Historical debates have a quaint air to them. As soon as they lose political relevance, which may or may not necessarily mean any intrinsic worth to a debate, they lose their shine. Think of the debate that took place more than thirty years ago in the seventies continuing to mid-eighties of the last century over mode of production in India, particularly in agriculture – whether India was semi-feudal, or semi-colonial semi-feudal or capitalist or retarded capitalist, something that must strike you is the scarcity of references to caste as a particular factor in shaping the mode of production. Yet today it is not difficult to recognize that caste works as an axis of post-colonial capitalism. Therefore there is the need to now theorise the ways in which agrarian capital or capital in the rural hinterland is influenced by caste and influences caste dynamics. Equally there is now a need to study the overwhelming presence of caste as a predicating factor in shaping the destiny of rural labour as agrarian labour or as surplus population absorbed in various forms of petty production, or as a reserve army of labour, or simply forming the vast army of footloose labour in the country. The relative absence of caste as a critical factor in a discussion on post-colonial political economy is stunning.
Partly the reason is with the categories which our discussants were working with at that time. With received categories they attempted to work out the implications of the dynamics they were studying – green revolution, mechanisation of agriculture, form of rent, form of wage, impact of grain trade on production, mode of investment in agriculture, etc. In this connection we may advance three tentative reasons for which caste was relatively absent in post-colonial political economy:
First, in this exercise there was little scope for an organisational analysis – in other words how the labour market is organised, or for that matter capital market. In other words we were not attentive to the organisational and structural process of accumulation in the countryside. But this should not have been so. This is because at the same time we were getting enriched with studies on the interface of caste and capital formation in trade and production, discussions on merchant castes, bonded castes - let us say a specific merchant banking caste like the Chettiar. We had studies of (for instance by David Rudner) of the interdependence among Indian business practice, social organization, religion, impact of colonial rule on indigenous commercial systems, and the inextricable links through formal and informal institutions between caste, commerce, production, and wealth. Practices crucial to the formation and distribution of capital also formed part of the study of the links, which could include factors like marriage alliance, status hierarchy, and related rational conduct of business. And remarkably these studies were suggesting ways ways to study caste and agrarian labour and rural labour in general.
Second, we were neglecting what Charles Tilly termed as durable inequality, which would mean the organizational factors determining patterns of employment, purchase, investment, etc. We can say that race, caste, gender, or tribe – all these constitute various aspects of the organization of society in which homogenizing tendencies like accumulation, equality, etc. have to operate. Inequalities endure, because for instance, civil rights legislation addresses only the situation in which a hiring boss, or committee, employs or promotes one person but not another in a formal system of straight competition. But as Tilly said, if we change the story, like this: We have a large grocery store where there are different departments - one of them meat, the other vegetables. It turns out that women move into the vegetables and the men into the meat. At entry level they are paid the same. No discrimination. Walking about the store the men go to places where they can talk to other men and swap stories about football. Women go to places where there are other women and they can swap stories about prices of vegetables, or children’s school fees, etc. But promotion structures work differentially so that managers mainly come from the meat department. There is no point at which you have two candidates, one of them male, one of them female. No discrimination at point of entry; or even promotion. But what you get over time is an invisible structure which is, in effect, discriminatory. Therefore what is needed is again to learn about linkages between enclaves and ghettos, about networks, trust stocks, and about migrations. So it is the process, no need to go to the economics professor or to Walmart or the Imperial Chemicals to know all these. So if we have to know the particularity of the dynamics in which caste works as a factor in accumulation we have to inquire the process in which some castes stream into some occupations and some into other. The answer is in the processes of self-reproduction. The phenomenon of “opportunity hoarding”, makes the dynamics of social mobility uneven. Overall social mobility has not changed much, and the possibility of moving from the agrarian sector into the elite meanwhile has diminished drastically. What is “opportunity hoarding”? It operates when members of a categorically bounded network acquire access to a resource that it considers valuable, renewable, supportive of the network, and enhanced by the network’s modus operandi, and thus subject to monopolistic control. “Opportunity hoarding” thus makes inequality durable. The concept allows insights into inter-group dynamics, race, ethnicity, identity, which include national units also in this inter-group dynamics. Economics rarely combines identity and interest; it takes into account only the latter. History emphasises mostly the former. Tilly wanted to give us an approach based on a combination. The implications of this approach, I suggest, is deeper. Tilly said large, significant inequalities in advantages among human beings correspond mainly to categorical differences such as black/white, male/female, citizen/foreigner, or Muslim/Jew rather than to individual differences in attributes, propensities, or performances”. In this case we have two elementary forms: types of social relations and inequality-generating mechanisms. We can see how a combination of these two elements leads us to an organizational analysis of inequality, organizational view of heterogeneity-generating mechanisms. Thus although historical accumulations of institutions, social relations and shared understandings produce differences in the day-to-day operation of various sorts of categories (gender, race, citizenship, and so on) as well as differences in various sorts of outcomes (e.g. landed wealth versus cash income), ultimately interactions of exploitation and opportunity hoarding explain them all.
Third, to grasp the relation between caste and accumulation, we have to use the notion of frontier in the way Bailey had long back used the word in his study of caste and the economic frontier. There Bailey had analyzed changes in the internal organization of a comparatively isolated village in the hilly section of eastern India, as a result of what he termed as the extension of the economic and administrative frontier, producing contacts qualified by caste, government policies, and the fact that the village studied by him consisted mainly of Hindu colonists in a predominantly non-Hindu tribal area. What is interesting in this is not the suggestion of a modernization thesis, which was there and is banal now, but the way the frontier of economy works in society. Think of the frontiers of accumulation, you will see how the multiple axes work. Caste is in fact the grease for the accumulation machine. The way low castes had their different destinies in districts and villages of West Bengal and Bihar shows how the accumulation machine works.
Now let us correct a bit to what was said in the first paragraph. It is not that caste was absent in the discussion. The presence of caste was considered as evidence that India was semi-feudal; the erosion of caste was evidence that India was capitalist. Nothing of the two, caste was simply sublimated in the political economy of post-colonial capitalist accumulation.
Let us now gather some evidence in support of our argument. Our evidence comes predominantly from the state of West Bengal, where it is variously argued that caste question appears to be subsumed in the larger issues of class and communalism that accounts for the continued hegemony of the upper caste bhadralok in society, polity and culture. Further, it is argued that in West Bengal there exists a neat ethnic separation between the composition of the commercial-industrial elite and the cultural-political elite. If the former is dominated by the Marwari and Gujarati houses of central Calcutta, the latter is populated predominantly by the upper caste urban bhadralok. Often, scholarly discussions on capital accumulation in West Bengal exclude the caste structure among Bengalis and focuse on the management of family and kinship capital of the Gujaratis and Marwaris. While discussing entrepreneurialism among members of religious groups other than the Hindus the scholars tend to look into family and kinship infrastructures without necessarily taking a caste optic. What follows then, we have very little idea about the dialectics of caste and capital in West Bengal.
We have chosen the Mahisya caste in Howrah for an initial exploration of the questions we have just mentioned. The 1931 Census (the last census before the Census of 2011 that enumerated caste) placed the Mahisyas in the middle order in Bengal’s caste hierarchy. Between 1920s and 1970s they played a very prominent role in running several engineering workshops in the city of Howrah. We extensively use the rich social survey of these workshops by R. L. Owens and Ashis Nandy in 1968-69 (later published as a monograph: The New Vaisyas: Allied Publishers, New Delhi, 1977) and the census data from 1921 to 1951 to arrive at certain conclusions about the relationship between caste and capital accumulation in Howrah. We then proceed to present some tentative results of a very preliminary field survey and arrive at a framework of analysis for a future project on the subject.
One of the authors of the current piece has a hazy boyhood memory of strolling through the crowded Balilious Road of Central Howrah with a prominent local leader of the CPI in a winter evening of 1988. The road was teeming with small industrial units in both sides. A moulding unit was situated amidst a dense cluster of shops selling food, tools and machines; a small unit hosting three precision grinders was choked with the smoke and heat of several surrounding units. In small workshops one could be easily baffled by the presence of all kinds of machines: lathe, capstan, shaping, grinding, punching, slotting, and so on stuffed in a dark shed measuring around 200 sqft., producing a range of simple tools and complex machines. The little sheds left very little space for the workers to navigate around the machines. Most of the machines were already consumed their expected lifetime. It was a time when deep recession starting nearly two decade back coupled with a crumbling electricity infrastructure and an emerging regime of environmental law already crippled the industrial climate of the Balilious Road.
The engineering industry in Howrah has a long career. It started in 1850s when Howrah became one of the central termini of a fledging national railway network. In the initial years, it serviced the construction of the rail roads. During the boom periods of the two World Wars and then the first decade of planning after Independence it developed into a vast and interlinked network of large firms employing up to a thousand employees and small workshops employing often less than ten workers clustering mostly around the Ward 39 of the city Corporation and the Baililious Road. The relation between the two types of units was usually of contracting work from large to the small. The surplus demand of engineering goods sustained such a system of interdependence for a long time in the 20th century. The ethnography of Owens and Nandy in the Ward 39 revealed that at least since the time of the World War II a majority of the small workshops were owned by the entrepreneurial Mahishyas who mostly employed Mahishya kinsmen as workers in the workshops. The rise of the Mahishyas in dominating the engineering industry surpassing the erstwhile dominant groups from the Brahman and Kayastha sections of the Ward 39 appeared to be intriguing to Owens and Nandy taking into account the fact that the Mahishyas were never previously associated with commerce. In vast fertile area comprising south and eastern Midnapore, Howrah, Hugli and 24 Parganas the Mahishyas were the enterprising peasants being enumerated in the census of 1931 as the ‘single most important middle caste’ in south-western Bengal. In Midnapore and Howrah they constituted 31.56 percent and 24.92 percent of the total population respectively.
How then to account for their interesting switch from agriculture to industry already in the second decade of the 20th century? Owens and Nandy appear to believe that this was not an abrupt or radical transition. Already in the mid 19th century, Howrah was more urbanized and industrialized than any other districts in Bengal. The wage differential between agriculture and industry was already considerably high coupled with a sharp differentiation among the agricultural work-force. According to the census of 1921, in Howrah, the share of landless labourer in the total agricultural work-force was 30.05 percent which was only second to Birbhum (35 percent) and much higher than other districts of western Bengal where, as we know, differentiation among peasantry was considerably higher than the jute producing eastern Bengal that sustained small tenure holdings and small scale peasant agriculture for a long time. In 1931, the percentage of the landless labour in Howrah further increased to 44.64 percent keeping pace with the general trend in many western Bengal districts. In 1951, however, the share of the landless labour in Howrah declined to 38.2 percent. If the presence of the landless work-force amidst wage differential between agriculture and industry meant a growing tendency among the population to change their principal source of earning, the gradual decline in the percentage of landless labourer in agriculture between 1931 and 1951 implied that part of this population was absorbed in the non-agricultural sectors. Additionally, one can observe that between 1911 and 1931the number of Mahishyas involved in cultivation in Howrah dropped by no less than 20 percent while there was an increase in the number of landless labourers in agriculture by 12 percent and 7 percent increase in the number of the ‘white color’ workers. We, then, find little reason to challenge the larger ethnographic finding of Owens and Nandy that the engineering workshops were run by the ‘local’ Mahishya labour-force for two generations.
In the next step, Owens and Nandy concluded that the Mahishya entrepreneurs of the 1960s and 1970s had prior knowledge of the operation of the workshops as they emerged from the Mahishya working class. During the boom of the First World War many of workers of the existing engineering forms opened their own workshops to respond to the surplus demand created by the war. They invested capital from their meagre savings and became millionaires in two decades. The old entrants to the industry, the propertied Brahmans and Kayasthas gradually lost ground. Their defeat was reflected in the upward mobility of the Mahishyas in socio-cultural world of the Ward 39 since the early 1920s. The first Mahishya founder of the engineering firm gradually displaced the two important Brahman families in terms of controlling the real-estate of the said ward and sent his son to a prestigious English school. The son, in his mature age, entered into the Congress politics and became MP from the Uluberia parliamentary constituency in 1967. Owens and Nandy argued that the ex-peasant and the ex-working class background of many of the Mahishya entrepreneurs enabled them to cultivate an ethic of work and self-help that the members of the erstwhile rentier castes crucially lacked. Such a privileged ethic of work became crucial at the time of slump and industrial unrest as unlike their upper caste counterparts the Mahishya entrepreneurs themselves had the capacity to run the work of production. Thus, in a separate paper, Owens observed the following:
I recall visiting the factory of one Mahishya entrepreneur in my sample who had come upon very hard times as a result of recession in process. The entrepreneur, who in previous interviews had been very much the ‘Babu’, dressed in spotless white dhoti-jama and supervising the work of several other labourers, was at our last meeting alone in his workshop, dressed in a workman’s greasy T-shirt and lungi and working at a lathe. Wiping the grease off his hands as he sat down for the interview, he observed, ‘I used to work at the lathe, and I can still do it when I have to’.
In another separate paper, Nandy (1973) confirmed Owens’ observations about the work ethic among the Mahishyas. He argued that the ‘positive attitude towards manual work’ gave the Mahishyas a ‘competitive advantage’ over the upper castes that ultimately accounted for their success in industry.
We hope that the readers will not take this note as a celebration of the long legacy of Weberian sociology. This prompts us to search for an alternative explanation. We hold that one cannot make a general argument about the Mahishya entrepreneurs that they stemmed from an erstwhile Mahishya working class without furnishing evidence about the formalized banking infrastructure and other sources of savings and credit available to kick off even a small workshop. Moreover, from a Marxist perspective it seems naïve to argue that the workers would own the means of production at a mass scale in a given industry within the capitalist mode of production by means of savings from their wage coupled with some meagre wage subsidy from agriculture in the extended family and the ethic of ‘self-cultivation’ (whose destiny is to become an entrepreneur one day!!). The initial source must have come from outside the particular circuit of production, i.e. from agriculture, trade, money-lending, and so on. Since we do not have data about banking and credit facilities available to the particular segment of workers in early 20th century Howrah, we search for some alternative data from other sectors of economy, particularly from agriculture and the prevailing agrarian structure in western and south-western Bengal. We then frame a conversation between such relatively macro data with our preliminary exploration of the history of one important Mahishya family in the industry.
Let us first look at the trends in land transfers in Bengal. Here, we have to remember that the reliable data comes to us only after the passing of the amendment to the Tenancy Act in 1928 that made the transfer of occupancy holdings a statutory right. Over all, scholars of agrarian change in colonial Bengal agree that transfer of land from the small tenure holders and ryots to the rich peasants increased significantly in the 1920s. The trend was checked significantly in 1929-30 as the 1928 Amendment made it mandatory to pay 20 percent transfer fee to the landlord. However, in the Depression years of the 1930s the rate of transfer augmented considerably. (Registration Department Sources, Binay Chaudhuri, cited in Partha Chatterjee, 1986: 143):
Number of Sales
Number of Mortgages
Between 1930-38, Partha Chatterjee calculates from the Tables on ‘operations under the Bengal Tenancy Act’ that the total sale and mortgage of land in the Mahishya belt of Midnapore (16.86 percent), Hoogly (12.72 percent) and Howrah (12,25) was to the higher limits to the prevailing south and western Bengal standard. In 1930-31 the total mortgage and sale of land in Howrah (2.38 percent) was lower than just three districts in the undivided Bengal—Noakhali (3.25 percent), Bogra (2,42 percent) and Pabna (2.42 percent).
The data reveals a greater concentration of land in the hands of a few. As the historians of Bengal’s agrarian transformation agree, those who benefited from this situation were not the members of the old rentier class such as zamindars and patnidars, but an upwardly mobile rural jotedars representing various middle strata of Bengal’s complex caste ladder. It has been well documented by historians of colonial Bengal’s agrarian relations that the late inter-war period witnessed the weakening of the control of the up-caste bhadralok zamindars in the villages of south-western Bengal. The power of the agrarian economy came to be decidedly controlled by middle-caste jotedars/moneylenders/traders and peasants having substantial land holdings who were “partially rooted in production at the village levels”, but operated generally as “a kulak class engaged in a variety of economic activities” such as grain trading, and agricultural processing industries(Chatterjee 1986: 178). The upward mobility of the mahishyas in Midnapore, the agurisin Burdwan, and sadgops and tilis all over the south Bengal has been well documented by historians (Ray 1975). While explaining the growth of nationalist mass movements in south-western Bengal, H. R. Sanyal, for example, tells us how political leadership emerged from the jotedar/moneylender/trading classes that substantially challenged the zamindari sections on the question of village control through newly established union boards, then by capturing anti-union board mobilization (see also Chatterjee 1986).
The Mahishyas of Midnapore and Howrah featured the emergence of several landholding families in 1920s and 30s who also invested substantially in getting land leases in Sundarbans. We interviewed one of the prominent members of the fourth generation of one such substantial family in Howrah. Already in 1910, the founder of this family, Anukul Mondal, shifted from Bauria to the Ward 39 of the city of Howrah. Until that time, as Owens and Nandy documented, the community in the said Ward was dominated by two Brahman gentlemen Jogin Mukherjee and Keshab Chandra Banerjee who owned most of the real-estate property of the area, conducted panchayats and a grand month-long feast at the time of Durga Puja. Between 1885 and 1920, Banerjee won local elections by combining his caste position with the glory of being a revered landlord. The Mahishyas were still far inferior in caste hierarchy and mostly engaged in low-paying industrial jobs. Around the first decade of the 20th century Anukul Mondal mortgaged 105 bighas of his agricultural land to buy 17 kotthas of land in Ward 39 from an improving Suvarnavanik family. Anukul build an impressive residential complex at one part of the land and kept the rest for an Engineering workshop which, by 1915 employed no less than 22 workers (18 of whom from his caste group). In 1921, he bought another prime plot of 10 kotthas from Keshab Banerjee to open a second unit of the same workshop employing an additional 40 workers, 32 of whom came again from the Mahishya caste. By 1920, the economic and political fortune of both Jogin Mukherjee and Keshab Banerjee declined drastically. Owens and Nandy recorded that while the former invested in a coal mine that produced water, the latter made an unsuccessful investment on a mining project that ultimately compelled him to sell his substantial real-estate holdings in the adjoining areas of the Ward 39 to Anukul Mondal who, by 1924, started a modest life insurance agency in Howrah with a branch office at Hazra Road, Kalighat. In 1938-40, Anukul sold off his possessions in Sundarbans and invested in a number of medical and departmental stores in Bhawanipore-Kalighat area. The descendants of his second and third sons still run these shops. Anukul’s first son, Jugal Mondal, was sent to the prestigious English medium school named Bishop Westcott Institution. Jugal Mondal, in his mature age, became one of the most influential politicians from the Mahishya caste in Howrah winning Lok Sabha elections in 1967 with a Congress ticket from Uluberia constituency. Later in the mid 1970s Jugal Mondal joined the Janata Party and in 1978, lost in a Rajyasabha seat from West Bengal. He is known to the community as a philanthropist who invested in schools and health centres in different Mahishya dominated villages. Jugal Mondal was associated with over 300 clubs and societies in Howrah and Calcutta and maintained strong linkages between Mahishya caste associations in rural and urban Bengal.
In West Bengal, the middle caste Kulak interest never succeeded in forming a powerful political party to safeguard its economic position and frame a distinct political goal. The Bangla Congress could at best be called one such moment in the late 1960s when kulak interests converged regionally against a potentially radical food movement (Chatterjee, 1997). It is well-known how that bloc crumbled within two years of its formation. This is not to deny their strong presence in district committees of all the major political parties in the state. By operating through the long-established and regionally dominant political parties they became the highest beneficiaries of the nationalization of Banks. They still control the commodity traffic in the complex network of national and state highways connecting different parts of the state with the major wholesale centres of Central Calcutta.
The small story of the Mahishyas in Howrah then is just one instance of a larger economic and social process in late colonial and early post-colonial Bengal. We have fragmentary evidence to show that these newly mobile agrarian kulak classes started investing in Calcutta’s real-estate and retail sectors in the early 1940s taking privilege of the war-time decline in the value of property. Here, one must mention that the impulse to diversify investment from the rice market and rice mills to real-estate development and urban retail came in 1930s, at a time when the price of rice started declining in the depression years and owing to among others the government’s procurement of cheap Burma rice. More substantial jotedars also purchased rent-collecting rights in the bazaars from the old zamindari houses in Calcutta. They also set up new wholesale spaces in central Calcutta. The Roy family of Arambag in Hoogly, for example, started investing heavily in the meat and poultry supply chain in Calcutta, while another Koley family from Kotulpur, Bankura invested initially in vegetable, fish, poultry wholesaling (establishing the Koley market in 1932), and later in food processing industry. The Koley market near Sealdah station became one of the major wholesale-retail centres for vegetables and fish in Calcutta. The Koley biscuit company in Narkeldanga (a segment of the same Arambag Koley family) also heavily invested capital in the city’s sports culture and thereby started claiming legitimacy in the city’s bhadralok public culture. Some others from the same family participated in active politics since the Subhas Bose-J. M. Sengupta era in the Calcutta Corporation. While more substantial middle-caste groups invested in urban provisioning from the third decade of the last century, the less substantial jotedar families invested capital in urban retail. However, the kulak-jotedar capital remained essentially fragmented across multiple caste lines and they tried in 1950s to capture the spaces left vacant by substantial Marwaris who, from 1930s, started replacing the British capital in Bengal’s veritable jute industry. However, they were extremely successful in establishing long-lasting ties with political parties through representation and everyday finance and subsequently gained from the abolition of zamindari, tenancy reforms, bank nationalization and subsidy.
The Great Bengal Famine in 1943 gave a serious blow to the already shrinking caste-based specialized artisanal retail chains in the city. The newly mobile middle caste groups started buying these moribund shops, and gradually replaced them with the more heterogeneous departmental and grocery stores. This change is well documented in Calcutta Corporation’s property and shop register books. A similar change can be found in less substantial real-estate properties too.
During our field research, we came across a large grocery store at Ashutosh Mukherjee Road in Bhawanipore. The store was established in 1944 by one Bimal Choudhury whose ancestral home, we were told by Choudhury’s great grandson, was in Kirnahar in Birbhum district. Bimal’s father Nagendra Choudhury belonging to the middle caste aguri community of the middle-West Bengal region earned a lot of money and landed property in Kirnahar by lending money to some of the declining segments of the Sircar family of the village who had the rent collecting right as a major patnidar of the Maharaja of Burdwan. In 1932, the Sircar family sacrificed its rent collecting right. The Chaudhuris earned one fourth of the rent collecting right from the Sircars as a part of the repayment of the outstanding loans that the latter had incurred from them. Bimal as the youngest son of Nagendra inherited a commercial land plot at Bhawanipore, Calcutta where he later established a grocery shop, a house in Sovabazaar that his father bought from the Deb family of Sovabazaar back in 1938, and some landed property at Kirnahar. Bimal also bought two stores in Shyambazaar (originally selling copper utensils) in 1948 and converted them to grocery stores. Our evidence from many such families in Calcutta attests to the fact that in 1930s and 1940s there was a major change in the caste composition in Calcutta’s wholesale and retail and real estate control.
The material wealth of jotedars did not only come from their superior land control. They also controlled agricultural labor by participating in credit chains. Sharecroppers lacking capital for procuring seeds and food cultivated for half share of grain with loans advanced by the jotedar. Owing to credit compulsions the poor farmers and sharecroppers sold their produce to the local grain dealer (byapari), who was often the jotedar, without waiting for a higher market price. The jotedar being equipped with storage capacity (gola) would hoard the grain for months to sell them in the middle of the cultivating season when the grain prices would sore up again (Ray 1975). However, due to the fragmented nature of the middle caste in Bengal, the jotedar capital could not establish a monopoly ethnic control over the real-estate or the segments of the wholesale trade. Nor could they come up with a formalized caste based banking network that a number of middle-caste trading communities could achieve in the South.
Let us go back to the two comments made at the outset. First, to understand the role of caste in accumulation process we have to go deep into organisational analysis, which as we have tried to show in this note would include analysis of supply chains, networks, property transfers, and several allied factors that result in durable (in this case ascribed by specific caste identity). The organisational analysis enables us to grasp what is at stake in the phrase “frontiers of post-colonial accumulation”, by which indicate a dynamic process of interface between the colonial and post-colonial society and the immanent logic of political economy. Second, the debates in the seventies and eighties of the last century over mode of production were like all social science discourses were acts of displacement, for while caste and capital were working in several ways in shaping and re-shaping one another precisely in those decades, caste had already been sublimated in the framework of an academic debate.
If we go back to the Mahishyas at the engineering workshops in the city of Howrah, what appears striking to us is the near absolute homogeneity over the caste composition of the entrepreneurs and the workers. If we accept Owens’ and Nandy’s conclusion that the entrepreneurs emerged from the workers, we are left with very little to proceed further. But, we believe that one needs to inquire into the operation of caste based informal associations in coordinating many extra-economic relations and loyalties in recruiting local labourers and in mitigating conflicts between labour and capital. It is then that we are able to understand the layered relation between caste and capital accumulation. We also intend to study the mechanisms of recruitment of workers in these small firms where one still has very little evidence in support of the presence of a migrant labour force. This paper is just the beginning of a larger project.