthcentury, puts a question-mark after 'De-urbanization' in a section-heading, and also pursues in the text the indefinite course that theheading presages.45Professor Kessinger's heading 'Growth of Towns' for the same periodis firm, and his conclusions absolutely definite. In the British territories'the expansion of economic activity' brought a 'period of rapid growthto the cities and towns of the Ganges valley.' In the native states thesituation was at variance with this: 'None of the court cities, not evenLucknow, show the kind of dynamic growth related to commercialcentres in British territories. Awadh in fact had a relatively smallnumber of towns in its territories at the time of annexation in I854(sic)'.46

Kessinger's specific statements about Lucknow and Awadh need to bechecked. First, Lucknow: there is no doubt as to an immense increase inits population under the Awadh kingdom. In 1799 its population wasestimated at half a million;47 and in 1858 at a million, so that it shouldthen have been larger than Calcutta.48 Even Kessinger admits thatLucknow became 'probably North India's largest city'.49 He offers noargument why such growth is not 'dynamic', while the much moremodest increase in the size of some towns in British territories deservessuch characterization. More; the moment Lucknow passed underBritish control with the annexation of Awadh in 1856, a steady decline

43 Gadgil, Industrial Evolutionof India, pp. I44ff.44 CEHI, II, 5 19-20, But see Gadgil, IndustrialEvolutionof India, pp. 42-3.45 CEHI, II, 275-7. 46 Ibid., 265-6.47 Tenant, Indian Recreations, II, 404, quoted in H. R. Nevill, Lucknow: a Gazetteer(Lucknow, 1922), p. 149. In I824, Heber 'guessed' that the population was 300,000 (R.Heber, Narrativeof a Journeythroughouthe UpperProvincesof India (London, I828), p. 90).48 William Howard Russell, My IndianDiary, ed. Michael Edwardes (London, 1957),P- 59. 49 CEHI, II, 266.

365


13/28

IRFAN HABIBbegan. The depopulation must have been heavy after the British seizureof Lucknow during the rebellion of I857-58;50 but the declinecontinued till 19I I, as the successive censuses show:1869 284,779

I88I 261,3031891 273,028190 264,049I9 1 259,798

Thus Lucknow would seem to have lost all its dynamism once it passedunder British rule.As for the supposedly small number of towns in Awadh beforeannexation, it is not clear from where Kessinger derives his information.One major detailed survey of the southern districts of Awadh by Butterin 1838 gives estimates of population of a large number of towns; thesemay be compared with the results of censuses beginning with 1869/1872.For this comparison, I have taken towns to which Butter assigns apopulation of 5,ooo or more,51 and then located them in the later Britishdistricts. From the modern censuses, I have taken all towns with

populations of 4,500 or more (the lower size for the census towns is tooffset any rounding in Butter's estimates). The results are tabulated inTable I.In these five districts of Awadh, then, the number of towns containingestimated populations of 5,000 and above was 43 in 1838; but the 191Census counted no more than 15 such towns (even including those of4,500 to below 5,000). The total population of the towns stood at anestimated 374,000 to 385,000 in 1838, and at only I86,483 in 1911.Kessinger's statement that Awadh contained only a small number oftowns before annexation is thus totally at variance with the quantitativeevidence we have access to. Even if we allow for a large margin of error inButter's estimates, the general tendency is still clear enough: towns inAwadh declined sharply in total population after the territory becamesubject to the benefits of direct British rule.S. Bhattacharya's hesitation in accepting a decline of urban popula-tion in Eastern India is also not easy to understand, since he has hadbefore him Durgaprasad Bhattacharya's detailed work on pre-censuspopulation estimates of Eastern India, which contains a very revealing50J. Nevill, Lucknow Dist. Gazetteer,63.51 Donald Butter, Outlinesof the Topography ndStatisticsof theSouthernDistricts of Oud'h,1839, reprint (photo reprod.), ed. Safi Ahmad (Delhi, 1982), pp. I 4-43; also pp.ioo-i6.

366


14/28

STUDYING A COLONIAL ECONOMYTABLE I

Number ofTowns of 5,000 Total(in Censuses, of PopulationDistrict Year 4,500) and above of the TownsUnao I838 9 103,000 to

I 14,000I869/72 5 37,842I88I 5 38,467i911 5 37,809

Rae Bareli1 1838 13 96,000I869/72 4 29,218i88i 4 38,379I911 4 40,545

Sultanpur I838 5 39,000I869/72 I 5,708I88i I 9,374I911 I 9,519Pratabgarh2 1838 II 102,0001869/72 I 6,240I88I I 9,756

I911 2 I6,04IFyzabad3 1838 5 I34,000I869/72 3 55,635i88I 6 1 ,773i911 3 82,5691The i88i population for Salon not being available, it has beenassumed to be the mean of the I869/I872 and I891 figures(5,699).2 Bela and Pratabgarh counted as one town.3Ayodhya and Faizabad (Fyzabad) counted as one town.

table comparing estimates for individual towns during 181-30 and thepopulation counted in I872.52 The trend towards decline is quiteunmistakable. The only city which shows a large increase is Calcutta,growing from 179,917 in I821 to 428,328 in I872. But this growth wasmore than counterbalanced by the decline registered in the populationsof almost all the other major cities: Dacca declined from an estimated200,000 in I800 to 68,595 in 1872; Murshidabad from I65,ooo in I815 to46,182; and Patna from 312,000 in 1811-12 to 158,900.53 S. Bhatta-

52 D. Bhattacharya, Reporton the PopulationEstimates of India, vol. III, 81 I1-20, pt A.Eastern Region (Censusof India I96I), pp. xvii-xviii.53 I have checked these figures with the volumes of Hunter's Statistical AccountofBengal; they diverge slightly, in the case of the I872 Census, from the figures given by D.Bhattacharya, op. cit.

367


15/28

charya suggests that the earlier estimates tended to be inflated becauseof the large area the Indian towns occupied.54 But the estimates we haveused were based on house- or hut-counts, such as the one for Patna madeby Buchanan in 181 I-12, and thus should not have been influenced bythe extent of land falling within the city limits.55It is not my intention to supply further demographic informationwhich should have appeared in CEHI. What I have offered is only byway of illustration: the case for de-urbanization in the nineteenthcentury has a much stronger basis in descriptions and statistics than theCEHI editors and contributors have allowed for.

IV. Per capita IncomeWhile the CEHI may be parsimonious in the space it devotes to matterssuch as the tribute, handicrafts, or urban populations, one cannot haveany quarrel with it on the generosity with which the estimation ofNational Income is treated, in full 87 pages.56 Heston's industry insetting out the detailed calculations is commendable; and one wishes onecould as forthrightly commend the calculations themselves.

Heston's conclusion which he states close to the beginning of hischapter is that the per capita income rose by at least 35 % between I86oand I920, implying an increase in real national income by over 700/.57One basic premise on which Heston builds his series to achieve thisresult, is a firm assumption that crop-yields have remained absolutelyconstant, an assumption which sets him against all the previous officialcrop-estimators, and against Blyn, Sivasubramonian and other statisti-cians, who have accepted long-term declines in crop-yields per acre as aplausible phenomenon.Heston says frankly enough that if we were to accept official yields, theincrease of output of principal crops per capita between 90 I and 1946would rise by a meagre 7% as against 23%, which he gets by taking the

crop-yields to be constant. He had elsewhere argued, especially withreference to the Bombay Presidency, that official estimates of crop-yieldstended to be inaccurate before the crop-cutting methods came to beemployed in early 1950s.58 But Ashok V. Desai took issue with him over54 CEHI, II, 278, quoting Rennel.55 Francis Buchanan, An Accountof the Districts of Bihar and Patna in 1811-1812 (ThePatna-Gaya Report) (Patna, n.d.), p. 6I. Buchanan fixed a ratio of 6 persons to a houseon the basis of trial enumerations; the houses were estimated to number 52,000.56 CEHI, II, 376-462. 57 CEHI, II, 379.58 IESHR, X, 4 (December I973), pp. 303-32; XV, 2 (April-June 1978), pp. 173-86.

368 IRFAN HABIB


16/28

STUDYING A COLONIAL ECONOMYthis wholesale rejection of the official data.59 In the CEHI volume itselfMcAlpin uses these very data to establish association between price-movements and variations in output of various crops.60Even if the official yields are sometimes suspect, the adoption ofconstant yields is a totally arbitrary proceeding. It is not to be supposedthat there should have been an upward movement in yields fromextension of irrigation and shift to more profitable crops. This is becausethe yields for irrigated and 'dry' land are separately given in thestatistics, and the acreages of the principal crops are also separatelyspecified. Thus an upward movement in total output per acre could welloccur even if there was a decline in the yields.

Heston, on the other hand, insists that these yields themselves werebeing pressed upwards through a number of factors. The mostimportant was 'the more extensive use of land due to more workers peracre'.61 But this does not agree with Michelle McAlpin's finding that thecultivated acreage per capita did not fall between the I86os and 1901 62The other factors which Heston specifies, viz. 'increased use of improvedseeds, implements and chemical fertilizers, improved dry-farmingmethods and rotations' can also hardly have had much significance innineteenth-century agriculture. During that period, Heston's own list ofcontrary factors was all the more important: 'the general expansion intolower quality lands, the substitution of better quality foodgrain lands forcash-crops and the decline in natural fertilizers, particularly oilseedcakes that were exported and cowdung which was increasingly used asfuel as deforestation accelerated.'63In the light of these factors, it is even less possible than otherwise toaccept a projection of constant yields from I952-55 to as far back asI875, and to give credence to the increase in per capita income whichHeston derives from such projection.Heston again strives to obtain the same result by, first, postulating avery large relative size for the income from animal husbandry, and,then, assuming a very high rate of growth for that income. In respect ofthe first he has been criticized by Angus Maddison, who would assign toanimal husbandry an income equal only to 25.15% of the income fromcrops in I900-01; in Heston the corresponding figure is 36.4% forI899-I900.64 An inflated size of pastoral income would naturally

59 IESHR, XV (2), pp. I73-86. 60 CEHI, II, 880-4. 61 Ibid.62 IESHR, XII (i), pp. 55-7 63 CEHI, II, 390.64 See Heston's table 4.3A in CEHI, II, 397, and A. Maddison, 'What did Hestondo?', unpublished note circulated at conference on the CEHI, Cambridge, 1984, wherethe original version of this paper was also presented. Maddison puts the income from

369


17/28

exaggerate the effect of its growth upon per capita national income. Thisgrowth Heston puts at no less than 58.7% between 1868-69 andI899-I900, implying an increase of 40.8%/ per capita.65Such a dramatic increase in pastoral income is opposed to almost allthat we know about conditions of animal husbandry of the period. Thesteady reduction in grazing lands and the disappearance of breedinggrounds of cattle, owing to the expansion of cultivation, adverselyaffected the supply of fodder and the quality of breeds. The RoyalCommission on Indian Agriculture, I928, laid stress on the fact that amere increase in cattle population could not, under such circumstances,add to the total efficiency of livestock.66 Heston ignores the circum-stances as well as the admonition: he proceeds on the assumption that anincrease in the number of livestock must have enlarged the pastoralproduct in an exact proportion.67 He then goes on to suggest an increasein the number of cattle by 60% and of sheep and goats by 45% between1868-69 and I899-I9oo.68 He assembles estimates of varied coverageand doubtful reliability which, however, do not necessarily support histhesis of a population explosion among cattle.69 Such a phenomenonwould be truly astonishing in the face of a dwindling fodder supply; andHeston, too, is none too sure of his cattle 'censuses'. He, therefore, tells usthat 'the main reason' for his argument in favour of a large increase inpastoral income is 'the expansion in the exports of hides and skins fromRs 5 million in 1859 to Rs I 15 million by 19oI'.70 But the moment weconvert these figures into those in terms of I946-47 prices, in whichHeston states all his figures in his table 4.3A, we can see how weak evencrops substantially above the estimates of both Sivasubramonian and Heston for goodenough reasons (Table 6 in the note). For Maddison's own original estimates, whichHeston seems to dismiss rather too casually, see his Class Structureand Economic Growth:India and Pakistan since the Moguls (London, I971), pp. I66-7.65 Heston explicitly assumes a 57% increase, but the figures he conjectures show oneof 58.7%: the discrepancy is left unexplained (CEHI, II, 397, 440). To convert this intoincrease per capita, I have used the population tables constructed by Morris D. Morrisin IESHR, XI (2-3), 312.66 Reportof theRoyal Commission nIndianAgriculture London, 1928), pp. 198-200. Seealso W. H. Moreland, Agricultural Conditions of the United Provinces and Districts(Allahabad, I913), pp. 26-3I; and Vera Anstey, The EconomicDevelopmentof India (4thedn, London, 1952), p. 172.67 CEHI, II, 440. 68 Ibid.69 CEHI, II, 438-9. Atkinson counted ioi.6 million heads of cattle in British India in1895; the official AgriculturalStatisticsreported only 76.7 million in i896-97, implying thedisappearance of a quarter of the cattle population within a year. For the whole of IndiaSivasubramonian estimates 153 million heads in I900-01. A comparison with theAgriculturalStatistics would make one infer that over half the cattle were to be found in theprincely states! 70 CEHI, II, 439-40.

370 IRFAN HABIB


18/28

STUDYING A COLONIAL ECONOMYthis 'main reason' is. In terms of those prices, the value of skins and hidesexported annually rose by Rs 385 million between I859 and I9OI.Heston, on the other hand, conjures up an increase of no less than Rs2,504 million in the income from animal husbandry between 1868-69and 1899-1900oo!71About 67.7% of the increase in national income that Heston estimatesfor the last thirty-two years of the nineteenth century is accounted for byagriculture and animal husbandry; and we have just seen how illusoryare the data of growth that he offers us for these sectors. It is difficult topursue him in detail the rest of the way, but let us note that he has noauthority for a 4% increase in employment in the small sector between1875 and 1895 ('I would not defend the figures ... very stoutly');72 norfor an increase in productivity per worker in that sector during the sameperiod by II.5%.73 Such assumptions need substantiation, since theconditions of small-scale industry at the time were hardly conducive togrowth. (See the preceding discussion on 'De-industrialization' in thispaper.) Heston's fellow-contributor, Sivasubramonian, finds that thework force in this sector went on declining in absolute numbers afterI900;74 that it followed an opposite trend during the thirty yearspreceding I900 cannot therefore just be taken for granted. For theincrease of productivity Heston has no evidence from the small sectoritself, but he invites us to accept it because productivity increased inagriculture (given his own assumptions of constant yields), in large-scaleindustry and in the railways!75 He even assigns the same increase inproductivity to domestic service-'where one might suppose there wereno gains in productivity'!76Such is the substance from which Heston weaves an increase in percapita income in the hey-day of Free Trade. Beyond his own effort,Heston had appealed in an earlier paper to Moni Mukherjee and MorrisD. Morris as two other proponents of the same view.77 Of these, M.

71 For converting the increase in value of exports of hides and skins into I946-47prices, I have used Moni Mukherjee's table on price levels in EconomicHistory of India,i857-i956, ed. V. B. Singh (Bombay, I965), p. 685.72 CEHI, II, 396, 45I. 73 CEHI, II, 451.74 Cf. Sivasubramonian in CEHI, II, 538-41, esp. table on p. 534.75 Heston gives a table (4A.ii) on p. 445 (where the column headings are a littlemisleading), which he describes on p. 450. This consists of Moni Mukherjee's real-wageindices for agriculture and industry (as recalculated by Heston) and his own indices ofproductivity per worker for agriculture and large-scale industry, and an index of outputper worker in railways. The real-wage series of Radhakamal Mukerjee, Kuczynski andK. Mukerji (all of which show declines in real-wages) are ignored. It is this table whichforms the basis for Heston's supposition of an increase in craft productivity (ibid., 45 ).76 CEHI, II, 451. 77 IESHR, X (4), 330 n.

371


19/28

Mukherjee looms fairly large in Heston's chapter in the CEHI, whichquotes from him this disarmingly illogical sentence: 'It is somewhatdifficult to believe in a continually declining sequence of per-capita realincome in as large, as varied and as great a country as India.'78Given this basic philosophy, Mukherjee links the estimates of theearlier writers in such a manner as to produce a rise in per capita incomein the second half of the nineteenth century. First of all, he takes F. T.Atkinson's two estimates, one for 1875 and the other for 1895. Both hadan upward bias because Atkinson was reponding to nationalistcriticisms. Yet when Mukherjee reduced them to common (I948-49)prices, Atkinson's two figures displayed a very slight change: a bare 3%increase in twenty years.79 Mukherjee, therefore, went to DadabhaiNaoroji's estimate of 1867-68, which was based on different assump-tions than Atkinson's, and relative to his, had a markedly downwardbias. Atkinson's estimate for I875 was therefore naturally a hefty 21%higher than Naoroji's (even after its being raised by V. K. R. V. Rao).80Mukherjee converts this difference into an increase in per capita incomeduring the intervening seven years.81 His further refinements consist,first, in building an annual series in which the indexed per capita incomeis made to rise from 169 in 1867 to 210 in I876,82 and, secondly, intransforming the annual series into a table of nine-year movingaverages, which results in artificially spreading this increase over alonger period, and so screening the sudden jump.83Heston does express some reservations about Mukherjee's use of thepoint estimates;84 but his reader never obtains the means of realizinghow arbitrary Mukherjee's linking of Naoroji's and Atkinson's estimatesis, and how, in effect, evidence has been artificially pressed to provewhat had previously been assumed. If one rectifies this, and restrictsMukherjee to Atkinson's two estimates, which follow a common methodand basis of calculation, then, not a rise in per capita income but'stability [stagnation ?]' over the period i875 to I900 would be theconsequence.85

78 From M. Mukherjee's chapter on 'National Income' in V. B. Singh (ed.), EconomicHistory of India, i857-1956 (Bombay, I965), p. 703; quoted in CEHI, II, 404.79 Table IV in ibid., 672. This results when Atkinson's fig. for I895 has been scaleddown by 20% (following R. K. V. Rao) and the i875 figure by 15%. There is littlejustification for the latter, except that 'a smaller scaling down would entail a drop in realper-capita income between I875 and 1895, an unlikely contingency' (ibid.)80 Ibid. 81 Ibid., 686-7, etpassim. 8Ibid., 689.83 Ibid., 70 1-2. Cf. also col. 5 in Table 4A. i. in CEHI, II, 422, and Heston's text, ibid.,418. 84 CEHI, II, 384, 4I8-20.85 See Mukherjee's table in V. B. Singh (ed.), EconomicHistory of India, 689-90: Thereis an actual decline from 202 in I875 (and 210 in I876) to I88 in I900. The words ininverted commas are from Heston's comment in CEHI, II, 418.

IRFAN HABIB372


20/28

STUDYING A COLONIAL ECONOMY 373Morris D. Morris's belief in an increase in per capita income in thenineteenth as well as the earlier part of the twentieth century rests on theabsence of'Malthusian checks'. To him, the increase in population after

1921 implies a stability or increase in per capita income.86 However, ifthe demographic argument is applied to the period before the first worldwar, with which we are here concerned, the opposite conclusion mustfollow. The average expectation of life at birth, calculated from thedecennial censuses, shows the pattern in Table 2.87TABLE 2

Male Femaleexpectation expectation(years) (years)

1872-81 23.67 25.581881-91 24.59 25.54I891-190I 23.63 23.96I901-I 22.59 23.3119I I-21 19.42 20.911921-31 26.91 26.56I93I-41 32.09 31.37

Given the enormous decline in life expectancy between I872 and192I, the supposition of a rise in per capita income would be hard to

justify.88 Directly faced with this dilemma Heston now tells us that thetwo are unrelated-'This seems any way to be the Indian experience'.89But it is, of course, far more likely that the 'Indian experience' has not atall been such as Heston has pictured it, and that the per capita incomeand the expectation of life-span moved in the same direction before1921.One particular set of data, which are significant, though not infallible,for checking the trend in per capita income, relate to real wages. Hestontabulates real-wage indices from Radhakamal Mukerjee and K.Mukerji, and both indicate a substantial fall in real wages fromI857/1860 to 1916/1920 (I34 to Ioo in R. Mukerjee and 145 to IOO nK.Mukerji).90 Jurgen Kuczynski's weighted real-wage index similarlyshows a large continuous decline between I88o and 1919 (127 in

86 IESHR, V (I), 14 and n.87 Kingsley Davis, The Populationof India and Pakistan (Princeton, 1951), p. 62.88 As Heston puts it: 'Up to I920, per-capita income and perhaps food availability arerising, while mortality experience is not improving[!] at all.' (CEHI, II, 414).89 Ibid. 90 CEHI, II, 402: Table 4.5, cols. io and i.


21/28

I880-89 to 98 in I9Io-I9).91 Dharma Kumar, the CEHI editor, herselfspeaks of official figures showing 'a marked decline in the real wage rates(of agricultural labourers) in six out of the seven districts (of the MadrasPresidency) covered in the last quarter of the I9th century.'92 How doesthis accord with the supposed rise in per capita income during the verysame period?93 Heston would replace these real-wage indices by a freshone calculated by himself from Moni Mukherjee's figures for wages andprices.94 The interesting point is that even these show a decline fromI880-8I to 1900-01, while an earlier rise demands scrutiny; andcertainly the large deviations from the results of R. Mukerjee and K.Mukerji have to be explained.

This long discussion of Heston's estimates of per capita income hasbeen necessitated by the seductive quality that detailed figures,percentages and indices have acquired today. Once it is realized thatHeston's figures merely express quantitatively what had been assumedin order to reach these figures, it would become obvious that theprecision of his estimates is deceptive. The estimates do not prove by anymeans that the conditions of life were improving; they have been madeon the premise of improvement; that is all.

V. The Drain and National IncomeI now offer another illustration from Heston's array of figures, wherethese are apt to be quite misleading. The National Income is the sumtotal of the value of goods produced and services rendered internallyplus income earned abroad, or minus wealth transferred abroad. Hestonaccordingly supplies a column on 'Foreign Earnings' in his table on 'NetDomestic Product, 1868-I900'.95 As may be expected, the figures are all

91 V. B. Singh (ed.), EconomicHistory of India, 6I I.92 CEHI, II, 238. Cf. also the same author's LandandCaste n South ndia(Cambridge,I965), pp. I63-7. The decline is all the more remarkable since the wage rates tended tobe under-reported until 1887, whereafter the understatement was 'somewhat less' (M.Atchi Reddy in IESHR, XV (4), 466).93 Cf. Heston: '.. . we again seem to find for the nineteenth century divergent trendsin the per-capita income and real wages, which is not a tidy result' (CEHI, II, 408). Adivergence between the two is, of course, possible if there is a radical change in thedistribution of national income, as in periods of rapid growth of industrial capitalism.Within the Indian economy the landlords apparently gained considerably in the latterhalf of the igth century, but essentially the process was one of'the slow impoverishmentof the mass [rather] than the enrichment of the few' (Stokes in CEHI, II, 65).94 CEHI, II, Table 4A.I I, cols. 2 and 5 and text, p. 450.95 CEHI, II, 397 (Table 4.3 A).

374 IRFAN HABIB


22/28

STUDYING A COLONIAL ECONOMY 375negative, since India was subjected to a continuous drain of wealth.Given in 1946-47 prices, the figures suggest detailed calculation; but, asHeston explains it, the device by which they have been obtained issimplicity itself. He first takes Sivasubramonian's figure for net loss byforeign trade in I900-OI, and then proceeds backwards by assuming1.5% annual growth in foreigners' earnings remitted from India. Forthis there is no greater authority than Heston's own supposition that theannual earnings of foreigners in India doubled between 1857 andI900.96 Such an inspired short-cut dispenses with all the detailedcalculations of the drain by Dadabhai Naoroji, William Digby, Y. S.Pandit and others.

The degree of accuracy that Heston with his simple device achievescan be tested if we turn to Kirti Chaudhuri's chapter in CEHI. Here weare told of S. B. Saul's estimate of Britain's favourable balance ofpayments with India in I880, viz., ?25 million, equal at the currentsterling ratio, to Rs 300 million.97 Converted into I946-47 prices,according to M. Mukherjee's table,98 this gives us a value of Rs I355million. This sum is nearly three times the value Heston enters for1882-83 (there is no entry for i880) under 'foreign earnings'-a mereminus 484 million in I946-47 prices. In 1882, the surplus export ofmerchandise, which can serve as a floor-figure for the net drain,amounted to Rs I237 in I946-47 prices.99 This accords with Saul'sestimate for 1880 and is two and a half times that of Heston. Similarly, in1898-99, according to Y. S. Pandit's estimate, cited by K. N.Chaudhuri, India had a total net deficit of payments of Rs 396.5 million,and an export surplus of Rs 355.6 million,100 equal to Rs 1397 million inI946-47 prices. In I898 the Home Charges alone amounted to Rs 244.8million, that is Rs 961 million in 1946-47 prices. 0 Yet Heston's entiretotal for net foreign payments in 1898-99 in 1946-47 prices, is just Rs625 million.This enormous inaccuracy in representing the size of the drain is notcited just to score a point: the drain of wealth at Rs 1355 million in 1882

96 Ibid., 452-3. This is surely an area where 'hunches' of this kind have the leastjustification, for the quantitative evidence is so rich. For the latest survey of that evidencesee A. K. Banerji, Aspectsof Indo-British EconomicRelations (Bombay, I982).97 Ibid., 873. A. K. Banerji, op. cit., I8-19, argues that Saul underestimates the extentof India's net payments to Britain in I880-8i.98 V. B. Singh (ed.), EconomicHistory of India, 685.99 See table in CEHI, II, 873, for figures in current prices. By 'net drain' I mean thetotal payments made to Britain less British capital exports to India (cf. CEHI, II, 873-4).100 CEHI, II, 874.101 William Digby, ProsperousBritish India; A Revelationrom OfficialRecords(London,I9I0, reprint, New Delhi, I969), pp. 217-18.


23/28

would reduce Heston's total for 'net product' in I882-83 to Rs 32,712million, and the drain would then amount to 4.14% of the nationalincome at that time. Such a continuous loss of savings would becrippling for any economy. Where would investments come from tostimulate any expansion of the economy, when the bulk of the possiblesavings was annually lost?102 And yet there is no discussion in the pagesof the CEHI of the basic quantities and implications of the drain in this'modern' phase of the Indian economy.103

VI. Commercialization of Agriculture and its ConsequencesThe railway construction of the second half of the nineteenth centurymay be said to have completed the 'colonialization' of the Indianeconomy, pulling all its erstwhile isolated segments inside the net ofBritish Free Trade. On the railways the CEHI offers a very competentstudy by John M. Hurd,104 and the relationship of prices and theexpanding markets for agricultural produce is explored by McAlpin. 105But from here the trail is abruptly abandoned. Starvation and famines ofthe utmost severity were a marked feature of the last quarter of thenineteenth century.'06 Was this simply due to a continuation of theolder conditions (then what of the relief expected from the railways?) orto a decline in the availability of foodgrains as a result of the transfer ofland to non-food crops and high-price grain (wheat), or, again, to 'achange in the distribution of income'?107 None of these alternatives are

102 It may be remembered that the total capital formation at the end of British rulewas no more than 5 or 6% of the NDP (Angus Maddison, Class Structureand EconomicGrowth,65; CEHI, II, 948).103 In her section on South Indian economy before 1857 Dharma Kumar doesrecognize that 'the "economic drain" was large', but urges that it 'is not clear that [thiswas] ... accompanied by the impoverishment of the people' (CEHI, II, 360, 375).104 CEHI, II, 762-803. 105 Ibid., 878-904.106 See the useful appendix to the chapter on population which tabulates famines,I750-1947, CEHI, II, 528-3I. The most deadly famine was that of 1896-97, with 96.9millions affected and 5. I million deaths.107 Cf. Michelle Burge McAlpin, 'Railroads, Cultivation Patterns and FoodgrainAvailability: India 1860-90', IESHR, XII (I) (Jan.-March 1975), 43-60, the quotedwords being from her conclusion (p. 58). Also see her earlier essay in JEH, XXXIV(I974), pp. 662-84, where too she disputes the effects of railways on foodgrainavailability. She ought, perhaps, to have treated acreage under wheat separately fromother foodgrains; for it was really the coarse-grain availability which was in question.Nor should one be so dogmatic about the fall in foodgrain availability. Even a 'smalldecline in the share of land planted with grain', plus foodgrain exports, could have had adevastating effect on a very large population permanently hovering upon the verge ofstarvation.

376 IRFAN HABIB


24/28

STUDYING A COLONIAL ECONOMYdiscussed in the CEHI: one looks in vain for the data on growth ofregional specialization, and relative crop-acreage; and on the distribu-tion of rural income, especially the movement of rents. Thus thequestion as to who benefited from the rise in agricultural prices and animprovement in terms of trade in favour of agriculturel08 is left high inthe air.Some of the CEHI contributors seem particularly anxious to avoidany suggestion that the expanding market and increasing differentiationcould have enlarged the ranks of agricultural labourers. This anxietymay explain the slip which Eric Stokes makes at one place in hisinimitable contribution on agrarian relations in Northern India: 'Thepressure towards reduction of the size of purely tenant holdings camefrom the increase of the numbers of the rural proletariat. The myth ofthe rise of the landless labourer . .' etc. What is stated for a fact in onesentence becomes a myth in the next.109

Similarly, Dharma Kumar writing in CEHI omits to refer to her ownimportant finding that in South India the proportion of agriculturallabourers to the total population increased from about 15 or 17 per centto 27 or 29 per cent during the course of the nineteenth century.1l0SurendraJ. Patel's pathbreaking book, AgriculturalLabourers n India andPakistan (Bombay, I952), does not even find a place in the bibliographyof the CEHI. By such silence apparently are old 'classics' to be renderedobsolete.

VII. Imperialism and Indian IndustryPart II of CEHIopens with a long chapter by Professor Morris D. Morrison the growth of large-scale industry."' Set nearly in the middle,Morris's contribution dominates the volume, being the only one wherefull-scale theorization about India and the British regime has beenoffered. To this the reader can relate the conclusions in many of theother contributions whose authors explicitly or implicitly share Morris'searly nineteenth-century, pre-List approach.Morris begins by taking a swing at Karl Marx, who, he says, was'excessively optimistic' in forecasting the development of modern

108 CEHI, II, 894.109 CEHI, II, 63. What Stokes appears to be saying is that large numbers ofagriculturalabourers eld smallbitsoflandandwerenottotally andless.This wouldnotbe seriously isputedby anyone.110LandandCaste n South ndia,168-82. 1ll CEHI, II, 553-676.

377


25/28

industry in India. Not only was Marx wrong here, but if Indian industrydid not develop, there is no reason to put the blame on 'British policywhich inhibited local initiative.'112It has been widely recognized that British tariff policy was mainlydirected towards making India an unprotected market for Britain.ll3Morris immediately proceeds to contest this. Given conditions as theystood, he says, 'scholars are beginning to agree' that the effects of 'avigorous protective tariff policy' would have been 'minuscule'. Heargues that 'the jute and cotton textile industries not only grew swiftlywithout tariff protection but many of their most important markets wereoverseas where tariffs could not have helped.'114This last sentence is typical of Morris's arguments. First, it is illogicalto lump jute and cotton industries together. The jute industry before1914 was almost wholly British-controlled and had, therefore, no serioustariff problems. As for exports, it is obvious that industries with

protected home markets can often afford to accept lower profit marginsin foreign markets; and so protection at home helps them abroad. In anycase, according to Morris, 'the market for Indian cloth was almostentirely domestic'.115 But, above all, it is the general logic of theformulation which is so weak. The textile industry was the only majorindustry, which, owing to close access to cheap raw materials and a largehome market, had some chance of growth. Other industries had nochance at all, and by imposing Free Trade, the British secured their totalsuppression or prevention.Morris repeats his remarks in the succeeding pages as well;"6 and hefinds his case so far proved to his satisfaction that he describes neither thetariff and excise nor the currency manipulation with which Britain triedto stifle the Indian cotton industry. Thus a reader of Morris would notknow that under Lancashire pressure all duties on imports into Indiawere abolished in 1879 and 1882, ostensibly to remove unfair competi-

112 Ibid., 553-4.113 Cf. Angus Maddison, ClassStructure ndEconomicGrowth,56, for a recent statement.114 Ibid., 555. Among scholars who share his outlook on tariffs Morris might have inmind B. R. Tomlinson. In his Political Economyof theRaj, g9I4-47 (London, I979), p. 13(also pp. 15-I6), Tomlinson argues that for India's industrial development protectivetariffs would have been less important than the transformation of 'the traditionalinstitutions of the internal economy.' He refers us in turn to T. D. Rider's unpublishedPh.D. thesis on 'The Tariff Policy of the Government of India and its DevelopmentStrategy, I894-I926'. I suppose if one can see an official development strategy in thatperiod, one can see anything. 115 CEHI, II, 578.116 'Given the widespread impression that industrial development was impossiblebecause of implacable British hostility to Indian competition, the career of the cottonmill industry seems particularly paradoxical.' (CEHI, II, 573).

378 IRFAN HABIB


26/28

STUDYING A COLONIAL ECONOMY 379tion and in the interest of 'the native population', who would getcheaper cloth.What happend in I894 went even beyond the sweeping measures ofI879-82. A tariff of5 per cent was now levied on imported cloth, but a'countervailing excise' was levied on Indian mill-yarn competing withLancashire. In 1896, in order further to protect Lancashire, such excisewas levied on all Indian mill-cloth. Gone was the pretence that FreeTrade was intended to cheapen cloth for the Indian consumer. Morriscannot but notice the setback the Indian cotton mill-industry suffered asa consequence, but the way he puts it is certainly odd: 'Between I894and 1896 tariff agitation generated considerable uncertainty'.17 Whatharmed the Indian industry was thus not the excise measures, but theIndians shouting themselves hoarse over them!Elsewhere in the CEHI K. N. Chaudhuri wonders why the abolitionof tariffs in 1882 did not overwhelm the Indian textile industry andrefers, as a possible factor, to the continuous devaluation of the rupeeowing to silver influx and free mintage.1l8 But this loophole, too, wasplugged when in 1893 the Indian mints were closed.1'9 One is gratifiedto learn from the CEHI that this was done at the insistent demand of theIndians: 'There was considerable agitation in India for the closing-downof the mints for the free coinage of silver and the ultimate adoption of thegold standard'.120 A. G. Chandavarkar has apparently omitted to readthe statement submitted to the Currency Committee, I893, by thatobsolete thinker, Dadabhai Naoroji, who opposed precisely the twomeasures which are here ascribed to an Indian agitation.121 Nor does itstrike him, as it did R. C. Dutt, that a revaluation on this scale (raisingthe rupee from 13. d. in 1894-95 to i6 d. in 1898-99) must have greatlyimperilled the Indian cotton industry.122 The rise in the value of therupee naturally made exports to East Asia and the Middle East lesscompetitive and cheapened British yarn and cloth in India.123If the Indian textile industry even then survived, it was not because of

117 Ibid., 577. Morris surely means the agitation in India over the countervailingexcise, not over the tariff imposed on imports. 118 Ibid.,868.119This step is commended by Chaudhuri: 'the adoption of the gold exchangestandardprovidedIndia with a reallymodern and automatic mechanismregulatingthesupply and demand for foreign exchange'. (CEHI, II, 874-5). 120Ibid., 770.121 Povertynd Un-BritishRule nIndia(London, I9I0), Government of India reprint(Delhi, 1962),pp. 495-8. Cf.BipanChandra, TheRiseandGrowthfEconomicationalismin India (New Delhi, I966), pp. 279ff.122 India n theVictorian ge, 2nd edn (London, 1906), Government of India reprint(Delhi, 1976), p. 433.123 See for a fall in the share of cotton goods in Indian exportsbetween I890-91 and1900-0 , CEHI, II, 844 (Table Io. I ).


27/28

any lack of trying on the part of the British government. In other sectorsits hostility to an Indian industrialization was bound to be more, evenfully, successful.In his sub-chapter on Railways, J. M. Hurd recognizes that theGovernment of India was 'interested not in India's financial andindustrial development but in Britain's,' and that it urged the railway

companies to 'buy British' right down to I924.124 Its refusal to utilizethe capacity of Indian railway workshops to produce competitivelocomotives, to which Hurd refers,125 is dismissed summarily in afootnote by Morris.l26 He has no time to consider the fact that owing toofficial policy, the railways failed to act as a stimulant for heavy andmachine-building industries as they did elsewhere in the world. Thisjudgement is left for Hurd to deliver:India's loss from the purchase policiesof the railwayswas not limited to herlackof progress n developing heavy industry. She also failed to reap the benefitsofthe spreadeffects to industrywhich would have occurred. Instead, the spreadeffects stimulated the Britisheconomy.127

The farthest Morris would go is to acknowledge that 'of course,official policy did not strongly [!] support the development of iron andsteel industry'.128 Seemingly oblivious to the irony of it, he applauds theaction of the Government in pledging itself to buy Io,ooo tons of ironfrom BISCO, if the price was 5 per cent lower than that of English ironupon its being landed in India.129 If iron and steel industry did not stilldevelop before 900oo, it was apparently the fault of 'entrepreneurialbehaviour'.130Whatever else might have been responsible for the failure of Indianindustry to develop, Morris exonerates the colonial regime from anyculpability throughout. The real culprit, he finds, was India's low per

capita income. It was only half that of Japan at the time of the MeijiRestoration: 'This suggests the inappropriateness of consideringJapanese development as a model of what would have happened inIndia had political conditions been different'.131 It is extraordinarilydifficult to compare the incomes of two pre-industrial countries, sodifferently placed in respect of climate, diet, and pattern of consump-tion; and yet Morris boldly makes the comparison the cornerstone of atheory. The theory must imply an unrelieved pessimism for India, for, as124 CEHI, II, 749 125 Ibid.126 'Railwayworkshopswere importantcentres of large-scaleproduction,butunfortunatelyhis sanactivityaboutwhich t is notyetpossibleosaymuch'.(CEHI,II, 566 n.) 127CEHI, II, 749. 128 Ibid.,587.129 Ibid.,586. 130 Ibid.,587-8. 131 CEHI, II, 60o n.

380 IRFAN HABIB


28/28

STUDYING A COLONIAL ECONOMYAlan Heston stresses, it has a receding per capita income relative to theadvanced countries (e.g. U.S.A.),132 and also, presumably, relative tothe oil sheikhdoms. We have, or ought to have, therefore, no industrialfuture, whatever be our 'political conditions'.This critique of the structure of contents and the dominant message ofthe CEHI II has necessarily been largely negative. It has addressed itselfto showing that the CEHI offers an interpretation of modern Indianeconomic history which is both incomplete and tendentious. The CEHIomits any serious scrutiny of colonialism and underrates its impact onIndian economy. To say this does not mean that the nationalist view ofthese matters as stated in the 'classics' is all that is required to reconstructmodern economic history; or that the nationalist view did not haveflaws, such as too often ignoring internal contradictions within theIndian economy or (much less often) presuming an earlier golden age.But the nationalists undoubtedly had a case; that case cannot bedismissed by mere claims to superior knowledge or better doctrine. Theopposite picture that the CEHI II constructs, pre-eminently through thecontributions of Heston and Morris, has far less to recommend it: in solarge a part here have mere impressions been transformed into statisticsand assumptions made to pre-determine conclusions. The purpose of thepresent critique would be served if the reader is sufficiently warnedagainst accepting this picture as anything more than yet anotherspeculative exercise by one side in the old great controversy.

132 Ibid., 415.

38I

habib studying colonial economy without perceiving colinialism modern asian studies critique of...

Documents