Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The spatial organization of a south Indian city : Coimbatore Joy, Annamma 1975

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1975_A8 J69.pdf [ 6.13MB ]
JSON: 831-1.0093470.json
JSON-LD: 831-1.0093470-ld.json
RDF/XML (Pretty): 831-1.0093470-rdf.xml
RDF/JSON: 831-1.0093470-rdf.json
Turtle: 831-1.0093470-turtle.txt
N-Triples: 831-1.0093470-rdf-ntriples.txt
Original Record: 831-1.0093470-source.json
Full Text

Full Text

TH3 SPATIAL ORGAMIZATTO" OF k SOUTH IMDIAN CITY - COT TORF. Annamma Joy v,k, i n S o c i a l Science, Madras University, 1973 A THESIS SrPWPTED I* 1 PARTIAL FHLFTL'V3NT 07 THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n tho Department of Anthropology and Sociology We accept t h i s t h e s i s as conforming to the requ i r e d standard THS UNIVERSITY OF BRITISH COLUMBIA August,, 1975 In presenting th i s thes i s in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make i t f ree l y ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th i s thes i s for s cho la r l y purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t i on of this thes i s for f i nanc ia l gain sha l l not be allowed without my writ ten pe rm i ss i on . Department of The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date Lf^ W -i ABSTRACT The s p a t i a l organization of a t r a d i t i o n a l Indian c i t y c l e a r l y reflected certain central s o c i a l values. At the heart of India's oldest c i t i e s lay a l l the administrative, commercial and r i t u a l l y s i g n i f i c a n t spaces, as w e l l as the residences of certain high ranking castes. At the peripheries were the residences of the lower castes and beyond the c i t y l i m i t s altogether were the untouchable settlements. The question raised by this thesis i s whether the physical layout of a modern i n d u s t r i a l c i t y refects these t r a d i t i o n a l s o c i a l values, or perhaps, new ones. Has the introduction of modern business and industry i n Coimbatore brought about a clear-cut d e l i n i a t i o n of commercial and r e s i d e n t i a l space? Has i t likewise necessitated the separation of administrative and r i t u a l areas? This analysis of the s p a t i a l organization of the modern, i n d u s t r i a l Coimbatore concludes that the o r i g i n a l s p a t i a l layout of t h i s South Indian c i t y has not been t o t a l l y replaced by a new one. Instead, t h i s vast urban complex has simply incorporated certain modern areas within i t s growing l i m i t s . Coimbatore c i t y , then provides a blend of old and new p r i n c i p l e s i n i t s physical layout. This blend can be seen as a result of a slow h i s t o r i c a l process, i n which the c i t y grew by small acretions. The new areas added by i t s nineteenth century c o l o n i a l residents became structured according to foreign p r i n c i p l e s , while some of the c i t y ' s older areas remained r e l a t i v e l y unchanged. The data on which this analysis rests include information on the caste id e n t i t y of every household on every street for major areas of the c i t y as of 1972, street maps of the c i t y from 1871, 1939 and 1974 bearing locale names that help i d e n t i f y the s o c i a l i d e n t i t y of par t i c u l a r areas, and also maps of the locations of telephones and .important businesses i i (also for 1972). Copies of a l l maps are provided with the thesis and could be used i n making a further longitudinal study of the ci t y ' s s o c i a l characteristics i n the future. In sum, the combined responses of the c i t y of Coimbatore to various external forces over time, some s o c i a l , some m i l i t a r y , and some economic, best explain the blend of t r a d i t i o n a l and modern elements i n i t s s p a t i a l layout today. i i i TABLE OF CONTENTS Page LIST OP TABLES v i LIST OF FIGURES v i i LIST OF MAPS . v i i i Chapter 1. INTRODUCTION • 1 SOURCE MATERIAL 2 2. TOWARDS A MODEL FOR THE CITY OF COIMBATORE 7 MODELS USEFUL IN THE STUDY OF URBAN CENTRES . . . . . . 9 The V i l l a g e Model . . . • • • • • 10 The City Model 11 The Regional Model • 13 Salient Features of the Supplementary Model 15 Central areas/peripheral areas . . . . . . . . . . 16 Pure areas/polluting areas • • • • • • 16 Areas associated with dominance/areas not associated (Right/Left) 16 Commercial/Residential . . . . . . . . . . . . . . 17 3. HISTORY OF THE GROWTH OF THE CITY 18 THE EARLY PERIOD: 1300-1759 . . • 19 THE MOSLEM PERIOD: 1759-1799 19 THE FIRST HALF OF THE 19th CENTURY • 20 THE SECOND HALF OF THE 19th CENTURY 21 The Origi n a l Nucleus 23 Central Areas/Peripheral Areas . 25 Pure/Polluting 26 i v Chapter Page Commercial/Residential 27 Right/Left Caste 27 Conclusion . . . . . 27 THE FIRST HALF OF THE 20th CENTURY 30 The Swadeshi Movement 32 POST-INDEPENDENCE YEARS 34 k. THE ANATOMY OF COIMBATORE CITY 33 THE NEUTRAL CASTES 49 The Brahmins . . . . . . . 50 The P i l l a i s 52 THE LEFT DIVISION OF CASTES 52 The Asaris 53 The Cettiyars 54 The Mutaliars 54 The Naickers 55 THE RIGHT DIVISION OF CASTES . . . . 56 The Kavuntars 57 The Natars . . . . . . . . . . . . 57 The Tevars 58 THE SERVICE CASTES . . • & The Vannar . . 59 The Navithar . . . . i . . 59 OUTSIDE REGIONS 60 The Naidus 6l The Boyar 6l The Malayalis 62 V Chapter Page NON-HINDU 63 The Moslems 63 The Christians 6h THE UNTOUCHABLES 65 INDEX OF SEGREGATION 6? INDEX OF DISSIMILARITY 67 CONCLUSIONS . . . 72 5 . CONCLUSIONS 7k Old Core and Old Periphery • • 77 The New Core • 78 Older Suburbs 78 The New Suburbs 79 The Industrial North and the New Peripheries . . . . 80 BIBLIOGRAPHY 82 APPENDIX . 89 v i LIST OF TABLES Table Page I. Classification of Street Names of the Map of 1871 on the Basis of Caste, Community, Commercial and P o l i t i c a l Identity 2h I I . Classification of Textile Mills i n Coimbatore With Respect to the Number of Looms, Number of Spindles, and Financial Layout 36 I I I . The Caste Composition of the Kongu Region, Coimbatore City, and the Sample City Population • Ul IV. The Index of Segregation of Castes i n Coimbatore City . . 69 V. Index of Dissimilarity Between Castes i n Coimbatore City . . . . . . . . . • • 71 v i i LIST OF FIGURES Figure Page 1. Diagrammatic Representation of Caste Data for the Kongu Region, Coimbatore City and the Sample City Population . . . . . . . . . 1;6 2. Index of Segregation „ 68 T W i i * * ^ S^oVl G l k W (uc3) ^ LIST OF MAPS Page MAP 1 - The sectoral division of Coimbatore City in 1972 « . . . h MAP 2 - The spatial layout of the village of Olappalayam in 1965 11 MAP 3 - Organization of space in Puranapur . 12~~) m eni/e'o^e MAP k - Coimbatore City in 1871 22 I MAP 5 - The organization of central and peripheral areas in the city of Coimbatore in 1871 . . . . . . . 25 MAP 6 - The demarcation of pure and polluting areas in Coimbatore in 1871 26 MAP 7 - The commercial and residential use of space in Coimbatore City in 1871 • • • • 27 MAP 8 - Map of Coimbatore City in 1939 3h m t a ^ e . MAP 9 - An illustration of the general density of population in Coimbatore City in 1972 U7 MAP 10 - Social character of the different sections in Coimbatore City in 1972 • h7 MAP 11 - Composite representation of all the caste concentrations in Coimbatore City in 1972 U8 MAP 12 - The spatial distribution of residential telephones in Coimbatore City in 1972 U8 MAP 13 - The spatial layout of business telephones in Coimbatore City in 1972 U8 MAP lit - The spatial location of large business firms in Coimbatore City in 1972 • • k9 MAP 15 - The spatial distribution of the Brahmins in Coimbatore City in 1972 50 MAP 16 - The spatial distribution of the Pillais in Coimbatore City in 1972 52 MAP 17 - The spatial distribution of the Asaris in Coimbatore City in 1972 .. 53 MAP 18 - The spatial distribution of the Cettiyars in Coimbatore City in 1972 . . . . . . . . . . 5U i x Page MP 19 - The s p a t i a l d i s t r i b u t i o n of the Mutaliars i n Coimbatore City I n 1972 . . . . . . . . . . . . 55 MAP 20 - The s p a t i a l d i s t r i b u t i o n of the Naickers i n Coimbatore City i n 1972 . . . . . . . . . 55 MAP 21 - The s p a t i a l d i s t r i b u t i o n of the Kavuntars i n Coimbatore City i n 1972 . . . . . 57 MAP 22 - The s p a t i a l d i s t r i b u t i o n of the Natars In Coimbatore City i n 1972 57 MAP 23 - The s p a t i a l d i s t r i b u t i o n of the Tevars i n Coimbatore C i t y i n 1972 58 MAP 2l| - The s p a t i a l d i s t r i b u t i o n of, the Vannars i n Coimbatore City i n 1972 59 MAP 25 - The s p a t i a l d i s t r i b u t i o n of the Navithars i n Coimbatore City i n 1972 60 MAP 26 - The s p a t i a l d i s t r i b u t i o n of the Naidus i n Coimbatore City i n 1972 6 l MAP 27 - The s p a t i a l d i s t r i b u t i o n of the Boyars i n Coimbatore C i t y i n 1972 . • 62 MAP 28 - The s p a t i a l d i s t r i b u t i o n of the Malayalis i n Coimbatore City i n 1972 62 MAP 29 - The s p a t i a l d i s t r i b u t i o n of the Moslems i n Coimbatore City i n 1972 63 MAP 30 - The s p a t i a l d i s t r i b u t i o n of the Christians i n Coimbatore City i n 1972 6U MAP 31 - The s p a t i a l d i s t r i b u t i o n of the Untouchables i n Coimbatore City i n 1972 65 :.x.. ACKNOWLEDGEMENTS I would like to take this opportunity to thank a l l ray good friends and well wishers both within this department and outside of i t for their advice and assistance i n the preparation of this thesis. However, there are a few to whom I would l i k e to express my gratitude i n a more specific manner* Fi r s t of a l l , I would like to thank the Chairman of my committee, Dr. B.E.F. Beck, for her unfailing enthusiasm, invaluable comments, and useful criticism. I would also li k e to thank the other members of my committee, Dr. M. Ames and Dr. T. Fernando, for their advice and helpful comments at various stages of the thesis. Dr. D. Ley, of the Geography Department, graciously agreed to read the thesis i n i t s draft stages and offerred useful comments as well. I would also li k e to thank my good friends, Dr. and Mrs. A. Misra and Mr. V.V. Baba, f o r a l l their unfailing enthusiasm and their kindly assistanceo This thesis was also written with the help of a grant from the Canada Council. Last, but not least, I would like to extend my thanks to Mrs. Joan Prentice for her remarkable patience i n typing this thesis. Chapter 1 INTRODUCTION This thesis examines the relationship between spatial organization and social values i n the city of Coimbatore, i n south India, The findings of this study reassert: the significance of a traditional social i n s t i t u -tion such as the caste system i n the organization of urban space. Further, the above conclusion allows us to integrate the study of the city with earlier anthropological studies of both the village and the region. Fin-a l l y , i t suggests that there i s a syncretism of the traditional and the modern elements i n the spatial order of an Indian c i t y . The name Coimbatore originates from the term 'Kovan-Puthur* by which the earliest settlement was known. Like many other similar terms, i t had been modified for the convenience of the colonial officers. But even after Independence, the city continued to be referred to as Coimbatore and not as Kovan-Puthur, Coimbatore, one of the leading centres for the cotton textile i n -dustry i n India, i s also the administrative centre for the d i s t r i c t of Coimbatore and one of the core c i t i e s of Tamil Nadu. According to the 1971 census of India, Coimbatore had a population of 35>3,U69 (Census of India 1971, Paper I of 1971 - Supplement llj.6). As Bose (1969) rightly notes, [Coimbatore like] "most Indian c i t -ies, although the handiwork of the colonial government, have been peopled by Indians who bring to these urban centres their traditional institutions" (quoted from Berry 1969s 1+67). To conclude, therefore, that the organiza-tion of space i n an industrial c i t y t e l l s us something concrete about the 2. traditional social structure i s not fallacious or far-fetched. On the con-trary, i t i s certainly clear from the mapping of the household distribution i n Coimbatore that traditional spatial patterns have not been replaced by Western spatial patterns, but that the former have been juxtaposed with the l a t t e r . Source Material The above conclusions are based on an analysis of an unusual set of caste data for the ci t y of Coimbatore collected by Brenda Beck from the National Malaria Office i n 1972. In order to study the spatial patterns inherent i n the 1972 data, a detailed current map of the city was obtained from Mr. K. Sundaram, who personally spent a l o t of his time i n the city perfecting his original sketches. Mr. Sundaram's private map i s probably the only detailed map of the city i n existence today.''" Our special thanks are due to both Mr. Sundaram and to Mr. G. Krishnaraoorty, the subdivisional officer of the National Malaria Board at Coimbatore who, by their kindly help, have made this thesis possible. The Coimbatore sub-division of the National Malaria Board, l i k e any other Public Health unit i n India, r e l i e s heavily upon the collection of door-to-door information by local level workers. The local level workers employed by the Board go around the ci t y every eleven days and record de-t a i l s of deaths and cases of fever and epidemics l i k e cholera and typhoid. There were 36 such workers employed at Coimbatore i n 1972. Each of them was provided with a l i t t l e register i n which they were to note the number and date of their v i s i t s , the street, the head of each household, and There have been other maps of the city compiled for tourist purposes, but none represent the layout of a l l the streets i n the c i t y . 3. certain sociologically important information such as family size, age structure, and the relationship of each resident individual to the house-hold head* In addition, Beck found that these workers had been noting the caste identity of about 90 percent of the households visited as a private means of helping them to remember the individuals they met. This information, of course, i s not o f f i c i a l l y required by the governmentj indeed, the o f f i c i a l collection of data relating to the caste of individuals i s greatly discour-aged. But such knowledge has proved to be very helpful to individual work-ers i n identifying families they must meet on their daily rounds. In 1972, Beck asked each of these workers to transfer the informa-tion they had collected i n the area under their jurisdiction on to large recording sheets, with the permission of Mr, Krishnamoorthy, the office head. The only problem that such data posed was the fact that the areas from which the information comes do not exactly coincide with the adminis-trative areas of the city as described by other government sources. As a result, there was a mine of extra caste information on the s a t e l l i t e munici-pality of Singanallur, but l i t t l e data on 3ome more central areas such as Race Course Road, I t was also very d i f f i c u l t to separate off the unnecess-ary data pertaining to the unmapped Singanallur sector of the c i t y . Since the only way to circumvent such d i f f i c u l t i e s was by mapping the available information on the identifiable streets of the map provided by Mr, Sundaram, a decision had to be taken to largely ignore those parts of the data obtained from the Malaria Office that could not be mapped. Out of the t o t a l number of 95,228 households that were described by the Malaria workers this l e f t 54,902 that could be spatially located. We decided that this would provide an adequate basis on which to proceed with the study. The 54,902 households thus selected w i l l , from now on, be referred to as the sample data from 1972» This sample, however, did represent both central as well as periph-eral sectors of Coimbatore c i t y . I t also included a l l the early extensions of Coimbatore such as R.S. Puram, Devangapet, and North Coimbatore. For analytic purposes, the city was divided into the following areas (See Map 1). I. the old core I I . the old periphery I I I . the new core IV. the older suburb (R.S. Puram) V. the older suburb (Devangapet) VI. the industrial sector of North Coimbatore VII. the new suburbs of North Coimbatore VIII. the new peripheries (west side extensions of the old core). Throughout this thesis, this sectoral sub-classification w i l l be adhered to. But neither the map of 1972 nor the spatial location of castes on the map gives us any understanding of the historical growth of the c i t y . The diachronic aspect of this thesis i s provided by (a) an 1871 map of the city (obtained from the India Office Library i n London), (b) the Manual of the Coimbatore d i s t r i c t (1898), (c) the Imperial Gazeteer (1908), (d) a journal article entitled "The Growth of Modern Coimbatore" by R. Chettiar (1939) i n which a map of the city i s also provided. 5. (e) The Kongu Country by M. Arokiasamy (1956), (f) The Coimbatore District Gazeteer (1966), (g) Castes and Tribes of Southern India by Edgar Thurston (1909), and f i n a l l y , (h) Peasant Society i n Konku by B.E.F. Beck (1972). In order to acquaint the reader with the above sources, a brief introduction i s essential. The Manual of the Coimbatore District (1898), and the Imperial Gazeteer (1908) are valuable, detailed sources of information about the various d i s t r i c t s , compiled by the colonial officers i n the last decades of the 19th century and the early years of the 20th century. In the years following Independence much of the above material was outdated and inadequate. Since the importance of such material was f e l t , the Records Office i n each of the states was given the responsibility of alterations and additions. The Coimbatore District Gazeteer (1966) i s an updated, revised version of the earlier records. The history of the Kongu Desa, recorded by M. Arokiasamy (1956), gives us an insight into the growth of the c i t y u n t i l the coming of the Bri t i s h . A further article by R. Chettiar (1939) on the growth of Modern Coimbatore gives us a bird's eye view of the historical and geographical factors that affected the growth of the city i n the Colonial Era. Finally, Edgar Thurston's (1909) monumental work on the ethnography of South India i s a comprehensive collection of data on the castes and tribes. In conjunction with the detailed ethnographic information of the Konku region recorded by B.E.F. Beck (1972), the above information has proved invaluable to our understanding of the social organization of the ci t y . With this as sufficient background, we w i l l launch into our study of the spatial layout of Coimbatore c i t y . In the next chapter we w i l l deal b r i e f l y with the history of urbanization i n India, and then examine a few models that are useful to our theoretical approach. In the third chapter we w i l l trace the history of the growth of the ci t y until the present time. This historical base w i l l make our analysis of the 1972 data i n the fourth chapter more significant and meaningful. Finally, i n the f i f t h chapter we w i l l summarize our conclusions, and offer a few generalizations. Chapter 2 TOWARDS A MODEL FOR THE CITY OF COIMBATORE Urbanization i n India i s not a recent phenomena. It has i t s roots i n the distant past. The earliest urban centres known to have existed are Harappa and Mohenjadaro (V. Smith 1958: 28). Unfortunately, l i t t l e i s known of the social and economic organization of these cit i e s because the few written records that have survived have not yet been deciphered. How-ever, archaeological excavations at the f i r s t site reveal certain important spatial patterns. The most distinctive feature of the c i t y of Harappa i s what remains of a citadel made of baked clay and mud. This fortress i s quite imposing i n size and extent. Inside the citadel towards the centre archaeologists have found a large bath, by the side of which were buildings for the use of r e l -igious leaders. Closer to the entrance of the citadel was a granary, where grain was stored. Well beyond the walls of the citadel l i e the workmen's quarters (V. Smith 1958: 29). What i s striking about this geometrical spatial pattern i s i t s close resemblance to other Indian towns known from a later period. As early as 300 B.C. Kautilya, a b r i l l i a n t minister i n the court of Chandragupta Maurya, wrote a treatise on statecraft. In this work he instructed the king on the ideal spatial layout of a c i t y . According to Kautilya, a royal c i t y was to be b u i l t on a square or rectangular grid and to have streets running i n both north-south and east-west directions. In the centre of such a ci t y was to be located the king's palace, a temple and/or a tank. Radiating from this nodal point were to be special streets allotted for the different castes according to their rank. 8 . The burial ground for the high caste people was situated i n the south side of the city, while the burial ground for other castes could be situated i n the north or the east. Even beyond the burial ground was a place where the Heretics and Chandalas (untouchables) had to l i v e (Dutt 1905: 151). By and large, ancient c i t i e s such as Madurai, Kanauj, Pataliputra, and Pukar were examples of these spatial principles stipulated i n Kautilya's Arthasashtras. Consider, for instance, the following graphic description of the c i t y of Puhar i n the Silapadikaram: "At the centre of the city were the wide royal street, the street of temple cars, the bazaar and the main street where rich merchants had their mansions with high towers. There was a street for priests, one for doctors, one for astrolo-gers, and one for peasants. In a wide passage, lived the craftsmen who pierce gems and pearls for the jewellers. In another quarter lived the coachmen, bards, dancers, astronom-ers, clowns, prostitutes, f l o r i s t s , betel sellers, servants and acrobats™ (A. Danielou 1965: 19)• The social history of the Indian people clearly indicates that such patterns of spatial differentiation were not just a feature of c i t i e s of the earlier period, but were commonly perpetuated by Islamic rulers of a later time (Ahmed & Spate 1950). What was remarkable about the moghul c i t i e s was that although Islamic style and taste greatly coloured their spatial charac-ter, moghul additions did not ravage or simply replace the older, trad i -tional forms. Indeed, i t was the policy of a few rulers li k e Akbar to ac-ti v e l y preserve a city's original format. Thus, the famous c i t i e s of Agra and Delhi have to this day retained some aspects of their traditional spatial layout (Ahmed & Spate 1950). With the advent of European traders, however, the major coastal c i t i e s expressing some features of western urban spatial organization began to make an appearance i n India. European trading posts usually grew around fishing villages. At the time of these foreigners' a r r i v a l the great c i t i e s 9. of Calcutta, Bombay, and Madras were only small settlements. As i n the Moslem era, the early period of colonial rule did not witness the destruc-tion of traditional spatial forms. In fact, colonial settlers tended to simply extend or build around the places that they conquered, and new prin-ciples of spatial organization were applied only to these marginal areas. As a result, i t i s s t i l l possible to distinguish, i n most colonial c i t i e s , the old city nucleus from newer British additions. "What was char-acteristic of these new areas was their spatial organization i n terms of economic c r i t e r i a rather than with regard to local ethnic or religious fac-tors. In B r i t i s h planning of such new urban areas the administrative and financial units were to be located at the centrej the surrounding zones then became residential and recreational areas (Smailes 1969). Thus, even after 13>0 years of European influence, Indian c i t i e s did not completely lose their traditional spatial forms: the old and the new were merely juxtaposed. Since i t i s the purpose of this study to analyze the spatial struc-ture of a post-colonial Indian c i t y , the following sections w i l l examine various models that may be useful i n formulating our approach. MODELS USEFUL IN THE STUDY OF URBAN CENTRES In traditional anthropological work on the spatial organization of human settlements i n India the unit of analysis has been the single v i l l -age. I t has been argued that within such a small territory i t i s possible to understand the intricacy of the traditional caste system (with the excep-tion of such factors as migration (Marriottl955: 1?1). But such studies have now raised at least as many questions as they once succeeded i n answer-ing. For a start, researchers have repeatedly asked how does one deal with 10. those aspects of organization that link a village to other social l o c i that clearly l i e beyond i t s bounds? And secondly, i t i s problematic whether a village study can adequately depict the patterns and social composition characteristic of a region as a whole. Thirdly, i t i s d i f f i c u l t to study the basic lines of economic and p o l i t i c a l dominance when one only looks at village phenomena. As Singer (1972) notes, i n some respects and under cer-tain conditions, the village (is) an organized unity with which a villa g e r identified^ but the numerous extensions of a village embedded i t inextric-ably i n a wider society and culture (1972: 258). More recent work has thus begun to focus on more extended spatial areas, and on the understanding of social processes as may only emerge clearly at higher levels of t e r r i t o r i a l organization. The Village Model Village studies since the 1950's (Gough I960, Srinivas 19$$, Gohn 1955, Marriott1955, Mencher 1972, and Beck 1972), to mention just a few, have, however, documented the existence of certain basic spatial forms that pervade lo c a l level organization. I t has been found, for example, that castes higher up i n the social scale tend to occupy the central and more prestigious areas of a village, while lower castes and untouchables occupy the more peripheral and non-prestigious space. Furthermore, since r i t u a l purity and social importance correlate well with high caste status, they easily become an attribute of that central space i t s e l f . Ritual neutrality similarly becomes characteristic of the semi-peripheral spaces where commer-c i a l enterprises are located. And r i t u a l pollution, f i n a l l y , becomes a quality of the extreme periphery of the village s i t e , the place where the untouchables l i v e . 11. In Beck's (1972) study of the Glappalayam settlement, such spatial expressions of social hierarchy involve yet one further dimension, which she has termed the Right/Left caste contrast (see Map 2). Here the econom-i c a l l y powerful 'right castes,' who are not necessarily r i t u a l l y very pure, are found i n the middle of the settlement; while the 'le f t groups' who are more particular about r i t u a l matters, but economically less powerful, can be found i n a single northern section close to the r i t u a l l y pure Brahmin quarters. The various touchable service castes (such as the potters, barb-ers, and washermen) are further separated spatially from a l l three of the above groups. They reside well to the south. Another area, the commercial strip surrounding the main settlement, has a large number of shops and a few houses belonging to various commercial castes. S t i l l further outside the settlement i t s e l f , quite by themselves, are the hamlets of various un-touchable groups. Having looked at the spatial organization of a village, let us now extend this analysis one step further to the c i t y . The City Model An increasing complexity i s necessarily introduced when settlements become very large. Such complexity i s greatest of a l l when we deal with a ci t y . Although city and town layouts i n India are based on similar prin-ciples as those which govern village organization, the cit y i s not just a mere spatial extension of a village unit. In addition to being a sub-system within a region, a c i t y i s also a 'hinge centre 1 (Cohn & Marriott19^8) that links up a local tradition to a greater tradition. Several studies hypo-thesizing the c i t y as a sub-system within a region have been conducted. In Singh's (1969) study of Bangalore he argues that "every ci t y , whether big or small, forms the nucleus of i t s region from which i t draws raw materials 12. and/or supplies the necessities" (1969: 82). In a similar vein, Janaki and Sayed (1962) (Quoted by Ginsberg 1969: Ulij.) have documented the importance of caste and religion to the town of Padra. A l l these studies have shown that c i t i e s and towns are linked to other c i t i e s and towns i n the region by complex networks of social and economic relations based on caste kinship or occupation. Further, cities also serve as focal points which f i r s t experi-ence the impact of outside forces. They, as Hoselitz remarks, "exhibit a s p i r i t different from that of the countryside. They are the crucial places i n under-developed countries i n which the adaptation to new ways, new tech-nologies, new consumption and production patterns, and new social i n s t i t u -tions i s achieved" (1969: 23). Because of this constant pressure to accom-' modate, the demarcation of such s t r i c t l y segregated social areas within a city i s more problematic. Hazlehurst's (1970) study of Puranapur attempts to grapple with this problem. In the following map (Map 3) Hazlehurst has abstracted some of the main principles of his analysis. The centre of Puranapur i s divided into r i t u a l l y pure and r i t u a l l y neutral sections. On the boundaries of this core are found the r i t u a l l y polluting areas. Distinct from these two areas are the sectors where the lower castes and untouchables l i v e . These latter sectors are also separated on the basis of social rank. Close to the untouchable settlements are the burial ground and the slaughter house. Each of these nesting tiers of soc-i a l space, i n turn, reflects the corporate or collective status of the residents associated with i t . While Hazlehurst provides a useful model that can be applied to Coimbatore city, his analysis does have certain limitations for our use. His i s a f a i r l y simplified and abstract picture; i t does not, for instance, take into account the fact that c i t i e s can have more than one centre. 13. Purariapur, i n spite of being an administrative unit of the B r i t i s h since the mid 19th century, does hot seem to be i n re a l i t y two c i t i e s — t h e old and the new adjacent to each other. This i s particularly true of colonial c i t i e s . When there i s more than one centre i n a city, demarcation of social areas along the lines suggested by Hazlehurst becomes problematic. Also, his model does not provide for drawing a distinction between spaces occupied by dominant castes who may or may not be r i t u a l l y pure, and the r i t u a l l y pure who may not be dominant. Without this added distinction we w i l l be hard pressed to analyze a distinctly south Indian ci t y l i k e Coimbatore. The importance of such a distinction w i l l become much clearer when we extend our unit of analysis s t i l l further to include the entire region. The Regional Model The village, town, and c i t y are sub-systems of a larger regional system and thus may be expected to ref l e c t , i n miniature, certain larger spatial and social structures. In order to test this idea that there may be a correspondence i n spatial layout, an analysis of the social organiza-tion of the region becomes essential. For our purposes, the region of Konku, i n which the c i t y of Coimba-tore i s situated, w i l l be treated as a model. The Konku region does not co-incide with the administrative division of the Coimbatore d i s t r i c t . However, as Beck (1972: 57) notes, the two geographical entities approximate each, other quite closely. The rural areas of Konku are marked by a large number of settlements or villages which depend on agricultural a c t i v i t i e s for their maintenance. The dominant caste i n rural areas i s therefore a land-owning caste, the Kavuntars. This caste owns 30 percent of the land. This position of 14. economic supremacy also gives that caste great social importance and allows i t to dominate many local level a c t i v i t i e s . The Kavuntars have for a l l i e s a group of castes that are occupation-a l l y dependent on them. Such castes were traditionally lumped together as belonging to the 'right division' of society. Groups who are not economic-a l l y dependent, and who vied for equal status with members of the right bloc, constituted a separate category symbolically labelled as " l e f t castes." Left division castes often try to emulate Brahmin manners and customs, but right division castes differ markedly from the Brahmins i n their value em-phases. Many references to these two divisions can be found i n the accounts of the l°th century travellers, though by now this bifurcation of social groups has become almost a mute issue. Certain other castes such as the Brahmins and the P i l l a i belonged to neither division but remained neutral and served as a symbolic head for the whole. Just as i n the v i l l a g e , the social organization of the Konku region exhibits a certain spatial patterning. The Kavuntars, who are the dominant caste, and head of the right bloc, are found i n nearly every major settle-ment. Members of this caste are especially densely settled i n the ecologic-a l l y defined centre of the region, which i s also their socially defined territory. The l e f t division castes, on the other hand, have a tendency to be found i n large numbers close to the region's commercial centres. Immi-grant groups and less powerful land-owning castes are found most heavily concentrated around the region's boundaries (Beck 1974). I t remains to be seen how far the c i t y of Coimbatore f i t s into this larger regional structure. Prom the above discussion some general conclusions can be drawn on the nature of spatial organization, be i t the village, c i t y , or region. 15. Salient Features of the Supplementary Model Some of the core contrasts that have emerged i n relation to the use of space i n the above studies are: (a) central areas/peripheral areas, (b) pure areas/polluting areas, (c) old core of the city/new core of the c i t y , (d) areas associated with social dominance/areas not associated with dominance (Right/Left), (e) commercial areas/residential areas. In addition to the above principles one further dimension was found essential i n providing a dynamic framework for the study of the spatial structure of the city, i.e., the temporal one. Without the time factor the study of urban centres becomes intractible. An historical perspective to the study of Coimbatore c i t y w i l l highlight some of the morphological addi-tions or omissions that have coloured i t s growth. As Hall (1968) has i n d i -cated, time and space are complimentary dimensions. We w i l l see this come true i n our study of Coimbatore c i t y . In the course of applying these principles to Coimbatore city, a number of questions have been raised. Is there more than one centre i n the city? Do each of these centres reflect particular social values? Or more specifically, which castes occupy these centres and which of them l i v e at i t s peripheries? Are these centrally located castes r i t u a l l y pure? If not do they economically dominate city l i f e ? Can such areas be described i n terms of the presence of Right and Left division castes as i s true for the region as a whole? Since these concepts can assume several shades of mean-ing we w i l l deal with each of them separately so as to be sure of what each means i n the context of our study. 16. Central areas/peripheral areas. In the ancient towns or ci t i e s of India religious, p o l i t i c a l , and administrative edifices occupied central space. Areas of economic activity were only secondary, and adjacent to these domestic institutions. Furthermore, the centre of the ci t y was char-acterized by a few broad streets (or Vidhi-Raja Vidhi) that made access to these central spaces easier, unlike the surrounding narrow winding streets. In Coimbatore the original nucleus had i t s sacred centre, and i n close proximity a commercial centre had i t s small beginnings. But with colonial rule, a new centre around which the ci t y developed was introduced. Thus, histo r i c a l l y speaking, Coimbatore has at least two distinct centres. They w i l l be referred to as the old core and the new core of the c i t y . In each of these centres a hierarchy of space can be discerned and each of these spaces given a particular social value. Pure areas/polluting areas. The difference between centre/periphery i s closely related to another contrast, that between purity and pollution. Most previous studies* i n South Asia have suggested that spatial separation i s commonly made to express concepts of social hierarchy. In the context of this study we w i l l use pure/impure to define the qualitative differences assigned to residential areas where Brahmins and untouchables are found. Areas associated with dominance/areas not associated (Right/Left). A dominant caste, once established, tends to expand i t s domain. This may be as Beck (1974) points out, an ecologically defined territory, or a soci-a l l y defined territory (1974: 5). But power need not always be an attribute of Brahmins or of other r i t u a l l y pure castes, as we noted ear l i e r . In the ^Particularly see Dumont i n Homo Hierarchicus (1969). 17. case of Kongu, and perhaps of several other agriculturally based regions, power has come to rest i n the hands of a numerically strong landowning cbmra-unity, yet not one that i s r i t u a l l y supreme. In this context, then, domin-ance i s used to mean the same thing as i t does i n the region as a whole. " . . . to mean near monopoly of management rights i n l o c a l resources (usually agricultural land) and considerable strength vis-a-vis an otherwise fragmented society" (Beck 1972). The dominance by a single caste of an entire region finds spatial express-ion i n the occupation of i t s centre by an economically powerful, manager-i a l l y oriented group. However, since agriculturally based power i s not par-t i c u l a r l y important i n an urban context, i t i s not clear that dominance w i l l take the same form i n urban areas. In this study, then, only the spatial separation of right and l e f t castes i s discussed. Urban power does not appear to be so caste specific. Commercial/Residential. This contrast relates to our earlier dis-cussion of the occupation of central space and i t s use for commercial as opposed to residential or r i t u a l purposes. In India i t is d i f f i c u l t to differentiate commercial from residential use of land. Even the predomin-antly commercial streets have a large number of residences on them. Perhaps what could be contrasted i s the more or less purely resident areas against the mixed residential and commercial functions. The f i r s t pattern i s the colonial one (a suburban trend) while the second is indigenous to the sub-continent from an early period. Finally, since the time dimension i s important to complete the study of an Indian city's spatial organization just as i t i s for a village or a region, we must next discuss the h i s t o r i c a l factors that have shaped Coimba-tore !s growth. The next chapter w i l l therefore deal with the early history of this c i t y . Chapter 3 HISTORY OF THE GROWTH OF THE CITY "If we would lay a new foundation for urban l i f e , we must understand the historic nature of the c i t y , and distinguish between i t s original functions; those that have emerged from i t and those which may s t i l l be called forth." — Lewis Mumford The word city i n Tamil has several synonyms. They are, according to the Dravidian Etymological Dictionary, Nakar (235), Fattinam (300), Padi (272), Pari (27U), Poral (300), Vity (369), and. Ur (57) (Burrow and Emeneau 1961). Of particular significance to this study i s the word 'ur' because as we shall see, the word Coimbatore i s actually the union of two other words—Kovan + Puthur. The prefix 'Puthu' i n Tamil means 'new.' So taken together, (Puthu + Ur) would make 'Puthur' which means 'new town.' Thus, Kovan Puthur refers to the 'new town of Kovan' or simply, 'Kovan's town.' Ramachandra Chettiar (1936: 102) relates a very interesting anec-dote regarding the origin of the name of the c i t y . "The Boluvampetty valley" when he writes: " . . . from which the river Noyal rises and flows eastwards, was a thick forest clad tract about a 1000 years ago. On the northern bank of that river i n a forest village, an Irula chief named Kovan lived. About the beginning of the 9th century a Siva chariyal with the help of his friend the Chera chief, travelled through these parts. Both being highly enthusiastic about religion and c i v i l i z a t i o n , the jungle was cut down; three temples were b u i l t , and a small fort and Petta constructed. In memory of the Irula chief, the place was known as Kovan Puthur . . . ." (Chettiar 1936: 102). From such small beginnings the modern ci t y of Coimbatore grew. ^Devotee of Lord Siva. 19 THE EARLY PERIOD: 1300-1759 Cooley's (I898) theory of transportation and Sjoberg's (i960) theory of political power seem applicable to the early growth of Coimbatore. Sit-uated at 11° N. and 76°58' E. on the left bank of the river Noyal and comm-anding the approach to the palghat gap, Coimbatore has from early times been a city of some strategic importance and military fame (Imperial Gazet-eer 1908: 371). The earliest mention of the city in inscriptional records stems from a period when i t was ruled by the Chera kings who, as we saw earlier, had established a settlement there. One other record in particu-lar suggests that Coimbatore was originally the settlement of Virakerallan-allur, named after the Chera king Virakerala (Arokiasamy 1956: 19U, 253). However, we not only lack evidence to support either hypothesis, but we also do not know how Coimbatore ranked with respect to the other strong-holds of the Chera Empire. During the next few centuries we do not even have records of the growth of the city, although we do know that the Kongu-desa in which Coimbatore is situated passed under the sway of various Indian powers such as the Rattasj the Gangas; the Cholas; the rulers of vxjayanagarj the Nayakasj and finally the Rajas of Mysore. With the excep-tion of being an important city under the latter, Coimbatore apparently did not play a vital role in the empires of this period. THE MOSLEM PERIOD: 1759-1799 According to R. Chettiar (1939) Hyder Ali and Tippu Sultan, who were the Rajas of Mysore, are said to have occupied the town on and off, and one building, now a hostel for girls of the secondary school, is said to have been the residence of the governors from Mysore and is pointed out 20. to be that which Tippu occupied when he visited the town (1939: 102). In addition, Coimbatore was also considered to be one of the most formidable forts held by the Moslems during the period of their rule. It was beseiged thrice during the Mysore Wars in 1768, 1783, and 1790 respectively until finally, in 1799, i t became an acquisition of the British (Chettiar 1939: 102). When Coimbatore passed into the hands of the British there was a remarkable alteration in the role the city played. Coimbatore, which was originally a military stronghold, was transformed into an administrative centre. Lewandoski (1975: 3^ 1-361) also offers a similar explanation in her study of the growth of Madras city. Coimbatore, with its mild climate and rainfall, became one of the most desirable stations in the Madras Presidency. A measure of its conven-ience can be gleaned from the following excerpt from the Imperial Gazeteer (1908: 372) which reads: "Situated at 1300 feet above the sea, in a pictur-esque position at the mouth of the Bolampetty valley, with the masses of the Nilgris and Annamalais rising into view on either side, its light rainfall of 22 inches and its moderate mean temperature, Coimbatore is at once healthy and pleasant." When Coimbatore became an administrative centre a number of changes were initiated. In the next few sections we will discuss some of the sal-ient features of colonial rule that had a direct impact on the growth of the city. THE FIRST HALF OF THE 19th CENTURY The Colonial period in many ways marked the beginnings of a revital-ization process. In 180J> Coimbatore was made the headquarters of the 21. d i s t r i c t , when the two earlier d i s t r i c t s with headquarters at Bhavani and Dharapurara were amalgamated into the single d i s t r i c t of Coimbatore (Chettiar 1939: 102). I t became the headquarters of the D i s t r i c t Munsifs 1 court i n 1855 (Coimbatore Gazeteer 1966: 1*86), and i n i860 the locale of the police de-partment as well (Manual of the Coimbatore Di s t r i c t I898: 248). But i t was not un t i l the introduction of the regulation system of government, i n i860, that Coimbatore became a municipality. As Beals (1955) notes: "Before i860, administrators had regarded themselves as care-takers and avoided altering local customs. After i860, how-ever, they sought to make sweeping changes. These changes included a continued expansion of the authority of the state government at the expense of the local authority; expansion and improvement of transport and communication; development of manufacturing; trading and diffusion of the European system of education and welfare" (1955s 79-80). THE SECOND HALF OF THE 19th CENTURY A regular municipality was established i n Coimbatore i n 1865 after the passing of the Madras Towns' Improvement Act X (Coimbatore Gazeteer 1 1966: U69). This Act, as documented i n the above gazeteer, primarily orig-inated with the intention of the government to make the inhabitants bear as much as possible the charges of maintaining a police force i n towns. Sub-sequently, however, the funds raised were not only applicable to the expen-ses of the police, but also to the construction, repairing and cleaning of drains, the making and repairing of roads, the keeping of roads, streets and tanks clean, and doing such things as may be necessary for the preser-vation of the public health (Coimbatore Gazeteer 1966: I4.69). The municipal-i t y derived i t s income from a tax on houses and buildings, a tax on carriages and animals, fees for registration of carts, fees for licensing slaughter 22. houses, carstands and offensive trades, fines and contributions from the government (Coimbatore Gazeteer 1966: 470). I t spent a major portion of i t s income on the police and the rest on roads, wells and sanitation. The construction of the central j a i l sanctioned by the government i n 1862 i n the north eastern section of the city, with 175 acres of land attached to i t , was also completed i n 1868 (Manual of the Coimbatore D i s t r i c t I898: 252). In 1871 there was a slight alteration i n the duties of the municipal-i t y , with the passing of the Towns Improvement Act I I I . Now the municipal-i t y had no longer to contribute to police charges, but instead had four new additional charges, namely those for hospitals and dispensaries, those for schools, those for birth and death registration, and those for vaccination (Coimbatore Gazeteer 1966: 471). The c i t y , i n the meantime, had also become the seat of a strong missionary movement and contained the cathedral of the Bishop of the French Societe" des Missions Etrangeres, as well as centres for the Leipzig Evan-gelical Lutheran Missions, which were also working i n the d i s t r i c t . These missions further took an active part i n educating the natives and set up several schools and colleges. The chief educational institutions were Saint Michael's College and Coimbatore College. The former was begun i n i860 by the French Mission and subsequently, i n 1891, a f f i l i a t e d to the Madras University (Manual of the Coimbatore Di s t r i c t I898: 126). The latter was established i n 1852 by Mr. E.B. Thomas, the then collector of the dist-r i c t (Manual of the Coimbatore Di s t r i c t I898: 125). These foreign structures were very prominent on the traditional landscape and Coimbatore, like other colonial c i t i e s , began to express this duality. The Map of 1871 highlights some of these opposing features quite well. In the following sections, therefore, we w i l l take a look at them (see Map U). 23. What i s most s t r i k i n g about t h i s map i s the clear separation of the Pre-Industrial and the new c o l o n i a l areas of the c i t y . And each of these areas of the map display d i s t i n c t i v e c h a r a c t e r i s t i c s . I n sura, while the s p a t i a l organization of the o r i g i n a l nucleus of the c i t y i s based on c r i t e r i a such as caste and r e l i g i o n , the s p a t i a l organization of the colon-i a l settlement was based on economic and administrative c r i t e r i a . The Original Nucleus ) What were the different castes and communities i n the c i t y i n 1871 and how were they distributed? A careful perusal of street names suggests the presence of a number of castes and r e l i g i o u s groups at t h i s time i n the old core of the c i t y and i n i t s new peripheries. These groups are l i s t e d and c l a s s i f i e d i n Table I . The c l a s s i f i c a t i o n i s based on the assumption that streets acquired the caste-names of the dominant castes l i v i n g on them. The categories used i n the c l a s s i f i c a t i o n were borrowed from Beck (1972), with modifications and additions wherever necessary. Thus, f o r instance, while Beck c l a s s i f i e d service castes as belonging to either the l e f t or Eight d i v i s i o n , i t was not possible to do so with the data on ser-vice castes and untouchables i n the c i t y . Castes i d e n t i f i e d as belonging to d i s t r i c t s outside Kongu were c l a s s i f i e d as belonging to outside regions. People i d e n t i f i e d as non-Hindu were c l a s s i f i e d as "outside r e l i g i o n s , n Streets that were named after the predominant trade were also c l a s s i f i e d . F i n a l l y , streets that were named a f t e r the King or the Colonial Officer were also i d e n t i f i e d . They w i l l be referred to as ' P o l i t i c a l Streets' i n column (9) of Table I , TABLE I: Classification of Street Names of the Map of 1 8 7 1 on the Basis of  Ca3te, Community, Commercial and P o l i t i c a l Identity ( 1 ) Neutral (2) Left (3) Right (U) Service (5) Untouchable (6) Immigrants (7) Outside Religions ( 8 ) Commercial ( 9 ) P o l i t i c a l Telugu Brahmin (Priests) V, Mudaliar Street (Profession-als) M. Nanjappa Gounder Street K. Gounder Street R. Gounder Street (Agricultur-a l land-lords) Idayar Street (Milkmen) Ottaisekar Street ( O i l m i l l workers) vannan Street (Washermen) Potters Lane O i l Mongers Street Chakkilayar huts Paracheri Oppara Street (Salt manufac-turers) Oppanagara Street (Telugu trad-ers + agri-culturalists) Odde Street (Construction workers who speak Telugu from Orissa) Okkilier Street (Canarese cul-tivation) Torayar Street (Canarese fishermen) Lingappa Cetti Street Komutti Cetti Street (Telugu trad-ers) Moslem Lane Pattuno olkaran Ismael Rawther Lane (Moslem Christian Lane Big Bazaar Street Fish Market Raja Street Sheristadar Street Sullivan Street 25. This classification produces several broad patterns. These have been mapped in condensed fashion (see Map 5) i n order to add c l a r i t y to the discussion which follows. Central Areas/Peripheral Areas One striking feature of the original settlement i s the organiza-tion of central space. As the map indicates, the centre of the original nucleus housed a few temples, the administrative units such as the Taluq 1 2 Cutacheri, the Chavadi, and the King's Palace. Radiating from this centre were the separate streets for the different castes. 'And on the peripheries were found the areas inhabited by the lower castes and the untouchables. One other feature of most Indian c i t i e s was a tendency for there to be a certain spatial segregation based on religious c r i t e r i a . This i s also illustrated clearly by the Map of 1871 where we find that Moslems and Christians located i n specific residential areas. Christianity, having arrived on the Indian scene later than Islam (with the exception of Saint Thomas and the Syrian Christian Missionaries), was relegated to the outer-most periphery of social space. Moslems, on the other hand, were accommo-dated within the original s i t e . In spite of being located i n the main settlement, Moslems by no means occupied central space. Instead, they were pushed outwards towards the peripheries, closer to the semi-commercial and polluting areas. Thus the mosques were situated near the Fish Bazaar, the slaughter house, and the f o r t . This finding, incidentally, f i t s well with The central government of the Raja of Mysore was organized more or less like a secretariat containing the various departments of which 18 cutacheris are seen i n the o f f i c i a l records of the time (Arokiasamy 1956: 39ljl Subdivisions within the department (Arokiasamy 1956: 396). 26. Fox's (1969) view of Moslems occupying a lower status i n the last century than do many of the lower castes i n society today. In'contrast, the central space of the B r i t i s h settlement was pre-dominantly office (commercial) space, and that i n turn was surrounded by religious, recreational and residential spaces respectively. This i s not surprising since land uses i n the industrial c i t i e s of the West were qualitatively different from those of the pre-industrial world. JPure/Polluting In terms of areas that are 'pure or polluting' there seems to be l i t t l e ambiguity. According to our earlier definition, areas occupied by Brahmins w i l l be considered pure and areas occupied by untouchables pollut-ing. And these polluting areas can be further subdivided. Note that the shoemaker (Chakkilayar - the lowest ranking group among the untouchables) i s found i n the most inauspicious and non-prestigious space i n the south which i s also associated with death (near the slaughter-house), while the Paraiyar (drummers - a slightly higher-ranking group) are found i n the eastern peripheries close to the f i s h market. This whole scheme of "out-side settlements" f i t s i n with the rules stipulated i n the Arthasastras where the untouchables were expected to li v e beyond the c i t y or village boundaries altogether. (See Map 6). What i s perhaps most distinct about the B r i t i s h settlement areas was their demarcation of peripheral space by t o l l gates and their use of peripheral space i n the north and south for burial purposes. Further, they improved the old peripheral (polluting) settlements adjacent to their site by constructing latrines. 27. Commercial/Residential If residential segregation on the basis of caste was a crucial t r a i t of the c i t y of Coimbatore i n 1871, another equally important element was the relative lack of such segregation with respect to land uses for economic purposes. Although we have no direct information regarding the commercial organization of the city i n 1871, by identifying two streets named after the commercial establishments of the streets (Big Bazaar Street and Fish Market Street) we do have some measure of i t s commercial importance. Also, by looking at the nature of i t s commercial establish-ments today and at the density of population (both of these w i l l be dealt with i n more detail i n the next chapter), we can say that segregation of residential and commercial land was not the order of the day. (See Map 7). Right/Left Caste The location of castes belong to the Right/Left and neutral d i v i -sions can also be identified on the map. The Higher Right, the Higher Left, and the Brahmins a l l have very prominent positions i n the core of the city, although the former two are more widely spread than the la t t e r . On the other hand, i n the British core there i s a conspicuous absence of caste names on the different streets. Conclusion From the above discussion, some general observations on the use of space i n Coimbatore city i n 1871 can be made. F i r s t , central space i n the original nucleus i s associated with r i t u a l purity, as well as r i t u a l neut-r a l i t y . As a result, central space i s predominantly occupied by the higher Left, Right, and neutral castes. ' 28. Peripheral space, on the other hand, i s polluting and non-prestig-ious space, occupied by the untouchables. Finally, commercial space i s not clearly differentiated from residential space. The spatial organization of the colonial settlement was based oh Western models. Underlying these specific principles, therefore, were certain new social values. Within the next 30 years a great many additions and alterations were made to Coimbatore and the city began to take on a new look. In 1884 the Madras Municipalities Act IV was passed, which for the f i r s t time introduced the term 'municipality' i n the t i t l e and adopted the new terms 'council' and 'councillors* instead of 'commission' and 'comm-issioners' employed i n the earlier acts. By this Act, 75 percent of the councillors were to be elected by the taxpayers. Coimbatore came to have 20 councillors of whom 15 were elected and 5 were nominated (Coimbatore Gazeteer 1966: 4). Owing to the recurrence of plagues i n the d i s t r i c t , the municipal-i t y began to pay attention to public health problems. One more hospital was founded i n 1888 and funded by the municipality (Manual of the Coimba-tore D i s t r i c t I898: 108). Further, i n 1892, a depot for the production of animal lymph was opened at Coimbatore under the immediate control of the Distr i c t medical and sanitary officer, and between 1888-1893 alone a tot a l of 7822 males and females were vaccinated (Manual of the Coimbatore Dist-r i c t 1898: 113, 115). In the process of such active institutional and spatial growth, transport and communication systems were improved. There were now four major state highways and one national highway, a l l of which converged on the c i t y . The town which had only 27 miles and 1 furlong of road i n 1866, 29. had 35 miles and 1 furlong of road by the turn of the century (Chettiar 1939: 105). Further, i n 1873, a railway line connecting Coimbatore with other urban centres i n the d i s t r i c t as well as the surrounding regions was i n -stalled. But i t was not until 1939 that Coimbatore became a junction of the South Indian Railway (Chettiar 1939: 10). Improvement of transport and communication had opened up Coimbatore to the population of the surrounding villages who sought employment i n the c i t y . By the turn of the century there was already a steam cotton press, a cotton spinning m i l l , a tannery, two steam coffee curing works, and a steam factory. They were a l l owned by the B r i t i s h . In addition there was a d i s t i l l e r y and sugar factory owned by the Indians (Imperial Gazeteer 1908: 373). According to the Manual of the Coimbatore Di s t r i c t (1898: 167) there were 4,169 landholders, 1,821 agricultural labourers, 4,100 other labourers, 10,826 traders, 5,954 weavers, 4,834 other artisans and 14,769 classified as 'others' employed i n the c i t y . Although there were a large number of traders, artisans, labourers and weavers, Coimbatore was not yet an industrial centre. The textile industry, for which Coimbatore has be-come well known, had hardly begun although experimental attempts at i n -creasing the output and production of cloth had been made by the B r i t i s h as early as 1842. Dr. Dwight, who established the experimental farm at Coimbatore, demonstrated that American cotton would grow very well i n the region. This was not the only farm i n operation at that time i n India. Perhaps i t could be said, however, that i t was the only farm that was an utter f a i l u r e . The peasants were not willing to grow American cotton, and more than that, they refused to use the cotton gin that was part of the experimental programme. They had several reasons for doing so: 30. (a) Indigenous cotton could not be cleaned by the American cotton gin. (b) And secondly, i t was not practical to grow American cotton because i t could not be cleaned by the indigenous gin. Although there were other reasons also cited (Leacock and Mandlebaum 1955s 33U-351) for the failure of the British i n controlling the cotton industry, the real impetus to the development of this industry was indigenous. I t could be said, however, that the British had l a i d the foundations upon which the cotton textile industry flourished. THE FIRST HALF OF THE 20th CENTURY At the turn of the century there were only two municipalities i n the d i s t r i c t , of which Coimbatore was the largest. Between 1915-1920, how-ever, four new municipalities were constituted. In 1920 the Madras Dist-r i c t Municipalities Act V was passed for repealing the Act of 1884 and for increasing the elected proportion of the members of the municipal council, as well as the resources and powers of the municipalities. In that year also an Inspector of the municipal council was appointed to supervise the workings of the municipalities. Any municipal council had to have 16 mem-bers i n municipalities with a population not exceeding 20,000; 20 i n those between 20,000 and 30,000; 2U i n those between 30,000 and 40,000; 28 i n those between 40,000 and 50,000; 32 i n those between 50,000 and 100,000; and 36 i n those exceeding 100,000 (Coimbatore Gazeteer I960: 475). Coimbatore had a total of 32 councillors, of which 75 percent were elected by the ratepayers and the others nominated by the government (Coimbatore Gazeteer 1966: U75). 31. The Dis t r i c t Municipalities Act of 1920 was further modified i n 1930 by Act X. This Act did away with the nominations and insisted that a l l councillors be elected. The government was authorized to secure u l t i -mate control over e l e c t r i c a l undertakings managed by the municipal councils and to appoint e l e c t r i c a l engineers. The changes that were made i n the local administration was increasingly giving the municipality more autonomy to guide i t s a f f a i r s . In the f i r s t three decades of the 20th century, a number of town extensions were made. Since the population had considerably increased, the authorities had to take drastic measures for the mitigation of insani-tation and congestion. By 1929 nine town extensions were made by which the outlying portions of the villages of Krishnarayapuram, Souripalayam, Puliakulam, and Ramanathapuram were added to the town (Chettiar 1939: 103). In 1866 the areas of the town was h»2 square miles and housed a population of 2l*,2Ul. By 1931, the population had increased to 95,198 and the area of the town had also increased to 7.5 square miles. According to R. Chettiar (1939: 103), by 1939 the population was over 100,000 and the area of the town incorporated 12 square miles. That Coimbatore was rapidly growing can be further adjudged by the gradual increase i n the number of houses and the tax paid on them. The number of houses rose from 2,000 i n 1800 to 9,283 i n 1915, and f i n a l l y to 13,71*2 i n 1939 (Chettiar 1939: 10U). Likewise, the property tax that was of the order of Rs 56,881; i n 1920, rose to Rs 106,1*23 i n 1931 and f i n a l l y to Rs 282,785 i n 1937 (Chettiar 1939: 10l*). By 1939, the municipality ran 18 elementary schools (with a to t a l of 5>205 pupils) to which were also attached weaving schools. A deaf and dumb school, likewise, was managed by the council. 32. In addition, there were also the agricultural college and forest college, which vi r t u a l l y constituted two independent colonies adjoining the c i t y . Industrial educational institutions such as Saint Joseph's and P.S.G., training schools, private schools of commerce, convents and orphan-ages also dotted the urban scene. The period commencing from the turn of the 20th century had wit-nessed rapid changes that were affecting the socio-economic and p o l i t i c a l structure. The Swadeshi1 Movement With the turn of the century the growing antagonism against the Brit i s h had erupted i n the form of an Indian National Movement. Although this movement had important Indian leaders l i k e Gandhi, i t also received a certain impetus from the activities of a minority of Europeans lik e Mrs. Annie Besant and Wedderburn. The lat t e r had, earlier i n his career, been a much respected collector at Coimbatore. His name i s s t i l l perpetuated by a street named after him i n that c i t y (Chettiar 1939: 103). Mrs. Besant had her stronghold i n Madras, where she founded the Theosophical College and conducted a number of "Home Rule Meetings." Coimbatore soon came into the picture, too, because Mrs. Besant made i t her home. As early as 1917, a Home Rule Conference was organized i n Coimbatore and had for i t s strong supporters a number of Brahmin lawyers i n the d i s t r i c t (Coimbatore Gazeteer 1966: 113). A number of demonstra-tions and meetings ensued, and soon the whole d i s t r i c t f e l t the impact of this new p o l i t i c a l activity. Soon the Home Rule Movement was enveloped Swadeshi' was the term used by Mahatma Gandhi to refer to the struggle for Independence. 33. into a larger movement called the Non-Cooperation Movement. This was largely due to the efforts of one man--Gandhi, who i n 1920 visited d i s t -r i c t s i n the south, of which Coimbatore was one. His v i s i t only caused these sentiments to multiply. T/Jhile a l l southern d i s t r i c t s joined i n this common cause, the seeds of divisive influences were already being sown. What emerged from this internal s t r i f e was the justice party headed by Mr. E.V. Ramasami Naicker and drawing a large number of supporters from the ranks and f i l e . The cause they valiantly strove for was the liberation of the lower castes from the influence of the Brahmins. This anti-Brahmin feeling took the form of a number of agitations and counter-agitations. In spite of this growing internal division, i t was not unt i l the beginning of the C i v i l Disobedience Movement that these agitations took a violent course. In January, 1932 one of the processions taken out by the Desa Bandu League of Tirupur was dispersed by a great deal of violence. The leader, S r i O.K.S.R. Kumarasami Mudaliar, popularly known as Tiruppur Kumaran, or Thiaygi Kumaran, was fat a l l y wounded (Coimbatore Gazeteer 1966: 122). His memory has been perpetuated i n not only a street named after him, but also a central market i n the heart of the original city nucleus (Thiaygi Kumaran Street, Thiaygi Kumaran Market). Also, as a response to the C i v i l Disobedience Movement, a number of hartals and strikes were staged by the millworkers. In September, 1937 there were two strikes i n mills near Coimbatore. In October a series of strikes began i n the Sarada Mi l l s , Coimbatore, and spread to eight other textile mills i n the c i t y . In February, 1939 a more serious strike began i n the Saroja M i l l s , Coimbatore. In July of the same year five textile mills i n Coimbatore and one i n Tirupur struck. In 1939 there was a strike 34. in the Kaleeswara Mills at Coimbatore that spread to other units (Coimba-tore Gazeteer 1966: 130). These strikes were organized by the leaders of the various political parties to achieve their ends. Thus, a few leaders like Mr. E.V.R. Naicker even wanted to make Coimbatore a stronghold for the Dravidians. These agitations did not cease even after the achievement of In-dependence. Political rivalries continued to make this city important as a seat of political power. Being an industrial city i t soon became an easy target for the communist party, and leaders such as Mohan Kumaramangalam and Aruna Asaf A l i held meetings in the city. Further, in I960, the comm-unist nominee was chosen as the mayor of Coimbatore. But before we enter into a discussion of the growth of the city in the post independence years we should pause and look at the map of 1939 in order to understand the spatial extensions and industrial locations of the city at that point in time (see Map 8). Although this map does not express the details of the spatial lay-out of the different castes in the city, as the map of 1871 does, i t s t i l l indicates the location of most of the important industries and institu-tions in the city. Also, i t indicates the directions and extent to which the city had grown. According to the map of 1939 there were $ cotton mills and 5 other ginning factories within the boundaries of the city. A number of offices and institutions established since the turn of the century also find a place ^ on the map. POST-INDEPENDENCE YEARS The years following Independence were years of planned change. In . 35. 1956 the Coimbatore municipality had 36 councillors. I t levied a property tax, a professions tax, a tax on vehicles and animals, a tax on carts, and also derived an income from surcharge on stamp duty, entertainment tax, motor vehicles tax, and government grants under several heads (Coimbatore Gazeteer 1966: 478). It maintained 83§ miles of road, l i t partly by elec-t r i c i t y and partly by kerosene. I t maintained 32 elementary schools with an average strength of 15,777 pupils as well as 5 high schools, 4 middle schools, one deaf and dumb school, one hostel, one library, and 24 reading rooms. I t ran 5 maternity and child welfare centres, 1 infectious diseases hospital, 9 dispensaries, treating 527,522 outpatients. I t had a f i r s t -class health officer, one assistant health officer, 2 selection grade sani-tary inspectors and 20 sanitary inspectors. It had 5 daily markets, 2 cart stands, and 2 slaughter houses (Coimbatore Gazeteer 1966: 478). Under the fiv e year plans, the cotton industry had also received a f i l l i p from the government. By i960 there were altogether 36 t e x t i l e mills within a radius of 7 miles from the heart of the town. The most im-portant of these are tabulated as follows with respect to the number of looms, spindles, and financial layout. (See Table I I ) . The f i r s t seven are within the municipal boundaries, but the other 7 are on the outskirts of the c i t y . This process of decentralization has had several advantages as well as disadvantages. While i t reduced the problem of commuting and overcrowding i n the city, i t also gave ri s e to a number of slums around the factories. .This historical sketch of the growth of the ci t y has brought us into the last phase of the 60's. At this point, therefore, we w i l l have to terminate our narration, for want of more detailed information, and look at some finer points of the spatial structure of Coimbatore. In the 3 6 . TABLE II: Classification of Textile Mills i n Coimbatore With  Respect to the Number of Looms, Number of  Spindles, and Financial Layout Mi l l s No. of Looms No. of Spindles Financial Layout (Rs) 1. Lakshmi 200 59,988 10,000,000 2. Srivenkateswara 268 47,444 5,000,000 3. CS. & W. Mills 407 72,832 2,500,000 4. Sivananda not mentioned 18,816 2,500,000 5. Pankaja M i l l n 30,790 2,500,000 6. Cambodia ti 38,110 2,500,000 7. Janardhana it 23,784 2,500,000 8. Kothari Mills -Singanallur 300 43,000 10,000,000 9. Vasantha Mills -Peela Madu 290 44,U24 3,000,000 10. Coimbatore Cotton M i l l s -Madhukarai - 30,829 2,500,000 11. Radhakrishna Mills -Tiruppur 300 51,664 3,000,000 12. Dhanalakshmi Mills -Tiruppur 201 34,708 2,500,000 13. Coimbatore Pioneer Mills -Singanallur - 24,466 2,500,000 14. Premier Mills -Udumalpet 20,736 2,500,000 (Coimbatore Gazeteer 1966: 322-323) 3 7 . next chapter, therefore, we w i l l see how the spatial organization of Coimbatore city reflects overarching social values that i s so much a part of traditional as well as modern Indian society. Chapter k THE ANATOMf OF COIMBATORE CITY In the last chapter we discussed how Coimbatore grew from a small settlement into a large industrial centre through a slow process of accre-tion. The historical data allowed us to divide the c i t y into I. an old core (the earliest main settlement), I I . an old. periphery (the outlying areas of the earliest settlement), III. the Bri t i s h core (the settlement area identified with Colonial rule), IV. an older suburb I (i . e . , R.S. Puram), V. an older suburb II (i. e . , Devangapet), VI. the industrial sector of North Coimbatore, VII. the new suburbs of North Coimbatore, and VIII. the new periphery (west side extensions of the old core). In this chapter the available caste data of the city i n 1972, the data on the spatial location of commercial concerns, as well as data per-taining to the location of business and residential telephones w i l l throw more light on the socio-economic and r i t u a l status of each of the above geographical sub-divisions. We w i l l therefore be i n a position to delin-eate areas of r e l i g i o - r i t u a l importance, such as the Devangapet sector (Brahmin enclave) which i s mainly a residential area. Areas of secular importance such as R.S. Puram, an adjacent, older, non-Brahmin suburb that combines commercial with residential land uses can also be identified. Further, i t w i l l allow us to distinguish traditional core areas of the city such as Selvapuram, that combines both r e l i g i o - r i t u a l and secular concerns. In striking contrast, are the peripheral industrial sectors of 39. North Coimbatore, which has the least r e l i g i o - r i t u a l and class s i g n i f i c -ance, as well as the adjacent cosmopolitan and newer suburbs of North Coimbatore. These distinct analytic categories do not approximate the administrative divisions through which the people perceive space i n Coimba-tore c i t y . We have made a clear distinction between the old core and the old periphery i n the 1972 map, for instance, while they are together re-ferred to as Selvapuram sector i n our original data. These various sect-ors make Coimbatore a 'mosaic of social worlds.' Three decades ago Wirth succinctly summarized the fundamental assumptions of ecological studies, noting that: " . . . diverse population elements (within the city) . . . tend to get segregated from one another i n the degree to which their requirements and modes of l i f e are incompatible with one another while persons of homogeneous status and needs unwittingly d r i f t into consciously select, or are forced by circumstances into the same area. The different parts of the c i t y thus acquire specialized functions " (Wirth 1938: 56). Now, while the history of Coimbatore can help to explain why cer-tain social worlds should be located where they are within the municipal boundaries, only a more detailed examination of the people who l i v e there can help illuminate the nature of these diverse social worlds that co-exist within the c i t y . For this purpose the various caste populations i n each of these sectors have been computed and combined i n such a way as to enable a broad comparison of the 'caste-composition' of these various sectors. To do thi s , each caste represented i n the 1972 sample was classified (in keep-ing with the regional classification) as belonging to one of the following categories: (a) neutral, (b) l e f t , (c) right, (d) service, (e) immigrant groups, (f) non-Hindu, (g) untouchables. In the regional classification, ho. the service castes and untouchables were classified as belonging to either the Left or Right division. Immigrant groups and non-Hindus are treated as distinct categories with respect to their numerical strength in the cities. Immigrant groups or castes in the city were further subdivided on the basis of numerical strength into North Indians and South Indians. Since the spatial patterning of a few South Indian castes was crucial to our understanding of the spatial layout of the city, they were given more importance than the others. They are the Naidus, Boyars, and the Malayalis. They are referred to as Immigrant Group I. The rest are referred to as other South Indian Immigrants II and North Indians. The unequal represen- • tation of these various caste categories in the several parts of the city described above forms the backbone of the following analysis. Since the purpose of this study is to analyze the extent to which the spatial dimensions of a city reflect overarching social values, we must devote a preliminary section to a discussion of the numerical repre-sentation of these various castes in the Kongu region as a whole.1 For this purpose, census caste data for the years 1901, 1911, and 1921 (after 1921 census data collection excluded caste identity) have been collected from the District Census Manuals, and an average for the three years com-puted (see Appendix I for caste details of each year). Although census data do not always prove to be accurate when exam-ined in detail, they will s t i l l serve our purpose when used in aggregate form. We will simply use the overall pattern of proportionate representa-tion at the regional level to highlight some of the salient features of the traditional population (see Table III). The Kongu region as Beck (1972) points out, is not quite the same as Coimbatore district, but the figures for the latter area represent the closest possible approximation under the circumstances. 41. TABLE I I I ; The Caste Composition of the Kongu Region3 Coiiribatore City 3 and the Sample City Population Kongu1 Coimbatore2 Sample City^ Castes (1) Region % (2) City % (3) Population % (4) Neutral Brahmin 1.95 6.2 9.46 P i l l a i — 4.42 4.82 1.95 10.62 U4.28 Left Asari 2.92 4.08 5.88 Cettiyar 2.85 11.86 14.30 Mudaliar 3.76 3.28 3.56 Naicker 0.48 2.70 2.71 10.01 21.92 26.45 Right 36.98 9.62 8.21 Kavuhtar Nadar 4.19 1.07 1.42 Tevar 0.14 3.02 0.78 Ul.31 13.71 10.41 Service Udaiyar 1.28 0.59 0.59 Vannan 1.54 1.46 1.75 Navithar 1,64 0.91 0.92 Pandaram 2.15 0.12 0.01 6.61 3.08 3.27 Immigrant I 1.57 7.43 4.34 Naidu Boyar 0.09 2.46 2.01 Malayali 0.12 8.41 8.62 1.78 18.30 14.97 Immigrant I I 2.84 0.58 Okkilier 1.00 Gowder — 1.93 1.72 Reddi 3.35 0.24 0.38 Arumthathiar — 1.02 1.56 Panikar — — — — Edayar — 1.60 1.65 Servai 0.01 0.01 0.01 Achangar — — — — Sivier — 0.15 0.25 Kollar mmmm 0.03 — Vaniyar 0.58 — — Ganga 0.01 0.01 0 Pachakerar — 0.01 0 Maratiyar — — 0.05 0.09 Vilayar — 0.04 0 Pannady — 0.11 0 Anglo-Indian — 0.20 0.04 Devanga 2.65 0 0 Konar 0.85 0 0 10.29 5".984 6.70 42 TABLE III: Continued Castes (1) Kongu Region % (2) Coimbatore' City % (3) Sample City- 5 Population % (4) Gurkha Punjabi Marwadi Sait Outside Religions Moslem Christian Untouchables (Castes) Pallar Valluvan Holeya Adidravida Paraiyan Chakkilayar Solaga Kadan Dommara Malasar Mudugan Kallar Kurumban Koraver Unknown (Tribes) 0.21  0.21 0.39 0.45 "oTBI 1.81; 0.22 0.21 3.99 11.04  17.30 0.17 0,02 0.07 0.07 0.03 0 0_ 5^735 17.66 0 0.071 0.03 0.62  0.721 7.50 11.17 4.79 0.08 0 0.53 2.57  7.99 0 0 0 0 0 0 0.13 0.19  0.32 6.06 8.31 0 0.06 0.06 1.07  1.19 11.15 4.20 u n 2.52 o.i4 0 0 1.59 n 0 0 0 0 0 0 0.22 0.09  0.31 J 4.56 2.62 100% 100% 100% Source: *The caste composition of the Kongu region i n column (2) Is the arithmatic average of the available caste data i n the census manuals for the years 1901, 1911, and 1921. 2+^The caste composition of the c i t y and sample was computed from the data obtained from the National Malaria Office at Coimbatore. 43. Column (2) of Table III summarizes the strength of particular castes i n the region i n the early decades of this century. There i s no reason to believe that these proportions have changed much i n rural areas since that time (Beck 1972). A number of important features emerge from this tabulation. F i r s t of a l l , i t il l u s t r a t e s several important differences i n pro-portional representation for the Right and Left caste groups respectively. For instance, i t i s evident that i n rural areas the Right division i s roughly twice the size of the Left (including the Chakkilayars). These groups comprise roughly about hi percent of the regional population, while the Left division constitutes only 10 percent. Of this hi percent, fur-thermore, more than two-thirds i s accounted for by the presence of a dom-inant agricultural group locally known as the Kavuntars. Unlike either of the above castes, the neutral castes ( i . e . , Brahmins and P i l l a i s ) are numerically a very small group constituting less than 2 percent of the t o t a l . The service castes i n the region ( i . e . , the cooks (Pandaram), potters (Utaiyar), washermen (Vannar), and barbers (Navithar) constitute almost 7 percent of the regional population. Of the remaining population of castes about 21 percent can be classified as recent immigrants to Kongu from elsewhere. They comprise a large number of castes and sub-castes that have come from a l l over India. However, the greatest numbers have been from regions that surround the Kongu ( i . e . , Andhra, Kerala), and from other d i s t r i c t s of Tamil Nadu. Another category, not mentioned above, which constitutes a small proportion, are the t r i b a l s . They account for another .36 percent. The Christian and Moslem community i n the region as a whole i s not very large (i.e., roughly 1 percent). In contrast to the two former groups, kk. the untouchables are a large group constituting a f u l l 17 percent of the whole. By comparing these figures to those for the city i n 1972 (column 3), some interesting observations can be made. Firs t of a l l , i t i s evid-ent that the neutral castes alone constitute almost 11 percent of the total number of households i n Coimbatore c i t y . The Right division castes, by contrast, contribute only Ik percent of the urban households while the Left division castes are now the largest category (22 percent). Note that whereas the Right division castes outnumbered the Left two to one i n the region, the Left outnumber the Right by roughly the same proportion i n the city ( i f we include Chakkilayars). Owing to a lack of information on the Right and Left caste a f f i l i a -tion of the various service castes i n our sample, they have been treated as a distinct category. This group forms 3 percent of a l l the households i n Coimbatore as contrasted to 6 percent i n Kongu region. We can conveniently divide the castes and communities i n Coimbatore that come from outside regions into two categories: those who have come from the north of India and the others who have come from elsewhere i n south India. Although the number of North. Indians i n the c i t y are negli-gible (1 percent) they s t i l l are an interesting subset because their pres-ence helps to explain certain nuances i n the spatial organization of the c i t y . Of the latter category three groups are strikingly numerous: the Naidus and Boyar (merchant castes from Andhra Pradesh) and the Malayalis (a l inguistic community from Kerala). They alone contribute to 18 percent of the household population of Coimbatore, as against their 2 percent i n the Kongu region. The other castes i n this category amount to approximately 6 percent. US. The Scheduled tribes i n Coimbatore are few and are found only i n the peripheral sector. Amongst the non-Hindu groups the Moslems have a particular impor-tance. They constitute about 8 percent of the households i n Coimbatore, while the Christians account for another k percent. In contrast to their proportion i n the region (1 percent), together they contribute to 12 per-cent of Coimbatore's household population. Finally, there are the 'untouchables' who are under-represented i n the urban setting as compared to the region (18 percent). Since most are manual labourers there i s more work for them i n the countryside than i n the c i t y . They contribute to 8 percent of our urban residents. The following bar diagrams (see Figure l ) summarize the contrasts between the caste population of Kongu and Coimbatore c i t y . Further, this comparison points up one important underlying principle: the p u l l the cit y has on those who have specialized marketable s k i l l s such as weavers and traders. Thus, castes who do not have a significant position i n the region ( i . e . , the non-agricultural castes) now assume a dominant role i n the c i t y . This contrast i s even more clearly illustrated by our sample popu-lation (Figure 1). Thus, Left division castes, neutral castes, and a selected number of castes and communities from outside regions dominate the significant sectors of the ci t y i n contrast to those who belong to the Right division. The details of the sample population w i l l be dealt with i n a later section that deals with the spatial distribution of these var-ious castes. Having compared the caste population of the region with that of the c i t y , we w i l l now turn our attention to some general aspects of 7H(k Cfitra e«ti**eine* ** THIS *««*U e«/<**7**<5 ciTy ,i. ..i. J.J.. Sample O -—-> r~~. '- J •H ;.j c to i> o I—i r_i C ' C H O hH • H d_' 'O C' .*} \> O - r l _p •'/} \.) •f-M •'"..J * , .~ • <*-i -P f—' 7* 'S; • H o * 7j 'J rj. !m .0 . c_;»- ' V O r..: 47. Coimbatore's resident population, as described i n social, demographic, and economic terms. Fi r s t of a l l , we w i l l discuss the general density of population i n Coimbatore. A rough measure1 of this was computed and mapped i n order to pinpoint the areas of high and of low concentration. Map 9 indicates that the large population centres i n the c i t y are located i n i t s old core and along i t s old periphery. As one moves out from this centre one can notice an appreciable decrease i n population density. Also, one can observe a corresponding increase i n the straight grid patterning, the total length, and even the sheer numbers of streets as one moves out. Such physical variation i s not unique to Coimbatore alone, but true of other Indian c i t -ies as well (Berry 1969, Rowe 1973, Smailes 1966). Just as a practised observer can examine a map of modern Coimbatore and distinguish the orig-i n a l site from the town extensions solely on the basis of street patterns, so the same observer can make a similar judgement for other c i t i e s such as Agra, Delhi, and Madras as well. In addition to these clear variations i n layout, however, one must know something about the social character of the city's different areas. Map 10 serves this purpose very well. Apart from the 'Industrial sector' of North Coimbatore, each of the other sectors can be seen to house a specific group of castes within i t s boundaries. Patterns of caste segre-gation, therefore, are not unique to the older areas of the ci t y but to H o t a l number of households E> density of population length of street * * * 2 I f a caste constituted 75 percent or more of the number of households on a street, i t was considered to be dominant. Streets with castes of relatively smaller numbers therefore had a greater amount of mix. 48. the newly developing ones as well. Map 11 is a more composite representa-tion of a l l single caste concentrations (i.e., 75 percent or more) on given streets within the city limits. It i s pretty obvious, as i n the earlier map, that the old core and the old periphery have the largest concentrations. Further, a separate mapping of the location of residential and bus-iness telephones i n the city gives us an independent measure of high and low status areas i n economic terms. Map 12 , for instance, demonstrates the spatial distribution of residential telephones within the city's muni-cipal boundaries. What is conspicuous about this map i s the fewer number of telephones located near the city's peripheries. Residential telephones are concentrated mainly along a few main streets (probably belonging to families who live at the rear of their businesses) and in the older sub-urbs (both Devangapet and R.S. Puram) which we shall see later on are mainly occupied by high caste families. It does appear, therefore, that there exists a certain amount of spatial congruence between c aste and status though the existing data does not allow us to explore this relation-ship in any great detail, A second map (Map 13) shows the spatial layout of business tele-phones in Coimbatore city. Of greatest interest is the fact that the very same streets of the old core are represented when i t comes to busi-ness telephones, as was the case with residential ones. This suggests that there i s l i t t l e specialization of land uses for commercial purposes in this central "core." Where we do find residential phones without business concentration of residential telephones concentration of business telephones *No. of residential telephones No. of households %o. of business telephones length of street k9. phones i s i n the older suburban areas. Furthermore, only a very few streets i n the new industrial area and along the main access routes leading to i t exhibit the western pattern of business telephones being spatially separated from residential ones. In an attempt to identify e l i t e business areas we also examined one further economic variable, i.e., the size of business advertisements i n the telephone directory. This time a rough estimate of the wealth of a firm was made from taking into consideration the size of i t s directory ad-vertisement. Map i i i 1 demonstrates the distribution of these presumably large and important firms. As expected, the major a r t e r i a l route that con-nects Coimbatore to other urban centres outside the region (Trichy Road) i s found to have the largest number of these firms. The next highest concen-tration i s an older suburban area called R.S. Puram, followed closely by the industrial sector of North Coimbatore, and one other business street i n the old core. We shall thus assume that the most prestigious firms are housed i n these areas i n the c i t y . Having looked at some of the general patterns of population d i s t r i -bution we w i l l now examine more closely the residential layout of some of the larger groups i n this industrial c i t y . THE NEUTRAL CASTES A few words are necessary to introduce the concept of 'neutrality' as i t w i l l be used i n the following discussion. In Kongu two caste groups, 'A h point scale was constructed and the scores were allotted proportion-ate to the size of the advertisement on each street. Then the scores were aggregated for each street and the results mapped. 1 . 2 Brahmin and Pillais, enjoy a very high occupational and ritual status. The former often serve as temple priests, and the latter as village account-ants. Both stand at the top of the social ladder where they occupy a pres-tigious niche that the Right and left castes find equally difficult to chal-lenge. Such castes, in turn, play on their special position and do not often engage in status disputes. They thus serve as a sort of 'head1 for an otherwise (at least traditionally) bifurcate social body. It is these groups who are here labelled as 'neutral.' The Brahmins •While the ritual status of the Brahmin is rarely questioned, the economic position of some of the members of this group in rural areas is actually quite low. This is especially true in areas where most of the agricultural land is owned by non-Brahmins. To increase their economic and social standing, therefore, this traditionally scholarly and priestly group has long looked to urban areas for employment. It is not surprising, then, to find large numbers of Brahmins in Coimbatore city. Other tradi-tionally neutral castes have followed their lead. Approximately 9 percent of al l the households in our sample are Brahmin by caste, and the largest concentrations of these are to be found either in the old core or in well-established but newer extensions of the city (see Map 15). Even today, ^For many centuries in India, Brahmins have been accorded the highest ritual status. 2The Pillais are a South Indian caste of accountants who enjoy consider-able power and prestige as village accountants. ^For other examples see Singer's (1920) study of Madras city, or Berry's (1969) study of Calcutta. 51. there i s a strong tendency for e l i t e groups to be found i n the inner belts of the c i t y . In these traditionally central areas of Coimbatore residen-t i a l , commercial, and administrative functions have always been combined. Specialization of land uses as i s found i n western c i t i e s has therefore occurred only a new 'core' locales established by the B r i t i s h . In expand-ing, the Brahmin community have not relocated i n these specialized British sectors intended for commerce (Trichy Road), administration (court collect-or's office area), or pleasure (Race Course Road). They chose instead the older suburbs of Coimbatore city, which, are mixed residential, religious, and commercial locales. In parts of these sectors Brahmin households a l -most completely dominate the scene. Furthermore, there has been l i t t l e change over the years i n the number and location of streets on which Brah-mins have lived (compare 1871 map with that of 1974 map). A sharp and per-sistent segregation of these Brahmin residential locales from the major un-touchable settlements of Coimbatore i s also quite evident. These 'Pure' and 'Polluting' spaces are well separated by buffering neutral zones. Probably for similar reasons the Brahmins have not located their houses along the 'exposed' a r t e r i a l routes of the c i t y . Instead, they seem to have always preferred more 'sheltered' and 'less accessible' streets for their homes. Further s t i l l , the Brahmins of Coimbatore do not reside i n close proximity to the Right division castes of their c i t y (described below). I t i s interesting to note that they are only moderately segregated, with an index of segregation of .33.1 However, i t i s not unusual for ^Index of Segregation (see Appendix II at the back of the thesis) i s c a l -culated: jjj € Xa^ ss total number of the subgroup i n the city ~ — ^ x a l = t ^ n l = * l i e ^°^ s^- population of the ci t y "A Methodological Analysis of Segregation Indices" by O.T. Duncan and B. Duncan, ASR 20: 210-17, 1955. 52. members of the Neutral castes (who are more concerned than Brahmins with ritual matters) to be found living close by to the l e f t division castes. The Pillais (see Map 16) The only other caste in our sample known to have traditionally held themselves aloof from Kongu society as a whole are the Pillais (Beck 1972). In Kongu today there are very few Pillais. Most members of this group are village accountants (about 1 percent) and a large number of them have found their way to the city of Coimbatore. They constitute roughly 5 percent of our sample. like Brahmins, a major cluster of Pillais are to be found in the old core of the city. However, this group also has substantial rep-resentation i n the more peripheral and more polluting areas of the city. The segregation index for the Pillais i s moderately low ( .25) . This leads one to speculate that various lower caste groups may have adopted the 'Pi l l a i ' t i t l e , who do not have the social connections or the means to en-able them to adopt the former's exclusive living standards. It i s also possible (although not so likely) that a number of traditional P i l l a i fam-il i e s have been forced by circumstances to move to less desirable areas. Thus, we find that the Pillais are distributed both in the better residen-t i a l and the polluting areas of the city. In this they are distinguished from their counterparts, the Brahmins, who live only i n the better areas. THE LEFT DIVISION OF CASTES By reason of their highly marketable business skills, the profess-ional or trader castes of India are over-represented i n a l l the urban areas of the subcontinent (Hazlehurst 1967, Fox 1969 and 1973). The left divi-sion castes (who are not untouchable), like the Right divisions show a 53. certain tendency towards spatial contiguity in their residential choices. Furthermore, they exhibit a clear preference for locales that are close to those of the ritually pure Brahmins. In Kongu, the Left division groups include the artisans (Asaris), the merchants (Cettiyars), the weavers (Mutaliars), and the earth diggers (Haicker). These groups constitute only 11 percent of the total regional population, but close to 26 percent of the households in our city sample. We have already noted that this Is more than twice the Right castes1 rep-resentation, an interesting reversal of the rural pattern where the Right castes well outnumber those of the Left. In this section we will discuss in detail only those "touchable" castes who are easily identified (Beck 1972) as belonging to the Left division. These are the Asaris, Cettiyars, Mutaliyars, and Naickers respectively. The Asaris This particular Left division group constitutes 6 percent of our city sample, and has a remarkable tendency to cluster residentially. Asaris are found mostly i n the old core and the old periphery.1 A few of them are found very thinly spread in the Brahmin enclave of Devangapet and the industrial sector of North Coimbatore. They have settled in well es-tablished commercial areas, although actually they are spread thinly over the city as a whole. They have a segregation index of .3U. In the old core the artisans' residences are often contiguous with Brahmin residen-ces. Even in an urban setting, therefore, they seem to associate them-selves with this ritually pure group (see Map 17). Same pattern in a village like Olappalayam. The Cettiyars The Cettiyars constitute the largest of a l l Coimbatore's urban groups, accounting for Ik percent of the total number of households i n our sample, (See Map 18), What i s striking about their spatial spread i s their near monopoly of the main ar t e r i a l routes of the c i t y . They also show a large concentra-tion i n the city's older core areas. While the Asaris remain clustered here even today, showing l i t t l e spread into the newer urban extensions, the Cettiyars have clearly moved from this original core along the various axial routes following, i t would seem, the path of their business interests (Maps of 1872 and 1972). They also have a low segregation index (.25). Thus, within the Left division specific patterns of residence seem to vary with the type of commercial activity undertaken.1 The Cettiyars appear to make l i t t l e distinction between their comm-erc i a l and residential spaces. Often they l i v e i n the back of the shops that they own. However, there does appear to be a trend among the richer Cettiyars to move into the better residential sections of Coimbatore, The Mutaliars The Mutaliars are referred to as mercenaries of the Chola army. Later many of them appear to have settled i n Kongu, and to have taken to weaving and to business (Beck 1972: 32). Besides being warriors and weav-ers, some Mutaliars used to perform important temple services. One a f f i l -iated group, for example, used to supply the Deva-dasis, or temple dancers, ""•Note, for instance, the percentage of Cettiyar houses i n the Brahmin locales. ^Evident on personal observation. 55. and their male counterparts, the dancing masters and temple musicians (Beck 1972: kl, n). In Coimbatore they constitute about 3.5 percent of our sample popu-lation (see Map 19). Again, like other castes i n the group, a few Mutaliars tend to cluster near the old core. Comparison of the map of 1871 and 1972 would show how l i t t l e change there i s i n this group's residential habits i n over 100 years. Being professionals, Mutaliars generally reside near or among other r i t u a l l y high status groups. A few can even be found l i v i n g i n pre-dominantly Brahmin areas as well. In spite of being few i n numbers, they are found i n general to be spatially closer to the households of the Cettiyars and Asaris, i.e., the other Left division castes, than to those of the Right division. Of the Left division castes, they are the least segregated (.16). The Naickers Members of this caste are generally known as well-diggers or earth movers. They have a rather low status; lower, for example, than the Natars of the Right division (Beck 1972: 5 ) . In the c i t y they probably do a l o t of generalized manual labour. They constitute about 3 percent of our sample (see Map 20). Large clusters are to be found i n the original settle-ment s i t e , both i n the inauspicious southern sections as well as i n the more r i t u a l l y neutral sections of the city. This variation i n locale i s perhaps due to variations i n status of the various subdivisions of the Naicker caste. Beck notes that the T o t t i 1 ^Totti refers to "sweepers" and washroom cleaners. 56. Naicker, for example, are a group that bears the t i t l e Naicker but who are labelled untouchables (Chakkilayar or Matari) by others. The wide range of residential locales for Naickers may thus be due to similar factors as were mentioned for the P i l l a i s . That i s , certain low status groups adopt high status names without attaining the associated l i v i n g standard. I t i s clear, however, that the Naickers are not found i n the 'better' residential areas of the city. Many of them can be seen to be clustered near the various commercial streets. In addition, they tend towards a spatial identification with other Left division groups. Thus, Cettiyar or Chakkilayar (Left division castes) residential areas are to be found near several Naicker clusters. Of the Left division castes they are the most segregated (.39). From the foregoing discussion, a single overarching pattern seems to emerge: the Left division castes are to be mainly found i n the older areas of the ci t y though they extend outwards along the communication net-works and commercial routes, just as they tend to form fan-like networks within the region as a whole. The only other f a i r l y substantial nucleus of touchable Left castes can be seen to be located i n the peripheral and industrialized sector of North Coimbatore, where many members of these castes are no doubt employed. THE RIGHT DIVISION OF CASTES The castes of the Right division, as we have noted earlier, are the economically dominant castes of the region. They tend to be centrally located, both i n villages and i n the region as a whole. In Coimbatore cit y , however, this pattern does not hold good. Instead, here the central areas (old core and old suburbs) are mostly dominated by the r i t u a l l y 57. neutral and r i t u a l l y pure castes. Right castes l i v i n g i n the core li v e i n small enclaves of their own. Furthermore, the Right castes constitute only IOJ4 percent of our c i t y sample as against roughly 40 percent of the house-holds i n the region as a whole. In the following discussion only those castes whose Right division status i s certain (Beck 1972) have been included. These are (1) the Kavun-tars, who are the agricultural landlords of the Kongu region: (2) the toddy-tappers, popularly known as the Natar; and (3) the Tevars, who have varied occupations but who take pride i n their reputation for military prowess. The Kavuntars The Kavuntars constitute 8 percent of our sample population. The largest concentration of Kavuntars i s i n the old core and the old periphery of Coimbatore c i t y , a carry-over, perhaps, from a time when Coimbatore had a village type of spatial organization (see Map 21). Given the number of Kavuntars i n the c i t y , many of them seem to occupy rather inauspicious and limited sections of the urban whole. In the old core the Kavuntar resid-ences are not always spatially contiguous with Brahmin residences. But one can find a few Kavuntars on or close to streets where Brahmins l i v e . Fur-ther, the Kavuntars neither gravitate to the well defined commercial streets nor to the administrative areas. They do not seem to choose to li v e i n mixed residential and commercial areas the way many Left castes do. They are also relatively low on the segregation scale ( . 2 4 ) . The Natars The Natars form a small minority (1.4 percent) i n Coimbatore. From the mapping of our household data (see Map 22) i t i s clear that the Natars and Kavuntars congregate i n the same general areas of the 58. c i t y . There i s also one substantial Natar concentration i n the industrial sector of North Coimbatore. Perhaps what i s important i n identifying the Natars' status i s their conspicuous absence from any of the Brahmin enclaves of the c i t y , unlike a few Kavuntars. As Hardgrave (1969: 22) notes, unlike the Parayan untouch-ables, the Natars had access to the streets of the agraharam (Brahmin ward or v i l l a g e ) . Yet they do share with untouchables the tradition of keeping their spatial distance from Brahmin residential locales. Like Kavuntars, too, the Natars are generally not found l i v i n g along the main a r t e r i a l or business routes within the c i t y . In terms of residential location, therefore, the urban Natars, like their rural counterparts, are found to be closely associated with their Kavuntar patrons and a l l i e s . They are a very highly segregated caste group i n Coimbatore ci t y (.54). The Tevars The term 'Tevar1 i s nearly synonymous with the term 'Maravar,' which i s the name of one of the earliest and fiercest of southern t r i b a l groups (Thurston 1909: v o l . 5, 48). The Tevar are generally of low, yet touchable status. Hardgrave (1969: 23) notes that Tevars are i n general not very different i n status from the Natars (see Map 23). what i s crucial to our understanding of Tevar spatial organization i s the recognition of their close alliance to the Natars and Kavuntars. It i s thus not surprising that we find them located i n clusters near these two other Right division groups. Their residential space can be said to be similarly 'neutral, 1 well away from the clearly pure or polluting areas. In spite of employment i n urban centres, the Tevars are neither a profess-ional nor a commercial caste. As a result they, too, are not found to live 59. along the main commercial streets of the c i t y . I f anything, the Tevars are even more closely a l l i e d to the Kavuntars, spatially speaking, than are the Natars. The Tevars, unlike the Natars, are not so highly segre-gated (.28). In spite of a few d i f f i c u l t i e s , i t does thus seem possible to trace a specific residential pattern for the Right division castes i n Coimbatore ci t y . The existence of some kind of common identity or sense of inter-dependence appears to be a strong factor i n the contiguous spatial place-ment of these groups within the c i t y . THE SERVICE CASTES Members of these groups perform specific services for higher rank-ing castes. Together they constitute only 3 percent of our sample. The castes that we w i l l deal with under this heading are (a) the washermen (vannar), and (b) the barbers (Navithar). The Vannar The washermen constitute 1.75 percent of our sample population. Washermen i n South India are quite low i n social status. Unlike other castes, they rarely cluster spatially i n particular villages. Usually only one or two such families w i l l serve each small settlement. In Coimba-tore c i t y , however, the washermen do tend to l i v e i n a few well defined areas and to service their clientele from certain nodal points (see Map 2l+). Their segregation index i s .27. The Navithar The barbers are also a relatively small group i n Coimbatore, fewer i n number than the washermen even, constituting less than 1 percent of the 60. households i n our sample. Like the washermen, however, the barbers are also found on certain streets (see Map 25). A substantial concentration on certain commercial streets i s particularly v i s i b l e . They have a segregation index of .32. Both barbers and washermen are indispensible to c i t y l i f e . Yet they can only travel limited distances i n their daily work, since they travel mainly by foot. For these reasons, i t would seem, small clusters of these impure groups are to be found i n several centrally located areas of the c i t y . OUTSIDE REGIONS Apart from the castes already discussed, a l l of whom are well rep-resented i n the population of the Kongu region as a whole, there are other communities that have made their way to Coimbatore from more distant places. Together these 'foreigners 1 form quite a substantial bloc of the urban popu-lation, constituting 22 percent of our sample. Broadly speaking, we can divide these people from elsewhere into two categories: (a) North Indian, and (b) South Indian. The former are very few (i . e . , about 1 percent of the sample city's household population). They can be further subdivided into three groups: (a) people from the Punjab (Punjabi), (b) people from Gujerat (Sait), and (c) people from Rajasthan (Marwadi). Of these, apart from the Saits, who are diffused, the other two groups cluster i n the middle class residential sector of R.S. Puram and Devangapet. The South Indian castes and communities are greater i n numbers. Of special significance to Coimbatore city are the Naidus and the Boyars (castes who are traders from Andhra Pradesh), the Malayalis (people from 61. Kerala), the Okkiliers (Canarese cultivators), and the Reddis (agricultural caste from Andhra Pradesh). The Okkiliers are a striking example of a segregated caste: a l l the Okkiliers are found i n the old periphery of Coimbatore. The Reddis, who are an agricultural group from Andhra Pradesh, are better represented i n the rural d i s t r i c t s than i n urban centres li k e Coimbatore. The Naidus The Naidus, who are traders from Andhra Pradesh, are not well rep-resented i n the region although they are substantially represented i n Coimbatore cit y (see Map 26). In our sample they contribute roughly h per-cent. The Naidus resemble the Cettiyars i n their residential distribution. Thus, the heaviest concentration of Naidus are found i n the commercial heart of the old city and i n the industrial sector of North Coimbatore. In terms of their social standing, the Naidus are 'outsiders,' although they are clearly not of low status. Because of their business interests they have been traditionally located i n the r i t u a l l y neutral areas of the ci t y , although their extreme wealth i s buying them entry into the Pure or Brahmin areas such as along Race Course Road. They are a highly westernized group who seem to have managed a certain spatial ex-pression of their values through their choice of old B r i t i s h residential space. They are f a i r l y segregated (.32). The Boyar The Boyar are supposed to be a subsection of the Naidus and a trad-ing caste by profession (Thurston 1909: v o l . 5, 138). They contribute to 2 percent of our sample population. In the c i t y , however, many of them 62. are forced to be manual labourers.* In accordance with their low prestige occupation, some Boyars can be found i n the old periphery and the less prestigious sections of the industrial section of North Coimbatore (see Map 27)• Others are found i n the better and older suburb of R.S. Puram, Almost no Boyars are to be found i n Brahmin locales. They are not so highly segregated (.22). The Malayalis The Malayalis constitute a large category i n our sample, nearly 9 percent. I t i s very d i f f i c u l t to identify the very varied professions of mig-rant Malayalis. I t i s , indeed, even problematic to say that they are res-ident i n any particular area of the c i t y at a l l . They are to be found both i n commercial and residential sections of Coimbatore (see Map 28). This wide distribution accords well with the Malayalis rather flexible attitudes towards occupations and towards pollution i n general. Immigrant Malayalis are almost always westernized i n their tastes and many of them are Christ-ians. They are the least segregated of a l l the castes and communities (like the Mutaliars) (.16). From the foregoing discussion i t appears that strong ethnic or communal ties do not characterize the spatial organization of most of these South Indian "immigrant" groups, with the exception of the Okkiliers. In-terestingly, however, the North Indians are much more l i k e l y to form small spatial enclaves. Perhaps the language barrier, much stronger for them than for South Indians, i s a par t i a l factor i n t h i s . The Southern immigrant Beck (1972) shows this to be true i n rural areas as well. 63. communities are more or less evenly and thinly spread throughout the city and do not seem to have clear nodal points for caste aggregation. They seem to behave spatially towards other castes, including the Brahmins and Pillais, in a neutral way. NON-HINDU This classification applies to two substantial religious communi-ties both in the region and in the city—the Moslems and the Christians. Historically the Moslem inroads on South India came first. In Kongu the Moslems constitute 3 percent of the population, 11 percent of Coimbatore city population, and approximately 13> percent of our sample population. Of all three major religious communities in Madras state (Hindu, Moslem, and Christian), the Moslems are proportionately the most concen-trated in urban areas. William Francis is said to have noted this phenom-ena in the 1901 census as follows: "Mossalmans, who are largely traders, show the greatest preference for town life (of the three religious commun-ities). In some of the sea-port towns, the Mossalmans, who a re generally tabbais, actually outnumber the Hindus." (Quoted from Mines 1973: 37). It is apparent, therefore, that a relatively high degree of urbanization has characterized the Moslems of Madras even before the beginning of the 20th century. The Moslems (see Map 29) This urban concentration is well borne out by our data. Large num-bers of Moslems (7.5 percent) are found within the municipal limits. In our sample the Moslems alone contribute roughly 11 percent. From the map-ping of our data i t was further evident that not only are they found in 64. large numbers within the c i t y boundaries, but within the boundaries of the old c i t y as well. Although the Moslems l i e outside the pale of Hindu soc-iety, they are s t i l l not treated as untouchables i n a spatial sense. It i s not surprising, however, that we find their conspicuous absence i n the •pure-residential' areas of the c i t y . An additional factor which affirms their tendency to aggregate i n commercial locations i s that the second largest concentration of Moslems i n the c i t y i s found i n the s a t e l l i t e nuc-leus of North Coimbatore. Furthermore, they are highly segregated ( . 3 8 ) . The Christians The Christian population, both i n the region and i n the city, i s not very large. Only 4 percent of the household population of our sample is Christian. From the mapping of our data (see Map 3 0 ) , i t is clear that their main aggregations l i e outside the old core, and most of the older suburbs as well. Large sections of Christians are to be found adjacent to the Bri t i s h site, around the main churches of Coimbatore, and also i n the i n -dustrial sector of North Coimbatore. Many low caste groups were converted to Christianity i n the 19th century by missionaries. Since these converts s t i l l l ive on the boundar-ies of the city's space, however, i t i s d i f f i c u l t to say whether these new Indian Christians managed to escape their previously polluting status by conversion. They are also very highly segregated ( . 4 7 ) . Both Moslems and Christians appear to be separated from the import-ant Hindu castes i n spatial terms. Of the two, Moslems seem to have stron-ger nodal ties while the Christians seem to be relegated largely to the peripheral areas. Those Christians found interspersed with other groups i n better residential areas are very l i k e l y higher caste converts including 65. Christian Malayalis. While Moslems are residents of already well estab-lished commercial sectors, such high-caste Christians can be found i n every sector of Coimbatore c i t y . THE UNTOUCHABLES India's untouchables have become a very d i f f i c u l t group to obtain information on ever since the turn of the 20th century. This i s because of the hesitancy, f i r s t of the British and later of contemporary Indians, to o f f i c i a l l y label any group by this derogatory term. However, the stigma of untouchability s t i l l exists despite i t s abolition by the Indian constitu-tion. To escape or avoid the issue various euphemistic substitutes such as 'Hari.jan' (a term popularized by Gandhi, meaning children of God), and 'Adi-Dravida' (earliest inhabitants of India (Dravidians) popularized by the D.M.K. Party) have been tr i e d . None have f u l l y succeeded i n masking the persistence of wide-spread traditional attitudes. The lingering contradic-tion makes for profound embarrassment on the part of many educated persons who prefer to deny the existence of untouchability by not labelling i t . We w i l l not dwell on the problem here, but merely state that i n Coimbatore the 'untouchables' are s t i l l a highly stigmatized group. This point comes ac-ross clearly on our ci t y map. The so-called 'Harijans' l i v e largely by themselves on the outskirts of the ci t y proper, or i n crowded slums along various public access routes, where they have become 'squatters' on munic-ip a l land. This isolation i s a legacy of the past. It may be viewed as the physical expression of more subtle social barriers, based on the higher castes' desire to avoid these groups' polluting presence. In Coimbatore today, these communities constitute a t o t a l of about 8 percent of the city's total household population. Probably another 66. smaller percentage are classified i n our data as 'Christians.' However, we have only roughly 5" percent who are clearly identified as untouchables i n our sample. Chakkilayaris, or the shoemakers, are numerically the largest of the ex-untouchable groups i n our sample (as well as i n the c i t y and the re-gion). Other castes that f a l l under this category are the Pallars (agri-cultural labourers) (Thurston 1909: v o l . 5, U73), Valluvar (priests of the above) (Thurston 1909: v o l . 7, 303), and Paraiyar (drum beaters) (Thurston 1909: v o l . 6, 77). In our sample we have no Paraiyar, although we have certain scheduled tribes who can be classified as untouchable. They are the Koravar (gypsy tribe especially found i n Coimbatore) (Thurston 1909: v o l . 3, liUl), and the Kurumba (Aboriginal tribes of the Nilgris) (Thurston 1909: v o l . U, 158) respectively. As a result of their social status, the untouchables have clearly been excluded from the good residential areas of Coimbatore. Their homes are mainly found i n the peripheral areas. Such settlements are clearly stigmatized and referred to as Cheris i n local speech. However, we do find some untouchable settlements located right i n the heart of the c i t y as well. In part this i s because the urban area has grown by stages. What used to be a peripheral areas i s so no longer. As time passed, the older untouch-able settlements have been enveloped within the city's growing boundaries. The location of the untouchable settlement of 1871, for example, remains unchanged today. For this reason such older untouchable enclaves can be used as useful markers of core-periphery areas of the past. In addition, one can now find untouchable settlements along the city's railroad tracks. Here undesirable public land has been made available for squatter settle-ments. Such proximity to the railroad tracks has always been a symbol of 67. low status, not only i n the East but also i n the West. These areas that untouchables locate i n soon become defined as 'Polluting' spaces that must remain spatially isolated from the larger whole. Furthermore, i t i s impor-tant to note that the 'untouchables' have spatially and socially subdivided themselves. Just as i n the rural areas, i n Coimbatore we do not find d i f f -erent groups of untouchables l i v i n g next to one another. The Chakkilayars, for instance, reside solely i n the south, while the Valluvars have a special niche on the west side of Coimbatore c i t y , while the Pallars are to be found mainly along the railroad tracks and on the peripheries of industrial North Coimbatore. Wherever they are found, they not only seem to be spa-t i a l l y cordonned off from each other, but from 'socially significant space' within, the urban environment generally. They are a very highly segregated group (.51). INDEX OF SEGREGATION In order to summarize the index of segregation for a l l the above r castes, a graph i l l u s t r a t i n g the extent of segregation of a l l the castes was drawn (see Figure 2). INDEX OF DISSIMILARITY As Timms (1965) notes, the most useful instruments yet devised for the quantitative description of geographical patterns are the Index of Dissimilarity and the various measures derived from i t (1965: 2h0). By calculating this index, we can assess the relative standing of each caste with a l l other castes. This index i s calculated from data giving for both castes the percentage of the total l i v i n g i n each areal unit. The index of dissimilarity i s then one-half the sum of the absolute differences between the two caste populations taken area by area. CO ST S* CM Q) bO w f « if e •H ha W o a V V •if 4 5 -f i Key :to' the Graph • j | X;4 Nadar [•  j j H .; H i M i l l :-: Asari' :  : 2 - Untouchable ; i - j = i ! - ; 112 !-: Cettiyar ; 3'- Christian; >'\ ''!" ] 13 - P i l l a i : 4 - Naicker | ! ^  M j , M lk -: Kavuntar • 5 - Moslem : i' j !• | j ; 15 -: Boyar: ; '• 6'4 Brahmin * . ; : ; i l 6 Mudaliar 7 - Naidu i i M 17 - Malayali 8 - Navithar U . : 9 - Tevar ; .'' .[ 10 - Vannan \ ! 1 • u ; ; j '- ' i i ; ' • <. j. ; ; i • ; • ! • f - l ! : ' ! ' 1 ' i. • 1 i ' • i ; j ! ; i •' i ' : • ; : ' • -i V I I I , 1 — • - f — i TABLE IV: The Index of Segregation of Castes i n Coimbatore City Neutral Castes Brahmin P i l l a i .33 .25 Right Division Castes Kavuntar Nadar Tevar .24 .54 .28 Service Castes Udaiyar Vannan. Navithar Pandaram .27 .32 Left Division Castes Asari Cettiyar Mudaliar Naicker .30 .25 .16 .39 Immigrants Naidu Boyar Malayali .32 .22 .16 Outside Religions Moslem Christian .38 .47 Untouchables Pallar \ Valluvar j Holeya \ Paraiyan [ Chakkilayar J .51 70. The following table sets out the index of dissimilarity for a l l the above castes. On a 1-100 scale the higher scores indicate greater dissimilarity. Castes that had a score of 60 and above were treated as highly dissimilar and those below 20 were treated as least dissimilar. The Brahmins and untouchables were the most dissimilar. The Left division castes i n general were also dissimilar to the untouchables and other non-Hindu communities. The castes that were least dissimilar were the Brahmins and Chettiarsj Chettiars and Kavuntars; P i l l a i s and Naidus"; and f i n a l l y , Naidus and Christians. Therefore, a few of the higher Left division castes, neutral castes, higher Right division castes, immigrant castes, and non-Hindus had the index of least dissimilarity. TABES V: Index of Dissimilarity Between Castes 71. in Coimbatore City to nl "SJ 1 h h & £ i—I 3 to to C 'H H > 73 !> S i> •H to «5 o> co «5 (i, w EH > a Brahmin 33 2k 49 45 42 36 Pillai 39 34 49 45 40 Kavuntar 68 38 45 48 Nadar 67 57 65 Tevar 30 49 Vannar 29 Navithar Asari Chettiar Mutaliyar Naicker Naidu Boyar • Malayali Moslem Christian Asari Chettiar 1 •H i H CO ft—J Naicker Naidu Boyar •H H & r-H M Moslem Christian Untouchable 45 18 30 59 34 47 25 62 4o 73 49 62 42 64 13 29 31 48 33 53 26 19 50 51 39 42 22 40 47 55 62 76 70 82 65 51 61 57 79 28 29 36 19 24 54 32 63 20 71 50 35 37 27 39 46 22 33 26 62 39 59 31 35 52 45 42 34 48 59 53 38 34 52 52 34 43 35 64 44 23 48 67 41 21 54 51 64 30 45 31 24 36 53 53 63 49 27 46 26 26 25 52 35 47 55 19 43 30 64 63 59 37 53 37 On a 1-100 scale the higher scores indicate greater dissimilarity-72. CONCLUSIONS Thus far we have seen how land space i n the city has been used for economic purposes as well as social purposes. Thus, we have seen how i n some areas like the old core there i s hardly a distinction between r e s i -dential and commercial space. These areas are also areas of r e l i g i o -r i t u a l as well as secular importance. The older, middle class suburb of Devangapet, on the other hand, i s mainly residential and has r e l i g i o -r i t u a l importance as compared to the adjacent middle class suburb of R.S. Puram that has secular importance and combines residential and commercial land uses. The industrial sector of North Coimbatore has the largest number of factories, around which l i e members of nearly every caste. It i s a lower class area as well. In contrast, the new suburbs on the west side of i t are gradually emerging as an e l i t i s t area. Finally, the old periph-ery and the new peripheries s t i l l remain as the undesirable locales of the lower castes i n Coimbatore. We have also seen how the various castes have carved out l i t t l e niches for themselves i n each of the above sectors, depending on their socio-economic status. Thus, i n the old core, the largest number of castes are the Left division castes, and the old periphery i s peopled by the immi-grants, non-Hindu communities, and untouchables. In the new core the immigrants and neutral castes appear to domin-ate. As we move away from the original settlement we encounter a large number of Left division castes, immigrant groups, neutral castes, service castes, as well as a small section of untouchables i n the secular suburb of R.S. Puram, while we find a large Brahmin enclave surrounded by Right 73. division castes and immigrant castes i n the more tradition-bound suburb of Devangapet. The greatest degree of caste mix i s encountered i n the industrial sector, while selected numbers of caste groups are to be found i n the new suburbs. In the west side extensions of the city (known as the new peri-phery) the lower caste groups and untouchables predominate. Further, the index of segregation and index of dissimilarity were calculated for a l l the castes. Some castes were found to be more segre-gated than others. The index of dissimilarity allowed us to assess the social standing of each caste with respect to a l l other castes. Chapter $ CONCLUSIONS "The sense i n which spatial arrangements reflect social values and social structure may be thought fanciful for some societies, but i n Hindu l i f e i t i s elegantly mani-fest i n ideology and i n physical reality." — William L. Rowe (1973: 2k2) This research has addressed i t s e l f to one basic question—Does the spatial organization of an Indian city reflect certain overarching social values held to be important by the culture at large? More specifically, we have asked whether the spatial organization of the different castes i n Coimbatore city i s concurrent either with the traditional organization of space i n a village or with the observed patterning of space i n the region i n which i t l i e s . From the foregoing discussion i t has become evident that Coimbatore, i n fact, represents a curious compromise between patterns found at the level of the individual village and of the region as a whole. Just as caste was primary to the analysis of the village of Olappalayam and of the region of Kongu, 'Caste' i s crucial to our understanding of the nature of the core ci t y of Coimbatore. The findings of this study of the c i t y inte-grate well with earlier anthropological studies of both village and region. In the following sections we w i l l recapitulate some of these findings to ill u s t r a t e how the city f i t s into the framework of these macro and micro social systems that surround i t . In analyzing the stages of growth of Coimbatore i t was seen that the origins of this urban centre date from as far back as the rule of the Chera Kings (9th century). Further, i t was also seen that descriptions of 75. the early spatial organization of this particular c i t y do not d i f f e r mark-edly from the spatial form of an Ideal Indian City as envisaged by the b r i l l i a n t Kautilya of Chandra Gupte Maurya. Coimbatore, i n i t s earlier period, gave spatial expression to certain traditional concepts of social order. In the centre of i t s original site stood the king's palace and a few major temples. Radiating from this centre were the residential spaces of the different castes, graded according to their social rank. And beyond the main settlement were the spaces allotted for the untouchable castes. In this way, the hierarchy of spaces radiating outward from the centre of the c i t y was at f i r s t f u l l y congruent to a traditionally accepted hierarchy of social statuses. During the next few centuries, as the meagre documentary evidence has shown, l i t t l e by way of innovation i n this basic layout was attempted. Successive rulers merely extended the city's original format. However, with the arrival of the British i n India, some new ideas were at last i n -troduced. Br i t i s h concepts of the ci t y layout did leave their mark on the city's landscape. This process was gradual and not revolutionary i n i t s approach. Soon the ci t y began to take on a new look. I t now represented the attempted merger,of two contradictory principles—one old and the other new to the Indian scene. In the course of time, inter-and intra-regional mobility, hastened by the improvements i n communication and the introduction of a new tech-nological order, caused the ci t y to expand i n a l l possible directions. In the process i t engulfed and over-ran i t s earlier peripheries. The small town of Coimbatore soon burgeoned forth into the large industrial complex that we know today. These various stages of the city's growth are evident from i t s 76. complex spatial format as i t can be seen today. As Hinduism i s said to engulf and accommodate a l l that tries to challenge i t , so the c i t y of Coimbatore has incorporated the spatial concepts brought by the Br i t i s h without abandoning i t s earlier patterning i n any radical way. Thus the new British core was b u i l t around the old core and the old periphery. This British influence skirted the city's older suburbs, R.S. Puram and Devanga-pet, but made i t s presence f e l t again i n the peripheral sectors of Indust-r i a l Coimbatore and i n the construction of the newer outlying suburbs to the north. At the same time, however, there was also an expansion to the west side of the original site of the ci t y enabling i t to form a new west-ern periphery. This last expansion i s of particular interest since i t has followed more traditional principles. This new periphery again tends to accommodate the lower and more "polluting" newcomers. These temporal and spatial sequences do not, by themselves, provide a framework for understanding each area's special social character. I t was an analysis of the caste composition of these various sectors that brought to light this last relationship; the important linkage existing betxreen the spatial and social dimensions of urban l i f e . The questions that were formulated to express these relationships were: (a) Which castes occupy the centre of the c i t y , and which of them live at i t s peripheries? (b) Are these centrally located castes r i t u a l l y pure? If not, do they economically dominate cit y l i f e ? (c) Can such areas be described i n terms of Right and Left Division caste presence as i s true for the region as a whole? We w i l l now attempt to answer these three questions by summarizing our observations i n the various sectors. We w i l l look at the caste 77. composition of each of these sectors i n turn. Further, our order of the analysis w i l l coincide with the temporal emergence of the various sectors. 1 Old Core and Old Periphery An examination of a l l the castes i n these two sectors brings to light some striking contrasts. The representation of the different castes i n both these areas approximates what we know to be the traditional rural pattern. Since the old core was the centre of the original settlement i t housed a large percentage of the socially higher and economically import-ant castes of the region's rural areas (see Appendix I I I ) . However, an important point to note i s that while the Right divis-ion castes are most numerous i n the centrally located village settlements of Kongu, i t i s the Left division castes who are numerous i n the c i t y (hh percent). The neutral castes follow with 21 percent and Right division castes constitute only 17 percent of the sample population. While the above castes are poorly represented (with the exception of the Left division castes with 32 percent) i n the old periphery, the immigrant group (with an aggregate of 21 percent), service castes (3 per-cent), non-Hindu or outside religions (25 percent - Moslems alone consist-ing of 2k percent), and untouchables (7 percent) have a f a i r l y high rep-resentation. In the original settlement, therefore, the Left division castes were numerically the strongest. See Appendix III. 78. The New Core Since the new core ( c i v i l lines) of Coimbatore is not well rep-resented i n our sample, the distribution of the various castes here i.s not very r e l i a b l e . However, i t is' clear that those few immigrant groups who are both highly westernized and wealthy have sought entry into this area (22 percent). Interestingly enough, there i s also an equally high percent-age of neutral castes (22 percent) i n the new core. Further, a large en-clave of untouchables (25 percent)' has been wedged between the old periph-ery and this 'new core.' One might speak of i t as defining a clear bound-ary between the two radically different parts of the city's whole. Older Suburbs (R.S. Puram and Devangapet) The better residential areas of the c i t y are the older suburbs. These form a substantial central belt to the north of the old core area. In both areas the Brahmins have carved.a niche for themselves along with their neutral a f f i l i a t e s , the P i l l a i (R.S. Puram = 20 percent; Devangapet = 18 percent). However, there are certain interesting d i f f e r -ences to be noted between these two sectors as well. While R.S. Puram had only 4 percent of the Right division castes, i t had 8 percent of the ser-vice castes and 7 percent of untouchables. Devangapet, on the other hand, had 13 percent of the Right division castes, only 2 percent of the service castes, and a negligible number of untouchables (less than 2 percent). Hence, while R.S. Puram represents a typically middle class secular area, Devangapet i s a more restricted middle class area and exemplifies a stronger emphasis on r e l i g i o - r i t u a l purity. Further, i f the significant difference between the original settle-ment and the older suburbs l i e s i n the fact that while immigrant groups and 79. non-Hindu communities (like Christians and Moslems) are peripheral to the old core, they are well represented in the central belt of Devangapet and R.S. Puram. Finally, the agriculturally important castes of the region are best represented in a l l these sectors. As we move outward, from R.S. Puram and Devangapet, we encounter the peripheral sectors of North Coimbatore. The northeast end of this sector is characterized by a number of small firms as well as by larger industrial installations. Further, this is the only sector that houses a l l the different castes of the city. It is consequently the area to have the least religio-ritual significance. Being peripheral, and housing a number of industrial units, i t is also a comparatively undesirable living quarter for the richer and "purer" residents of the city. The immigrant castes, as the table indicates, constitute by far the largest group (27 percent in aggregate). The Moslems and Christians closely follow them with an aggregate of 20 percent. There is also a 14 percent representa-tion of neutral castes as well as 14 percent of the Left division castes. Finally, h percent of the sectoral population is made up of untouchables. The new suburb which lies to the west of North Coimbatore repres-ents a striking contrast to the former. Although i t , too, is spatially peripheral, this area is symbolic of an emergent or non-indigenous spatial pattern. It is a fairly high class area, occupied by westernized and wealthy members of the "purer" castes (30 percent), but also accommodat-ing Moslems and Christians (lU percent), immigrant groups (28 percent), and a few Right (9 percent) and Left division castes (16 percent) as well. Conspicuously absent here are the service and untouchable castes, in other words, those of l i t t l e economic means. While ritual purity is not an im-portant variable for entry into this area, sifting and selecting of resid-ents here occurs solely through the manipulation of economic criteria. 80. Here we see a true "class" as opposed to a "caste" principle at work. The new peripheries, which are additions to the ci t y on i t s west side (close to the old core), best represent the other s i d e o f this new trend. In this sector, a large number of Moslems, Christians, Right d i v i -sion castes and untouchable castes are spatially contiguous* Here the representation of the pure castes such as the Brahmins, as well as of the professional and trading castes such as the Cettiyars i s negligible (see Appendix III, column 8). The most conspicuous groups i n this sector are the Moslems and Christians (21 percent) and the untouchables (2k percent). The Right division castes follow with a total of 16 percent (13 percent of which i s constituted by the Nadars alone), followed by the neutral castes (12 percent) (of which the P i l l a i constitute 10 percent). The immigrant groups constitute 10 percent and the Left division another 10 percent. In terms of spatial patterning, this area represents the emergence of a "low class" as opposed to a "low caste" suburb. In analyzing the spatial organization of the various castes i n the c i t y , then, two important patterns have been seen to emerge. They are: (1) the traditional Indian urban pattern (which combines elements of traditional village and regional structure), (2) the modern Western urban pattern. Rather than being arranged stably, as on a continuum, however, these two types are f a i r l y independent. The old core and periphery and i t s later extensions best approximate this region's traditional rural structure. The largest aggregations of the agriculturally important castes are to be found within these sectors. In these areas we find what I shall term a traditional urban patt-ern of l i f e . By traditional, I mean that the type of economic activity 81. (essentially consisting of small, localized, non-mechanized enterprises) a resident pursues continues to be closely linked to the type of relation-ship he i s enmeshed i n and the type of values he holds. Here residential and business areas are not well separated just as the economic, social and religious spheres of l i f e i n general are not extensively segmented. A modern Western pattern of l i f e , by contrast, typifies the more European areas of the c i t y . Here economic enterprise i s spatially separ-ated from residential a c t i v i t y . In the peripheral sectors of North (new suburb) Coimbatore entry appears to be based solely on earning capacity and the resultant l i f e styles of neighboring families to be highly variant i n the home and carefully segmented off from l i f e at work. Neither pattern i s invariant or stagnant. Each of these i s the end product of consistent directional changes that have occurred over the years. The one pattern i s an adaptation of a traditional order extant i n the region at large, while the other i s the product of external and modern forces that are affecting the ci t y at present. Both these blend, i n vary-ing proportions, to form the modern industrial Coimbatore of today. A number of questions have been raised by this research, although they have not yet been answered. Areal differentiation affected by macro-social processes answers only part of the questions. The effect of indiv-idual activities or micro-social processes, within this spatial framework, s t i l l remains to be explored. BIBLIOGRAPHY 82 Aberle, E. Kathleen I960. The Social Structure of a Tanjore Village. In India's Villages, Second Edition, rev. pp. 90-102. M.N. Srinivas, Ed.- Bombay: Asia Publishing House. Abu-Lughod, J.L. 1969. Testing the Theory of Social Area Analyses: The Ecology of Cairo, Egypt. American Sociological Review, 3h'- (2), 198-212. Adegal, Prince Ilango 1965. Silapadikaram. Trans, by Alain Danielou. London: George Allen and Unwin Ltd. Ananthakrishna Iyer, L.K. 1909-12. The Cochin Tribes and Castes. Two volumes. Madras: Higginbotham. Arokiasami, M. 1956. The Kongu Country. Madras: Madras University Press. Bailey, F.G. 1957. Caste and the Economic Frontier. Manchester: Manchester University Press. Baliga, B.S. 1966. Madras Dist r i c t Gazeteers. Coimbatore: Government of Madras. Barth, F. (Ed.) 1969. Ethnic Groups and Boundaries. Boston: L i t t l e , Brown & Co. Beals, Alan R. 1955* Interplay Among Factors on Change i n a Mysore Village. In Village India. McKim Marriot, Ed. Chicago: University of Chicago Press. Beck, Brenda E.F. 1970. The Right/Left Division of South Indian Society. Journal of Asian Studies, 29: (4), 779-98 1972. Peasant Society i n Konku - A Study of Right and Left Sub-Castes i n South India. Vancouver: University of British Columbia Press. 1975• Core and Boundary Systems i n South India: Towards a Regional Model of Stratification. To be published as a chapter IN Regional Analyses: A New Approach to Complex Social Systems. New York: Academic Press. Berry, B.J.L., and Rees, P.H. 1969. The Factorial Ecology of Calcutta. American Journal of Sociology, 7U: (5), UU5-91. 83. Berry, Brian J.L., and Horton, E. Frank 1970. Geographic Perspectives on Urban Systems. New Jersey: Prentice Hall, Inc. Beteille, Andre 1965» Caste, Class, and Power. Berkeley: University of California Press. Burrow, T., and Emeneau, M.B. 1961. Dravidian Etymological Dictionary. Oxford: At the Clarendon Press. Buttimer, A. 1969* Social Space i n Interdisciplinary Perspective. Geographical Review, 59: 107-26. 1971* Society and Milieu i n the French Geographic Tradition. AAG Monograph Series Number Six. Chicago: Rand McNally. Carter, A.T. 1974. E l i t e P o l i t i c s i n Rural India. Cambridge: Cambridge University Press. Chopra, P.N. (Ed.) 1969. Who's Who of Indian Martyrs. New Delhi: Ministry of Education and Youth Services. Cohn, Bernard N. 1955. The Changing Status of a Depressed Caste. In Village India. McKim Marriot, Ed. Chicago: University of Chicago Press. Coimbatore 1966. Madras Di s t r i c t Gazeteers. Compiled by Dr. B.S. Baliga. Madras: Government Press. Crane, Robert I. (Ed.) 1967. Regions and Regionalism i n South Asia Studies. Papers presented at a symposium held at Duke University, A p r i l 7-9, 1966. Duke University Program i n Comparative Studies on Southern Asia. Monograph and Occasional Paper Series, Monograph Number Five. Durham, N.C.: Duke University Press. Davis, Ringsley 19!&-55» Urbanisation and the Development of Pre-Industrial Areas. Economic Development and Social Change, 3: 6"-24. 1969. The Urbanisation of World Population. In The City i n Newly Developing Countries. Gerald Breese, Ed. Englewood C l i f f s , N.J.: Prentice Hall, Inc. Dumont, Louis 1966. Homo Hierarchicus. Paris: Gallimard. Dumont, L., and Pocock, D.F. (Ed.) 1957» Village Studies i n Contributions to Indian Sociology. Number One. Paris: Mouton and Company. Duncan, O.T., and Duncan, B. 1955. Residential Distribution and Occupational Stratification. American Journal of Sociology, March. Dupuis, Jacques I960. Madras et le Nord du Cororaandel. Paris: Librairie d'Amerique et d 1Orient. Dutt, B.B. 1925. Town Planning i n Ancient India. Calcutta: Thacker, Spink & Co. Firey, W. 1945. Sentiment and Symbolism as Ecological Variables. American Sociological Review, 10: IJ4.O-48. Reprinted i n Theodorson, (l96l), 253-61. Forrest, G.¥. 1905. Cities of India - Past and Present. London: Archibald Constable & Company, Ltd. Fox, Richard G. 1968. From Zamindar to Ballot Box. Ithaca: Cornell University Press. . Urban India - Society, Space and Image. Durham, N.C.: Duke University Studies on Southern Asia, Monograph Number Ten. Ghurye, G.J. 1962. Cities and C i v i l i s a t i o n . Bombay: Popular Prakashan. Gi l l i o n , Kenneth L. 1968. Ahmedabad: A Study i n Indian Urban History. Berkeley: University of California Press. Gist, Noel P. 1957. The Ecology of Bangalore India - An East-West Comparison. Social Forces, 35: 356-65. Gupta, S.M. 1971. Plant Myths and Traditions i n India. Netherlands: F.J. B r i l l , Hall, E.T. 1959. The Silent Language. Garden City, N.Y.: Doubleday & Co., Inc. Hall, R.W. 1974. Spatial Behaviour i n Victory Square. Unpublished M.A. Thesis, University of British Columbia. Hardgrave, R.L. 1969. The Nadars of Tamilnad - The P o l i t i c a l Culture of a Community i n Change. Berkeley: University of California Press. Hasan, Mohibbul 1951. History of Tipu Sultan. Calcutta: Published for the World Press. 85. Hawley, Amos 1950. Human Ecology - A. Theory of Community Structure. New York: The Ronald Press Co. Hazlehurst, L.W. 1966. Entrepreneurship and the Merchant Castes i n a Punjabi City. Monograph Number One. Durham, N.C.: Duke University Press. Hoselitz, B.F. 1954-55. Parasitic Cities. Economic Development and Social Change, 3: 278-94. 1959• The Cities of India and Their Problems. Annals of the Association of American Geographers, 49: (2), 223-31. Ikram, S.M. 1961. History of Muslim Ci v i l i s a t i o n i n India and Pakistan. Bombay: Star Book Depot. Irschik, Eugene F. I969. P o l i t i c s and Social Conflict i n South India. Berkeley and Los Angeles: University of California Press. India, Dominion of I898. Madras Di s t r i c t Manuals - Coimbatore. Madras Presidency. Volume Two. Madras: Government Press. 1902. Census of India, 1901. Vol. XV-A. Madras. Part II. Imperial Tables. ¥. Francis. Madras: Government Press. 1912. Census of India, 1911. Vol. XII. Madras. Part I I . Imperial and Provincial Tables. J. Chartres Molony. Madras: Government Press. 1922. Census of India, 1921. Vol. XIII. Madras. Part I I . Imperial and Provincial Tables. G.T. Boag. Madras: Government Press. 1962. Census of India, 1961. Vol. IX. Madras. Part X - l . Dis t r i c t Census Handbook, Volume I I . Coimbatore: Government Press. 1972. Census of India, 1971. Paper I, Supplement 146. Provisional Population Totals. A. Chandra Sekhar, Registrar General and Census Commissioner, India. India, Republic of 1963. Census of India, 1961. Vol. IX. Madras. Part II-A. General Population Tables. Madras: Government Press. Kane, P.V. 194l. History of the Dharmasastras. Volume II (2 parts). Poona: Bhandtarkar Oriental Research Institute. Kalkarni, M.R. 1971. Industrial Development. India: National Book Trust. 86. Leacock, S., and Mandelbaum, David A. 1954-55* A Nineteenth Century Development Project on India - The Cotton Improvement Program. Economic Development and Social Change, 3: 334-351. Lewandosky, Susan 1975* Urban Growth and Municipal Development i n the Colonial City of Madras 1860-1900. Journal of Asian Studies, 34: (2) Lynch, Owen M. 1967. Rural Cities i n India: Continuities and Discontinuities. In India and Ceylon: Unity and Diversity. Philip Mason, Ed. New York: Oxford University Press. 1968. The Poli t i c s of Untouchability. In Structure and Change i n Society. Milton Singer and Bernard H. Conn, Eds. Chicago: Aldine Publishing Co. Mahar, M.J. 1972. The Untouchables i n Contemporary India. Tucson, Arizona: University of Arizona Press. Marriot, McKim 1955. L i t t l e Communities i n an Indigenous C i v i l i s a t i o n . In Village India. Chicago: University of Chicago Press. Mason, Philip (Ed.) 1967. India and Ceylon - Unity and Diversity. Oxford: Oxford University Press. Mattison, M. 1974. Tamil Muslim Merchants i n India's Industrial Development. In Entrepreneur-ship and Modernization of Occupational Structures i n South Asia. Milton Singer, Ed. Monograph and Occasional Paper Series, Monograph No. 12. Durham, N.C.: Duke University Press. Mehta. S.D. 1954. The Cotton Mills of India, 1854-1954. Bombay: The Textile Association, India. Mencher, Joan 1972. Continuity and Change i n an Ex-Untouchable Community of South India. In The Untouchables i n Contemporary India. Michael Mahar, Ed. Tucson, Arizona: University of Arizona Press. 1966. Kerala and Madras: A Comparative Study of Ecology and Social Structure. Ethnology, 5: (2), 135-71. Misra, B.R. 1970. The Administrative History of India. Oxford: Oxford University Press. 87. Murvar, V. 1966. Some Tentative Modifications of Weber's Typology: Occidental Versus Oriental City. Social Forces, 44: ( 4 ) , 38I-89. 1928-36. Mysore Tribes and Castes, The. (4 volumes). Mysore: Mysore University. Neelankant Shanterprasad (Ed.) 1971. Election Manifestos. Bombay: Hemangini Mehta. Opler, Morris E. 1972. North Indian Themes - Caste and Untouchability. In The Untouchables i n Contemporary India. Michael Mahar, Ed. Tucson, Arizona: The University of Arizona Press. Owens. Raymond 1974. Peasant Entrepreneurs i n a North Indian Industrial City. In Entrepreneurship and Modernization of Occupational Cultures i n South Asia. Milton Singer, Ed. Monograph and Occasional Paper Series, Monograph No. 12. Durham, N.C.: Duke University Press. Parathasarathy, V.S. 1958. Caste i n a South Indian Textile M i l l . Economic Weekly, 10: (33), 1083-86. Piggot, Stuart 1952. Prehistoric India. London: Penguin Books. Pocock, David F. 1960. Sociologies - Urban and Rural. Contributions to Indian Sociology, 4: 63-81. Poole, M.A., and Boal, F.W. Religious Residential Segregation i n Belfast i n Mid 1969: A ' Multi-Level Analysis. In Social Patterns i n Cities. B.D. Clark and M.B. Gleave, Eds. Porteous, J.D. 1974. Social Class i n Atacama Company Towns. Annals of the Association of American Geographers, 64: (3). Radhakamal, Muherjee 1961. Ways of Dwelling i n the Communities of India. In Studies i n Human Ecology. G.T. Theordersen, Ed. Evanston, 111.: Row, Peterson & Co. Ram, Mohan 1968. Hindi Against India. New Delhi: New India Press. Ramachandra, Chettiar 1939. Growth of Modern Coimbatore. Journal of the Madras Geographical Association, 14: (2), 101-16. 88. Ramachandran, Ranganathan 1969. Spatial Diffusion of Innovation i n Rural India: A Case Study of the Spread of Irrigation Pumps i n the Coimbatore Plateau. Ph.D. dissertation, Clark University. Redfield, R., and Singer, M.B. 1954. The Cultural Role of Citi e s . Economic Development and Cultural Change, 3: (1), 53-73. Rowe, William L. 1973. Caste, Kinship and Association i n Urban India. In Urban Anthropology. Aidan Southall, Ed. ' Toronto: Oxford University Press. Sangave, Vilas A. 1961. Changing Pattern of Caste Organization i n Kolhapur City. Sociological Bulletin, 11: (l&2), 36-59. Singh, R.L. 1956. Mirzapur: A Study i n Urban Geography. Geographical Outlook, 1: (1), 16-27. 1961+. Bangalore - An Urban Survey. Varnasi: Tara Publications. Singer, Milton 1972. When a Great Tradition Modernises. New York: Praeger Publishers, Inc. Sitcan, D.C. I960. Studies i n the Geography of Ancient and Medieval India. Delhi: Mohtal Banarasidass. Sjoberg, G. 1955. The Preindustrial City. American Journal of Sociology, 60: (5), 438-45. 1963. The Rise and F a l l of Cities: A Theoretical Perspective. International Journal of Comparative Sociology, 4: (2), 7-20. 1965. Cities i n Developing and i n Industrial Societies: A Cross Cultural Analysis. In The Study of Urbanisation. P.M. Hauser and L.F. Schnore, Eds. New York: Wiley. Smailes, A.E. 1969. The Indian City. Geographische Seitschrift, 57: Smith, V. 1958. The Oxford History of India. Third Edition. Percival Spear, Ed. Oxford: At the Clarendon Press. Sopher, David E. 1967. Geography of Religions. Englewood C l i f f s , N.J.: Prentice Hall, Inc. 89. Southall, Aidan (Ed.) 1973* Urban Anthropology. Toronto: Oxford University Press. Spate, O.H.K., and Ahmed, E. 1950. Five Cities ;of the Gangetic Plain: A Cross Section of Indian Cultural History. Geographical Review, 1*1: (2), 260-78. Spear, Percival 1965. The Oxford History of Modern India, 1740-19U7. Oxford: At The Clarendon Press. Spratt, P. 1970. D.M.K. i n Power. Bombay: Nachikete Publications Ltd. Srinivas. M.N. 1962. Caste i n Modern India and Other Essays. Bombay: Asia Publishing House. 1962. Industrialization and Urbanization of Rural Areas. In Caste i n Modern India. Bombay: Asia Publishing House. 1962. Social Anthropology and the Study of Rural and Urban Societies. In Caste i n Modern India. Bombay: Asia Publishing House, Theordorson, George A. (Ed.) 1961. Studies i n Human Ecology. I l l i n o i s : Row, Peterson & Co, Timms, D. Quantitative Techniques i n Urban Social Geography. In Frontiers i n Geographical Teaching. R.J. Chorley and P. Haggett, Eds. Vanhatarayappa, K.N. 1957, A Socio-Ecological Study. Bombay: Bombay University Press. Volwahsen, A. 1969. Living Architecture: Indian. Great Britain: MacDonald & Co. Weber, Max 19^ 8. The City. Translated and edited by Don Martindale and Gertrude Neuwirth. I l l i n o i s : The Free Press. Wirth, L. 1938. Urbanism as a Way of L i f e . American Journal of Sociology, kk: 90 APPENDIX I Caste Data For the Kongu Region i n 1901, 1911, and 1921 19011 19H 2 19213 Castes NO. $ No. % No. % Average Neutral Brahmin P i l l a i 3,111 1.73 0 36,820 2.11 1,741,667 39,4l4 2.00 1,969,188 1.95 1.95 Left A sari Cettiyar Mudaliar Naicker 50,377 2.84 45,258 2.55 63,553 3.57 SSylS^ 3.16 64,448 3.70 61,227 3.51 54,847 2.78 45,650 2.31 83,000 4.21 28,505 1.44 2.92 2.85 3.76 0.48 10.01 Right Kavuntar Nadar Tevar 690,155 38.96 79,416 4.48 2,605 0.16 639,557 36.72 76,907 4 .41 3,105 0.17 694,906 35.28 72,923 3.70 2,148 0.10 36.98 4.19 0.14 41.31 Service Udaiyar Vannan Navithar Pandaram 20,470 1.28 28,752 1.62 27,396 1.71 40,099 2.26 22,090 1.26 29,731 1.70 29,511 1.69 39,594 2.27 25,625 1.30 26,707 1.31 30,485 1.54 38,344 1.94 1.28 1.54 1.64 2.15 6.61 Immigrants (a) Naidu Boyar Malayali 24,402 1.37 1,826 0.10 33,348 1.91 28,344 1.43 5,711 0.29 5,623 0.28 1.57 0.09 0.12 1.78 Immigrants (b) Okkilier Gowder Reddi Arumthathiar Panikar Edayar Servai Achandar Sivier Kollar 53,139 3.00 64,018 3.6I 698 0.03 53,878 3.09 59,364 3.40 48,228 2.44 60,090 3.05 2.84 3.35 0.01 91 APPENDIX I - Continued 1901 1911 1921' Castes No. No. No. Immigrants (b) Vaniyar Ganga Pachakerar Maratiyar Vilayen Pannady Anglo-Indian Devanga Konar N. Indian Gurkha Singh Marwadi Sait 8,597 751 0.48 0.04 12,833 0.73 10,887 0.59 48,640 ll+, 887 2.74 0.84 43,911 2.52 16,951 0.97 53,105 14,811 2.69 0,75 12,562 0.63 Immigrants (c) Uppilayar Odde Jangam Dasari Bestha Batrazu Vedans Vettuvans Sudarman Valaiyan Sembadavan Nattaman Muppan Agamuddian Ambalakaran Dasi Golla Telaga Janappan 39,521 2.23 1,267 195 110 582 998 49,407 331 16,798 2,908 363 23,729 6,733 95 0.07 0.01 0.03 0.06 2.78 0.01 0.94 0.16 0.02 1.33 0.38 24,488 l.4o 27,998 1.49 12,119 20,666 19,188 2,759 28,650 3,567 3,873 11,301 O.69 1.18 1.10 0.15 1.64 0.20 0.22 0.64 3,477 72,817 23,312 4J988 34,499 13,975 4,490 0.17 1.26 3.69 1.23 1.18 0.91 — 0.01 0.25 0.08 0.01 — 0.24 1.75 1.90 0.01 0.77 0.93 0.22 0.17 0.00 1.34 1.43 0.19 0.52 0.24 0.28 0.09 — 0.21 8.89 Outside Religion Moslem Christian 23,286 27,003 1.19 1.37 92 APPENDIX I - Continued 19011 19112 19213 Castes No. % No. % No. % Average Scheduled Castes Pallar Valluvar Holeya Adidravida Paraiyan Chalckilayar Koraver 61,181 2.32 3,730 0.31 76,239 U.30 197,1+69 11.14 36,631 2.10 4,628 0.26 68,969 3.95 198,380 11.39 28,910 1.46 4,187 0.21 12,718 0.64 73,665 3.74 209,017 10.61 1.84 0.22 0.21 3.99 11.04 17.30 Scheduled Tribes Solaga Kadan Dommara Malasar Mudugan Kallar 3 5,1+60 0.30 H+1+ 1,142 0.06 4,102 0.23 510 0.02 1,861 0.10 — 4,571 0.23 4,023 0.20 0.17 0.02 0.07 0.07 0.03 1,593,025 ioo 1,741,667 100 1,969,188 100 100 Census of India 1901: Vol. XV-A (Madras), Table XIII, Part 2, pages 158-191+. Imperial Tables by ¥. Francis, I.C.S. (Madras: Printed by the Superintend-ent, Government Press, 1902). 2Census of India 1911: Vol. XII (Madras), Table XIII, Part 2, pages 115-123. Imperial and Provincial Tables by J. Molony, I.C.S. (Madras: Printed by the Superintendent, Government Press, 1912). 3Census of India 1921: Vol. XIII (Madras), Table XV, Part 2, pages 118-126. Imperial and Provincial Tables by G.T. Boag. (Madras: Printed by the Superintendent, Government Press, 1922). 93. APPENDIX II 1. Index of Segregation When the index of dissimilarity between one caste and a l l other castes i s combined, i t i s referred to as an index of segregation. When calculating the index of segregation, the areal unit under consideration i s crucial to the outcome. When the unit i s larger, the index i s lower. The formula used i n the calculation of this index i s as follows: Is = Irj 1 -tfXa-i_ i i n i •EXa^ = total number of a caste i n the c i t y . CXn^ = the total population of the c i t y . 2. Index of Dissimilarity To compute this index one calculates for each caste the percent-age of a l l the caste members residing i n each areal unit. The index of dissimilarity between two castes i s then one-half the sum of the absolute values of the differences between the respective distributions taken area by area. (1) Proportion of Brahmins (2) Proportion of area total to the city total (3) Difference between (l) and (2) Area I .21 .12 .09 Area II .05 .23 .18 Area III .28 .17 .11 .38 ID = .38 = .19 z Further, i n the calculation of this index the algebraic sign i s not taken into consideration. APPENDIX III Representation of the Number and Percentage of Each of the Castes i n the  Separate Sectors of Coimbatore City i n 1972 Castes Neutral Brahmin P i l l a i Xeft Asari Cettiyar Mudaliar Naicker Right Kavuntar Nadar Tevar Service Castes Udaiyar Vannan Navithar Pandaram Immigrants I (S. Indians) Naidu Boyar Malayali (1) Old Core No. 1097 336 1131 1604 225 1 1101 41 56 16.26 4.98  21721? 16.77 23.78 3.33 0.01 H37B9 16.32 0.60 0.83 17713 (2) Old Periphery No. 268 162 928 1569 665 961 978 20k 11 303 69 0 112 268 823 2.09 1.26 7.24 12.25 5.19 7.05 327IB 7.63 -1.15 1.59 "5707 0.08 2.36 0.53 0 2.97 0.87 2.09 6.42 973B (3) New Core No. (4) R.S. Puram No. % 24 14.11 1479 15.82 1126 12.98 14 8.23 355 3.79 416 4.79 2 2 7 5 * 19.61 17.77 3 1.76 274 2.93 64 0.73 11 6.47 1849 19.78 1806 20.80 7 4.11 471 5.03 296 3.41 0 0 223 2.23 173 1.99 12.34 30.12 26.93 12 7.05 312 3.33 1123 12.09 0 0 64 0.68 9 0.01 2 1.17 70 0.74 18 0.02 •BT22 4.35 13.02 0 0 278 2.97 1 0.01 2 1.17 276 2.95 63 0.72 2 1.17 187 2.00 89 1.02 0 0 3 0o03 6 0.06 2735 7.95 T78i 20 11.76 290 3.10 456 5.25 0 0 224 2.39 33 0.38 17 10.00 952 10.18 1122 12.93 21.76 15.67 18.56 (5) Devangapet No. (6) Industrial North No. 734 6.93 715 6.-75 137SS 589 502 286 132 910 379 1122 5.56 4.74 2.70 1.24 8.59 3.58 10.60 22.77 (7) New Suburbs No. (8) New Periphery No. 377 20.19 47 1.99c 192 10.28 230 9.76 30.47 11.75 61 3.26 91 3.86 234 12.53 139 5.89 0 0 5 0.21 0 0 0 0 15.79 9790* 170 9.00 47 1.99 0 0 312 13.24 0 0 10 0.42 9.00 15.65 0 0 0 0 16 0.85 56 2.37 0 0 52 2.20 0 0 0 0 oTBT E H 212 11.35 111 4.71 29 1.55 56 2.37 204 10.92 70 2.97 23.62 10.05 4=-APPENDIX III - Continued (1) Old Core (20 Old Periphery (4) R.S. Puram (5) Devangapet (6) Industrial North (7) New Suburbs (8) New Periphery Castes No. No. No. % No. No. No. Immigrants II (S. Indians) Okkilier Gowder Reddi Arumthathiar Panikar Edayar Servai Achanar Sivier Kollar Vaniyar Ganga Pachakeran Maratiyar Vilayar Pannwady Anglo-Indian 0 2 0 0 2 4 o 0 0 0 0 0 0 0 0 0 0 (N. Indians) Punjabi (Singh)19 Marwadi I 39 Sait I 0 Outside Religion Moslem Christian 159 55 2.35 0.81 552 226 7 0 0 442 0 0 138 0 0 0 0 50 0 0 0 U.30 1.76 0.05 0 0 3.45 0 0 1.07 0 0 0 0 0.39 0 0 0_ 11.02 3149 24.58 63 0.49  2^ 707 16 0.18 0 0 269 3.10 298 3.43 626 7.21 0 321 185 629 0 30 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3.03 1.74 5.94 o 0,28 0 0 0 0 0 0 0 0 0 0 c_ 10.99 0 0 _0_ 0 973 9.19 1128 10.65 19781 110 157 5.89 8.1+0 1O9 0 0 0 0 0 0 o 0 0 o 0 0 0 0 0 0 0 0 0 0 li86 20.62 10 0.1*2 2iToH NO APPENDIX III - Continued (1) Old Core (2) Old Periphery (3) New Core (4) R.S. Puram (5) Devangapet (6) Industrial North (7) New Suburbs (8) New Periphery Castes No. No. $ No. $ No. $ No. $ No. No. No. Scheduled Castes Pallar Valluvar Paraiyan Chakkhayn Devendrar Koravar Kurumbar Unknown 42 0.62 3 o.oU 0 0 7 oao o77§ 0 0 15 0.22 0 0_ 0.22 195 2.89 302 0 0 587 0 0 34 0.23 0 0 4.58 4781 o 0 0.26 "5T2S 43 0 0 0 0 0 0 25.20 0 0 0_ "23720 0 0 0 0 0 77 0 0 0 0 0 0 0.82 0 0 oTBl 0 0 0 0 0 0 0 0 0 0 0 0 0 0 _0_ _0_ 0 0 _p_ 0 305 0 0 0 83 39 90 2.80 0 0 0 2785 0.78 0.36 0.85 1789 15 0 0 0 0 0 0 0.80 0 0 0 "5780 0 0 0 0 243 1.89 100.00$ 100.00$ 8 4,7 100.00$ 535 5.72 i4o i .6i 205 1.93 0 0 436 0 0 i4o o o o 58 100.00$ 100.00$ 100.00$ 100.00$ Source: The caste composition of each sector was computed from the original data collected from the National Malaria Office i n Coimbatore. VO ON 97 GLOSSARY Acari (Asari) A caste whose traditional occupation i s considered to be that of a craftsman who works with gold, brass, iron, stone, and wood* The two subcastes described i n this account are the CoLi and the Konku Acari. Both are considered to be members of the l e f t division. Brahman A caste whose traditional occupation i s considered to be scholarship and the priesthood. The members of this group enjoy an elevated r i t u a l status i n the Konku region. Except where otherwise specified, the persons described i n this account are members of the Aiyar sub-caste. They are considered to be above the r i g h t - l e f t division. ceri (cheri) Ah untouchable settlement or l i v i n g area. CeTTiyar (chettiar) A caste whose traditional occupation i s considered to be business. Except where otherwise specified, the persons described i n this account are members of the KomuTTi sub-caste. They are considered to be members of the l e f t division. Chavadi Public building i n a village. Cutcheri Assembly. KavuNTar (Gounder) A caste whose traditional occupation i s considered to be farming. This group controls most of the cultivable land i n Konku today. Except where otherwise stated, the persons described i n this account are members of the Konku (or VeLaLar) sub-caste. They are considered to be leaders of the right division. Kohku (Kongu) A large cultural and historical region, comprising most of the north-west corner of the present Tamilnadu (Madras) State. Kovan Puthur The name by which the earliest settlement of Coimbatore was known. Kuravar A caste that i s classified as "untouchable" and whose traditional occupation i s considered to be basketmaking. The subcaste discussed i n this account i s known as KuTai. Its members are said to belong to the l e f t division. 98 Matari (Madari) A caste that i s classified as "untouchable" and whose traditional occupation i s considered to be leatherworking. The caste i s known elsewhere as Cakkiliyan. Several sub-castes are discussed i n this account. They are a l l said to belong to the l e f t division,, Mutaliyar (Mudaliar) A caste whose traditional occupation i s considered to be weaving, military service, and business. Except where otherwise stated, the persons described i n this account are members of the Kaikkolar sub-caste. They are considered to be members of the l e f t division, but their position appears to be ambivalent as, i n practice, they exhibit some characteristics that are typical of the l e f t and others that are typical of the right. Nakar, Pattinam, Padi, Pari, Poral, Vity, Ur Synonyms for the word c i t y . NaTar (Nadar) A caste whose traditional occupation i s considered to be that of palmyra-palm climbers (toddy-tappers). Except where otherwise spe-c i f i e d , the persons described i n this account are members of the Marameri sub-caste. They are considered to be members of the right division. Nlvitar (Navidar) A caste whose traditional occupation i s considered to be that of barber. The two subcastes discussed i n this account are the Konku and the PaNTiya. The f i r s t are considered to be members of the right division, and the second members of the l e f t . Nayakkar (Naicker) A caste whose traditional occupation i s considered to be well-digging, stone excavation, and road-building. Except where otherwise speci-fied, the persons described i n this account are members of the VaTuka sub-caste. They are considered to be members of the l e f t division. Paraiyar A caste that i s classified as "untouchable" and whose traditional occupation i s considered to be drumming. The sub-caste discussed i n this account i s known as Konku. Its members are said to belong to the right division. Petta Market place near a town. PiELai ' A caste whose traditional occupation i n the Konku area i s considered to be accountancy. Except where otherwise specified, the persons described i n this account are members of the KaruNikar sub-caste. They are considered to be above the rig h t - l e f t division. Puthu New. Ur Village or settlement. UTaiyar (Udaiyar) A caste whose traditional occupation i s considered to be that of potter and house-builder. Except where otherwise specified, the persons described i n this account are members of the Konku sub-caste. They are considered to be members of the right division. VaNNar A caste whose traditional occupation i s considered to be that of washerman. The two subcastes discussed i n this account are the Konku and the VaTuka. The f i r s t are considered to be members of the right and the second members of the l e f t division. Vidhi Street. 1 0 0 . A Note on the Limitations of this Thesis A thesis such as this, that deals with spatial organization, has many strong as well as weak points. Its greatest strength l i e s i n i t s sensitivity to the variety of physical form and patterns and, to some extent, the uniqueness of a colonial city such as Coimbatore. Further, i t lays considerable emphasis on a historical approach. It also takes the individual household as i t s unit of study. From the analysis i t was concluded that the spatial organization of Coimbatore City hest represents a curious compromise of the traditional village and regional patterns i n conjunction with emerging western spatial patterns. However, while the above conclusions i l l u s t r a t e the way space defines and separates the urban landscape into distinct areas of status equivalence (e.g., economic, residentially pure and polluting) (Hazlehurst 1971: 1 9 0 ) , i t throws l i t t l e light on the social processes that link to-gether the urban landscape. I t i s actually these social processes that necessitate spatial distinction and separation. It i s hoped that other studies of a related nature w i l l overcome these shortcomings and take into consideration both space and social processes. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items