"The entire history of civilization," wrote the French geographer Demangeon, is reflected in present forms of human settlements."1 Though possibly exagger- ated, that statement recognizes a fundamental truth: that the tangible manifesta- tions of man's current occupance of the earth bear abundant testimony to the historic currents to which his forebears have been subjected. In giving expression to his present culture, they tell also of his past. And where the buildings of past eras persist, whether in monumental architecture or in simple peasant dwellings of durable construction, the fossilized past appears before our very eyes. It is primarily in the light of these considerations that a special section of this atlas is given over to the representation of settlement patterns.
Obviously, however, history and culture are not the sole determinants of settle- ment form. The physical environment offers man different sets of building ma- terials in different regions. The availability of water and the fertility of the soil constrain the spacing and size of settlements. The intensity of rainfall suggests the desirable slope for a roof, the nature of summer heat the desirable thickness for a wall. While the elements of physical geography are not among the topics to which this atlas gives major emphasis, some of the more important among them are considered in section I as a background for the better comprehension of history in any age. And, since the physical elements themselves have been in- fluential in shaping the course of history—helping, for example, to make some areas more attractive or powerful than others—they have exerted an indirect as well as a direct influence on the development of culture and the establishment of settlement forms.
XII.A.1 and 2. Rural Settlement Patterns
Although South Asia displays an enormous variety of settle- ment types, persons traveling extensively through that area will be struck by the tendency for roughly similar settlement forms to be associated with particular regions, even though other forms may also be found therein. A rapidly growing body of literature is available from the Anthropological Survey of In- dia, from the India Census, and in the work of university schol- ars in South Asia and elsewhere to illustrate these varying re- gional patterns of settlement. The Anthropological Survey's admirable study Peasant Life in India, first published in 1961, provides what is to date the only noteworthy attempt at a com- prehensive systematic study of the subject in India; but there is as yet no comparable work for the whole of South Asia. What is presented in map (a) of plate XII.A.1 is based largely on the work just cited for India and on a wide variety of addi- tional sources, including personal observations in the field for both India and other areas and detailed scrutiny of topographic maps, at the scale of one inch to the mile, for the entirety of Sri Lanka. It seems quite likely that in the course of future research the need for significant modifications of our map will become apparent. Nevertheless, we offer it now in the hope that even in its present, fairly crude form it may be of some utility.
Maps (b) through (m) illustrate the various settlement forms whose distributions are presented on map (a). The spe- cific site (or in one case, in Nepal, an approximate site) of each settlement depicted is also given on map (a). Obviously these maps (a number of them might better be described as "plans") vary greatly in quality, depending on the sources used, each of which is indicated on the map itself. At one ex- treme we have the exceedingly detailed and accurate map of "Piparsod" (map (c)); at the other are a number of highly generalized sketches that, despite their simplicity, may convey some sense of the spatial layout and pattern of residential seg- regation of the settlements shown. Using whatever information was available to us, we have tried to standardize the scale and legend for all the settlement maps. Even in those cases where we have noted "not drawn to scale," since no scale was pro- vided in the original source, we have tried to approximate the scale used elsewhere. But the similarity of scale may not be evident insofar as what are depicted in certain maps are indi- vidual houses or structures, several of which may make up a single homestead (e.g., maps (c), (i), or (m)), whereas other
To round out the information on rural settlements in South Asia and to provide some useful comparative information on urban settlement, a statistical table has been included on plate XII.A.1 based on 1961 census materials, or, for Ceylon and Nepal, where those were not available, on those of the closest census year. The data make clear the preponderantly rural locus of the populations of all the countries of South Asia. Additionally, they indicate, for India and Nepal, the propor- tion of the total and rural population living in villages of par- ticular size categories. Census data on average village size, however, must be used with caution, for in a number of regions of South Asia, including major areas of dispersed rural settle- ment (e.g., Kerala, West Bengal, and Bangladesh) and several areas with numerous hamlets (e.g., Tamil Nadu and parts of Bihar and eastern Uttar Pradesh) the census village is a mere administrative convenience, the bounded area of which often bears little relationship to the actual layout of settlement on the ground. With this important caveat in mind, one may wish to consider the average village population of the following countries as of 1961 (1952–54 for Nepal): India, 635; Pakistan (then West Pakistan), 898; Bangladesh (then East Pakistan), 747; and Nepal, 142. Overall, the relationships among Paki- stan, India, and Nepal are probably essentially correct. The figure for Bangladesh, in our judgment, is meaningless. No figure could be derived for Sri Lanka (Ceylon) because of the absence of data on the number of villages in the country.
Sources (in addition to censuses cited in text, highly selective)
For Map XII.A.1 (a) (books only, save for one lengthy article, relating to the whole of India)
N. K. Bose, ed. (1967); K. H. Buschmann (1954); E. K. Cook (1951); S. F. De Silva (1954); W. Donner (1972); T. Hagen (1971); J. Humlum (1959), (1969); P. P. Karan (1960); R. Mukerjee (1968); A. Sievers (1964); R. L. Singh (1971); O. H. K. Spate and A. T. A. Learmonth (1967).
For Maps XII.A.1 (b) and (c) and XII.A.2 (d)–(m)
Sources appear on the maps themselves.
XII.A.3. Rural House Types of South Asia
While poverty condemns a very large part of South Asia's population to live out their lives in simple mud huts and other
The sources utilized in preparing plate XII.A.3 are indicated on the bottom of the plate itself.
XII.B.1–3. Calcutta, Lahore, and Other Cities
In this section of the atlas we consider fourteen cities of highly diversified form and function, ranging from ancient pil- grimage centers and an entrepôt for caravans to modern indus- trialized centers and spatially planned administrative capitals. Our intention is to illustrate the ways in which the diversity of urban form and the specific content of each city reflect the purposes the city has been intended to serve and the historical periods over which it has developed. For twelve of our cities (maps (e) through (h) of plate XII.B.2 and (a) through (h) of XII.B.3) a single view is provided, presenting the en- tire city, if the scale permits, or a portion of it that epitomizes its essential character. Two cities, Calcutta and Lahore, have been singled out for more detailed treatment, on plates XII.B.1 and 2 respectively. Calcutta, established by the British as a trading post in the late 17th century, had become by the close of the 19th century the largest metropolis of South Asia and the second largest city of the British Empire. From 1833 to 1912 it served as the capital of British India and the principal link between India and the West, which latter function it largely retains. Lahore, by contrast, has deep indigenous roots, per- haps extending back for nearly two millennia. Noted by Hsüan- tsang as a large Brahmanical city in the 7th century, its fate for roughly eight centuries following 1036, when it became the