The more significant events in the development of modern industry in South Asia and in the concomitant organization of industrial labor are pointed out on the end cover chronology in the column headed "Economic and Demographic Evolution." Cotton milling was the earliest type of modern manufacturing to undergo rapid expansion. Around the turn of the 20th cen- tury, when factory employment passed the half-million mark, cotton milling accounted for roughly two-thirds of that total. It was not until 1911, when iron was first produced at the Tata plant at Jamshedpur, that any noteworthy development of pro- ducers' goods industries took place. By 1931 the total factory employment was approximately two million, roughly 1.3% of the then-existing labor force. Of that total more than a third was found in and around three cities, Calcutta, Bombay, and Ahmedabad, and only two others, Kanpur and Rangoon, had more than 25,000 factory workers each. Under the stimulus to production provided by World War II and by the nation- building élan of the early years of independence, South Asia witnessed a major expansion and diversification of industrial employment and output, as well as a diffusion of modern man- ufacturing throughout the subcontinent. The rapid rate of growth is documented by the figures of our table and, more strikingly, by the map showing the distribution of factory em- ployment as of 1961.

Comparing the 1961 map of factory employment with that for 1931, one notes that, while the leading centers of the ear- lier date maintain their preeminence, there has been a great proliferation by 1961 of second- and third-order centers. The principal centers of relative growth were in South India as far north as the Krishna-Godavari delta region; the upper Ganga- Yamuna Doab, plus the area around Delhi; the Punjab, in both India and Pakistan; and the isolated districts of Karachi and Dacca. One is also struck by the remarkable proliferation of minor inland centers of manufactures in 1961, extending to all but a few districts of the subcontinent. While the aggregate numbers of employees in such centers are certainly not large, the fact that the seeds of the Industrial Revolution have now been disseminated over so wide an area is, in our judgment, of considerable significance.

Total employment in registered factories in India and Paki- stan in 1961 exceeded 3.5 million; yet, despite gains since 1931, the growth of the population as a whole was such that factory workers still constituted less than 1.7% of the com- bined labor force of the two countries. Indeed, as is suggested by maps (b) and (d), showing factory workers as a propor- tion of the total labor force (though comparisons are diffi- cult because of the noncorrespondence of boundaries), certain areas actually lost ground, in a relative sense, over the three intervening decades. In neither 1931 nor 1961 did any prov- ince/state have as much as 6% of its total labor force engaged in factory manufacturing, and there were still a number of major states of India in the latter year where even the 1% figure had yet to be attained.

The proportional squares on map (d) showing value added⁵ by manufacturing by major administrative subdivisions con- tain no great surprises, reflecting closely the regional pattern in the district level data of map (c). Maharashtra, West Ben- gal, and Gujarat—thanks to the continuing, though no longer extreme, dominance of Bombay, Calcutta, and Ahmedabad respectively—lead all other states. Madras, the fourth-ranking Indian state, though lacking any single manufacturing center comparable to the three just named, fares as well as it does because of its notable number of second- and third-order cen- ters. Within Pakistan the pronounced industrial lead of the western over the more populous eastern wing is evident. The industrial stagnation of the latter area was among the griev- ances that contributed to its ultimate secession.

Sources

General References

East Pakistan, Statistical digest . . . (1965); G. L. Harris et al. (1973); India (Republic), Central Statistical Organization, Statistical abstract, . . . (various years).

Government Documents

Afghanistan, Ministry of Planning (1965/66); Ceylon, De- partment of Census and Statistics, Census of industry . . . (1954); East Pakistan . . . (1965); India, Census (1961); India, Department of Commercial Intelligence and Statistics, Large industrial establishments . . . (1933); India (Republic), Central Statistical Organization, Annual survey of industries, (1961); United Nations, ECAFE, Statistical yearbook (vari- ous years); United States AID Mission to Pakistan (1966); West Pakistan, Directorate of Labour Welfare (1964).

Other Works

G. S. Bhalla (1959); V. G. Kulkarni (1968); B. P. Shreshtha (1966), (1967); D. N. Wilber (1962).

XI.C.2 and 3. Factory Employment in Selected Industries, 1931 and 1961

Of the multitude of industries now to be found in South Asia we have selected for special treatment on plates XI.C.2 and 3 four groups that together accounted for slightly less than two-thirds of the employment in large-scale industrial es- tablishments⁶ in 1931 and nearly three-fourths of such em- ployment and of the value added by manufacturing in 1961. The relevant figures for these industries within a broader South Asian industrial context are provided by the table on each map plate. As on plate XI.C.1, data on these more specialized maps are, for reasons of scale and comparability, aggregated by dis- tricts, despite the fact that a more detailed locational break- down was possible for 1931. For all maps plotted data were centered on the leading town/city of each district.

Of the four industrial groups mapped, the textile industries enjoy a commanding lead in both 1931 and 1961. Despite their doubling in employment between those two years, their share of the total employment in large-scale manufacturing in South Asia declined from about 45% to about 41% in 1961. This decline is symptomatic of growing industrial maturity and par- allels a historical trend in industrial development that many other parts of the world have also experienced. Within the combined textile industries cotton continued to hold a pre- eminent position, and save for the Calcutta area, where jute leads, there were few manufacturing centers where cotton did not account for at least three-fourths of the total employment. Calcutta, Bombay, and Ahmedabad, though still leading all other textile centers in employment in 1961, faced much greater competition from other cities than they did in 1931. The rise of the jute and cotton textile industries, in East and West Pakistan respectively, since independence is striking, the former being largely responsible for the Calcutta-Hooghlyside conurbation's loss of its preeminent position among textile- milling localities in South Asia.

The growth and diversification of metallurgical industries in the period 1931 to 1961 has been dramatic. Employment dur- ing that period increased by well over 500%. Whereas only a single center, Jamshedpur, had more than 10,000 employees in 1931, the number of such centers had risen to six within the next thirty years, and a still larger number of lower-level centers had sprung up. Of special note is the belt of basic metal producing centers in northeast India, from Rourkela in Orissa to Calcutta. Here lies the country's core region of heavy in- dustrial development, particularly in the manufacture of iron and steel. The relative insignificance of basic metallurgical in- dustries in Pakistan is striking and poses a fundamental prob- lem for its future economic development.

Of the four industrial groups considered on plates XI.C.2 and 3, the engineering industries displayed the most even dis- tribution of employment in both 1931 and 1961. Although Calcutta (i.e., the Calcutta-Hooghlyside conurbation) and Bombay were the leading centers in both years (with Bombay taking over Calcutta's leading position in the intervening pe- riod), the degree of their dominance, as in the textile industry, has been appreciably reduced. The rise, by 1961, of a number of sizable centers of engineering activity in southern India is striking, especially the mushrooming of Bangalore as South Asia's third largest center. The relatively rapid growth of engi- neering industries in West Pakistan is also noteworthy.

From 1931 to 1961 employment in South Asia's chemical industries more than tripled, and, as one would expect, the range of goods produced was greatly diversified. Compared with the other industries considered on plates XI.C.2 and 3, enterprises in the chemical industries are characteristically small in scale. Whereas in other, more developed countries, chemical industries tend to agglomerate in a few localities, often close to heavy industry of other kinds or to centers of petroleum refining (many plants depending largely for their raw materials on the by-products from other nearby plants), that tendency is not apparent in South Asia. The industry appears, in fact, to be almost as highly dispersed as the engineering industry. Even 6 For the employment criteria used in compiling the industrial data on which our maps were based, one should consult the second paragraph of the text for plate XI.C.1. in 1961 there were no more than a handful of centers with more than 10,000 employees each.

Sources

Same as for plate XI.C.1.

XI.D.1–4. Growth of Road, Rail, and Irrigation Canal Networks, 1872–1961; Growth of Air Services, 1947–70; Electric Power Generation and Transmission Network, 1961

The four map plates discussed in the following paragraphs depict the expansion in modern times of the more important components of the economic infrastructure of South Asia. Though many historical accounts of modern South Asian his- tory touch only lightly, if at all, on this general theme, a proper appreciation of the processes of colonial expansion, economic and social development, formation of national consciousness, and, in recent decades, national consolidation requires that attention be paid to the important role that roads, railroads, and other infrastructural facilities have played in relation to them.

In our presentation of the growth of the road, rail, and irri- gation canal networks on plates XI.D.1–3 we portray the avail- able data for the four benchmark years—1872, 1901, 1931, and 1961—that we have selected for mapping many other phenomena in sections X and XI. Because of the recency of development of air services and electric power, however, we have portrayed their development in the post-independence period only. The sources of our data, both cartographic and textual, are sufficiently detailed so that we can be reasonably certain of the existence of a particular rail line, canal, air route, or portion of the power grid as of the dates of the re- spective maps on which they appear. In the case of roads, however, it is often a matter of judgment whether or not a particular route was important enough to be shown. Where doubts existed, we have attempted to check sources from vari- ous dates close to that specified for the map in deciding on what to include. Similarly, some judgment was called for in identifying roads classified as major thoroughfares.

Before studying the four maps of plate XI.D.1, relating to highway development in South Asia, one might profitably con- sider James Rennell's route map of 1788, a facsimile of which is presented on plate VII.A.6. This is, to the best of our knowl- edge, the first "modern" road map of India. While it shows that virtually every town of consequence was even then linked by a road of sorts with other nearby towns and cities, it says nothing about the arduousness of overland transit along most of the routes depicted. Except for a few major thoroughfares, such as the Grand Trunk Road, roughly following the route of the ancient "royal road" of the Mauryas and restored and improved under Sher Shāh and his Mughal successors, roads in precolonial India were ill-maintained and beset by obstacles to movement. The road development that took place under the British during the first three-fourths of the 19th century was also unimpressive, and such main routes as were constructed (cf. plate VII.B.2) were intended largely to speed the move- ment of troops in case of civil disturbances such as that of 1857. Goods movement over short distances was predominantly by bullock cart—a very important form of transport in the South Asian countryside to this very day—and with economic life focused on the coastal presidency towns and cheap rail trans- port developing apace, there was little economic incentive for extensive highway development in the interior. By 1901, how- ever, some of the more productive portions of the Indo-Gan- getic Plain, the Deccan Lava Plateau, and the far South of India witnessed the beginnings of a highly ramified route net- work. Automobile transport was slow to come to South Asia; it was not until 1929 that the first motoring map of the area was published by the Survey of India. Although by 1931 motorable highways had penetrated to virtually every town of consequence in India not located on a navigable stream, large parts of the subcontinent, such as eastern Bengal, Assam, the Chota Nagpur, and Sind, continue to be very badly served. The principal burst of road-building activity came only in the post-independence era. Both the 1961 map and, indirectly, the table on plate XI.D.1 testify to this rapid phase of develop- ment. The map, however, completely fails to suggest the re- markable proliferation of short motorable feeder roads, largely built by volunteer labor, linking tens of thousands of villages with main metaled routes, which has occurred under the aegis of community development programs in many parts of South Asia.

Although India's first steamship was constructed as early as 1823, another thirty years were to pass before the country wit- nessed its first steam locomotive. Thereafter, however, the Brit- ish were quick to perceive the military and economic advan- tages of rail transport, and, as plate XI.D.2 shows, by 1872 the key arteries had already been established. The pace of new railroad construction probably peaked before the century's