Unfree labour relations have attracted considerable attention in the recent literature (Bhaduri, 1983; Patnaik and Dingwaney, 1985; Bharadwaj, 1985; Brass and van der Linden, 1998).  Some argued that unfree wage labour reflects not only underdeveloped labour markets, but also conjoint exploitation in interlocked markets (Srivastava, 1989) and an incomplete transition to the capitalist mode of production.  By contrast, Brass (1990, 1992, 1997a) and Miles (1987) take an opposing view that unfree relations are compatible with capitalism, comprising part of capitalists' assault on the autonomy and wages of labour.  Brass has argued that convict labour and indentured labour were subject to capitalist exploitation.  Analysis of the relations of production should be located in concrete historical context, and we find evidence that suggests bonded labour has roots in precapitalist structures (and ideology) and is used by capitalists in modern contexts.  Given the remarkable diversity in Indian agriculture, any generalisation from one case study would risk trivialising or oversimplifying the situation.  The present paper is concerned more to identify the concrete elements of exploitation and oppression in one case study.  The concept of interlocking markets is used to help grapple with the nature of class exploitation.  We also link interlocking markets to other institutions of oppression, such as caste and patriarchy.  The case study quantifies contracts at household and individual units of analysis while also broaching the qualitative question of inter-caste and intra-household social relations.  

The paper is divided into four sections.  Firstly, a historical perspective on bondage in India is given and a brief review of literature is provided.  Secondly, we describe the origin and growth of Palamuuru contract labour, and present Ramana Murthy’s field observations on the working of the contract system in six villages of Andhra Pradesh.  Thirdly, we provide some analysis of exploitation in the system before concluding.  Our main innovation is to stress the dialectic between interpersonal modes of oppression and structural forms of oppression, thus extending the labour relations agenda to include interpersonal violence, sexual acts used to control or manipulate others, and other dyadic interactions that support and perhaps embody structural forms of oppression.  
 

I. Introduction

Thus the colonial rulers inherited specific forms of forced labour.  They introduced new variants such as indentured labour for plantations and for export to the Carribean and Fiji.  Indentured labour was only opposed by the Indian Congress in 1909, after it was a well entrenched practice that had underpinned huge flows of migrants.  However, the practice of indenturing labour continued for plantations and for public works within India even after that time.  Public works such as the building of roads, railway tracks, irrigation facilities, government buildings, military airports and barracks emerged as an important sector for which bonded labourers were recruited by contractors (Sarkar, 1985).  The terms of bondage in all these systems were not alien to the labourers.  Instead existing discourses of debt and obligation were adapted to create new variants.  The post-colonial state inherited many of the practices of its predecessors which facilitated blatant exploitation of workers in large-scale public works (Kerr, 1998).  Private contractors have thrived by exploiting cheap labour in several ways, such as: advancing wages and trapping people in bondage; transporting people to places of work alien to them; paying abysmally low wages; controlling the food and housing of seasonal migrants; imposing hazardous working and living conditions; and extracting phenomenal amounts of surplus value, thus reproducing the debt relations. 
 

Interlocking markets and unfree labour.  A historical understanding of unfree labour relations in India shows that bondage has been conditioned by usury and denial of subsistence.  As others have pointed out, the imposition of everyday forms of oppression often operates through the institutions of caste.  Lerche (1998) offers a convincing case study of political tactics in two regions of Uttar Pradesh, where the caste basis of political action is having a profound effect on development trajectories (see also Lerche (1995) for a theorisation of the caste-class nexus).  In most contemporary Indian bonded-labour studies, surplus value exploitation and ‘non-economic’ forms of exploitation are mutually reinforcing elements, and their analytical links must be explored.  If a factor is non-economic but affects economic outcomes, then it is of course relevant in any analysis of the political economy.  Three schools of thought may be mentioned although these three are not necessarily mutually exclusive theoretical positions on all fronts.  The three that we will distinguish are the semi-feudalism thesis; the studies of bondage that see it occurring within capitalism; and the recent neo-institutional analyses. 

In the literature on semi-feudalism in Indian agriculture, Bharadwaj (1985) brilliantly summarised the nature of forced commercialisation where people are circumscribed by the interlocking of credit, output and labour markets.  She argued that underdevelopment of one market can mute the development of other markets.  A dominant party conjointly exploits the weaker party in two or more markets, and the weaker party in the exchange loses the option to exercise choice in other markets due to their commitment in one.  The label semi-feudalism was applied by Bhaduri (1973, 1983) to this sort of situation.  Basu (1983, 1986) offered a mathematized version of a similar argument.  Bhaduri was pessimistic about the prospects for capitalist growth, and he developed mathematical models of class interaction to show that it is possible for semi-feudal landlords to inhibit technological change if they find it in their own interest to focus on extracting surplus rather than allowing the overall productivity of the land they control to increase.  Bhaduri’s 1977 paper explores the role of usury in such a scenario.  A number of case studies that take this strand of theory as a starting point have shown capitalist landlords to be taking diverse strategies in different regions of India (Olsen, 1996; Harriss-White, 1996; and Harriss, 1982).  Some academic observers would not accept a simple characterisation of Indian agriculture as semi-feudal, semi-capitalist although all the rural studies do show landlords and landlord-moneylenders using social power as well as economic position to extract surplus value.  The semi-feudalism thesis per se sees the function of landlords’ power as holding back agriculture (Bhaduri, 1983), but the empirical studies mentioned above show rapid technical change interacting with changing social and economic power relations over time.  The social relations do not necessarily hold back technical change. 

Nevertheless there has been a consensus about the semi-feudalism thesis among a large number of marxist writers in India (Bhaduri, 1973, 1977, 1983; Patnaik, 1983; Bharadwaj, 1985; Rao, R.S., 1995; Balagopal, 1988; Ramachandran, 1990).  These writers have attempted to stress forms of unfreedom that exist in agrarian relations to distinguish the ‘free’ wage labour in the classic double sense in capitalism i.e. free of means of production (nonpeasants), and free to sell their labour power.  The use of the term semi-feudalism aimed to stress the incomplete growth of capitalist relations between capitalist landlords and labourers and the inability of capitalist relations to eliminate social modes of oppression. 

Partly as a result of his own findings, and partly in order to improve current strategies for social change, Brass has argued that it is a mistake even to refer to semi-feudalism (or especially feudalism) in the context of analysing contemporary unfree labour.  Brass’s concern is that the semi-feudalism thesis focuses too much attention on individual landlords or their class, instead of focusing on the developing capitalist system as a whole.  Brass’s 1990 case study, based in Haryana, showed that unfree labour exists even in ‘developed’ regions.  This point is also made by Lerche and Kapadia.  Brass called this phenomenon ‘deproletarianisation’, and argued that such labour relations can exist within the capitalist mode of production.  Bondage may even increase, or its forms proliferate, in the context of the constant aggression of capitalists on labour.  Brass argues that free wage labour in the classic double sense has never existed anywhere in history (1997a).  Capital always invents and employs various forms of unfreedom to control and manage labour.  Even in the most industrialised countries, labour is not free in an absolute sense.  Therefore, he argues that the forms of unfreedom among attached farm servants in India are misconstrued if described as feudal remnants.  Instead deproletarianisation is a strategy of capital.  

However, in Indian agriculture the prevalence of attached labour (in the sense of farm servants) has declined considerably, as confirmed by Jodhka (1994), and especially so in the more irrigated and commercialised regions.  Venkateswarlu (1997) makes a similar finding for coastal Andhra Pradesh.  Attached farm labourer relationships are still found in dry and backward regions.  One report on bonded labour in Medak District, another dry area of Andhra Pradesh, noted the widespread prevalence of attached labour (Subrahmaniam et al, 1994), putting the figure at 500,000 with bonded labour among them as 13,000.  From the interplay of the semi-feudalism thesis and the deproletarianisation thesis we draw two conclusions.  Firstly the forms of labour relations are diverse across India, and one therefore expects the dynamic trajectories arising from them to be varied.  Secondly, the labels used for classes and power groups in analytical theories (e.g. semi-feudal landlords; capitalist class; workers) are potent in themselves when they become integrated into political action.  Instead of making a conscious choice between competing labels for social classes, it may be important to recognise diverse forms of exploitation—– between castes; between genders; patriarchal exploitation of the young; and class exploitation.  Each of these forms in turn interacts with the others.  A social relations approach rejects the dualism offered by the Brass-vs.-Bhaduri debate in favour of integrating feminist insights with marxist theories.  A similar integration is achieved by Kapadia (1992, 1995) in her gem-cutting and agricultural-labouring case studies.  We also see a variety of social structures stressed by Kalpagam (1994: chs. 1-2), who is analysing women’s informal-sector work.  The social relations approach is also consistent with the analysis offered by Kabeer (1994).  Brass himself explains that kin relations and gender are integrated into many unfree-labour analyses (1997b: 38).  However Brass ignores the problem that the conceptions of ‘worker’ and ‘labour’ used by most contributors to Free and Unfree Labour: The Debate Continues are anathema to feminists who wish to transcend the dualism of productive-vs.-reproductive work (e.g. Himmelweit (1995) who criticizes the assumption of a work-nonwork divide).⁴  For transformative politics, it is important to identify explicitly oppressive modes of social behaviour, both within and between households.  

A third school of thought lies out side of the political-economy tradition.  Neoclassical economists such as Srinivasan (1989) have examined the labour-credit linkage.  These economists analyse contractual debt bondage as risk-averting behaviour of the bonded labourer, who is assumed to be a rational maximiser of his/her utility in light of the conditions they live in.  By accepting a year long contract, one reduces the transactions costs of finding alternative employment avenues and one avoids unemployment.  Unemployment would be worse than the contract which guarantees some work and food for a person and their children.  The oppression is taken to be an acceptable cost of reducing risk.  Such an approach is shallow in its understanding of the nature of bondage and poverty.  Authors in this tradition claim to be more objective than the marxists.  However Olsen (1998) has stressed the subjective positioning of such authors and the specific norms that are embedded in their claim that bondage contracts must be efficient.  The claim is also functionalist since research on transactions costs and risk aversion could not falsify the claim that all contracts are efficient.  Contracts are seen merely to function to keep the capitalist economy working efficiently.  Neoclassical economics has been evolving from demand-supply models, through transactions-costs analysis, to information economics, and now centres on the ‘new information economics’ which we have called neo-institutionalism.  Hoff and Stiglitz (1990) give a presentation of the theory, while Hodgson (1993) distinguishes the old and new institutionalisms.  However the apolitical economists still underplay the role of local politics and social power in the development of labour relations.  Caste, gender and household dynamics play no role in these models.  In common with methodological individualism generally, the new institutional economics has an implicit gender bias and a claim to value-freeness, whereas the political economists are prepared to make their normative positions explicit and to open them up to debate.  These epistemological differences help to explain why both of the political economy schools are so open to the introduction of feminist ideas, whereas the neo-institutionalists are not. 
 

Exploring the underlying social structure.  Much of the marxist debate on unfree labour has been conducted on narrow class lines ignoring other institutions of oppression such as patriarchy and caste.⁵  In order to consider such factors explicitly along with class exploitation, a useful starting point is the one suggested by Hodgson (1982: 208-216).  He classifies exploitation into: i) exploitation through bargaining power; ii) exercising authority to force someone to do something—especially 'work'— but taking work to include unpaid and ‘unproductive’ services; iii) corporeal exploitation through violence, restriction of movement, and sexual advances; and iv) value exploitation through the appropriation of surplus value in the labour process.  Such a division might widen the scope of analysis from the narrow economic/monetised/public realm to a wider landscape involving more subtle and private forms of oppression.  All the forms may be found co-existing.  They are interlinked and are often mutually reinforcing.  Workers’ subjective understanding of their ‘proper’ role(s) can help to entrench exploitative relationships, quite apart from the deliberate exercise of power through explicit intentional action.  The converse is that changes at a subjective level may reverberate upon the real social relations, such as class relations. 

One can also try to understand the partial embodiment of hierarchical social relationships in the body itself.  The musculature, body size, eating habits, ability to withstand outdoor work, and various manual skills that workers acquire, are all examples of bodily traits that both reflect and perpetuate the current distribution of work.  Fay (1987) argued in support of this thesis.  He warned that critical social scientists should understand and study embodiment, as it places limits on the immediate prospects for change in labour relations.  Ultimately the purpose of his and our analysis is to guide activists in planning effective strategies for change. 

Some forms of oppression are divided into various modes in Table 1.  As Hodgson proposed, we have four modes of exploitation: bargaining; corporeal; authority; and value.  In addition we can separate interpersonal forms from general structural locations that create the conditions for apparently anonymous exploitation (first row versus second row).  For instance, any bond or contract with outstanding loan obligations reduces the bargaining power of the worker.  The employers may restrict workers’ physical mobility, thereby denying access to information about alternatives to bondage.  This is particularly important in places where the labour market is underdeveloped, with high transactions costs and few employers.  Authority structures arising out of the institutions of caste and patriarchy (and thus with roots far back in history) limit people’s verbal negotiating freedom, especially for women.  The ideologies of caste and gender roles place a personal price on breaches of behavioural norms.  Similarly the caste structure claims dominance and high status for the high castes and thus underpins the unequal power relations with labourers.  

Underlying the edifice described above is the class structure.  At the same time, patriarchy is a separate dimension that imposes an unequal gendered division of labour within households as well as in public arenas.  We define patriarchy as the subordination of women as a gender to men as a gender.  It is not independent of class, but rather interacts with it; in all we have here several social divisions that all reinforce unequal access to property (Agarwal, 1998) and the denial of some people’s control over resources, including their own bodies. 
 

II. Case Study of Palamuuru Contract Labour

Mobilising forced labour for public works such highways, railway tracks, irrigation projects, etc., had been a common colonial practice (Kerr, 1998).  Evidence suggests similar recruitments of labour originated in Mahabubnagar district.  For instance, since 1934 labour groups were used to build the Nizamsagar dam by the Nizam government.⁸  

The introduction of large-scale irrigation in the post-independence period turned this region into a source of cheap labour with a remarkable network of middlemen.  Such a relentless form exploitation in the region, has received surprisingly little attention in the literature on labour.  Walker and Ryan (1990), writing about several well-funded long-term village studies of which one was based in a palamuur labour village, say very little about it.  Nor do any official statistics exist to capture the extent of it, as the Indian Census and National Sample Survey have ignored seasonal migration.  However, several researchers have noted a significant level of presence of seasonally bonded labour in at least eight out of twenty-four mandils of the District (Ravinder, 1989; Ramana Murthy, 1991; Reddy, 1990; Rao, Usha, 1994).  The migrants from Mahabubnagar district can be classified into three groups: firstly, those who migrate for seasonal agriculture work in cotton, paddy and sugarcane growing areas; secondly, those who migrate to Hyderabad and other cities to work in the construction industry, usually also seasonally; and thirdly, those who migrate on contract work.  See for details Ramana Murthy (1991) and Usha Rao (1994).  In Mahabubnagar District alone local experts estimate that at least 150,000 people go away to migration works every year.  Of these, about 50,000 (one third) are thought to be bonded labourers. 
 

How the Palamuuru contract system works.  There are two ranks of middlemen: group maistries (labour contractors) and maistries (supervisors).  (We use the gender-specific term middlemen, although it is possible to have women taking on this role.⁹  All the intermediaries in this particular study area in 1991 and 1994 were male.)  Construction companies place an order for the requisite labour force and advance a mutually agreed amount of money.  Each group maistry patronises a team of maistries, each of whom actually mobilises a group of 20 to 30 labourers.  The company forwards some money to the group maistry, who in turn passes some on to the maistry, after retaining his commission.  The usual terms of the contract in both 1991 and 1994 were 8 to 9 months’ work, 12 hours work each day, a holiday for a fortnight, a pack of cheap cigarettes (beedies) for a man every week, and hair oil and food for the labourer and his/her dependent children.  The advance was Rs. 2000 per worker in 1991 for an eight month contract for a man or a woman.  Nominally, this was thought of as a monthly wage of Rs. 200 and provision of free food at the work site.  Two days’ wages are deducted for a day's absence (one for the day's wage and the other for the free food).  If a worker cannot work continuously, and falls sick, any medical expenses are deducted from the wage and counted as debt.  Workers renew their contracts yearly, sometimes because they cannot redeem their loans and sometimes in order to obtain further loans.  For labourers coming from landless and small peasant households struggling to subsist, the maistries are practically monopoly creditors and monopsony buyers of their labour power in the absence of alternative sources of credit and employment.  Such contract labourers hail from the lower castes, enabling the middlemen to command enormous control in enforcing the contract, restricting their mobility at labour camps, and so on.  The maistries have to guard the labourers to keep them from going away as their money is locked up in the investment.  Accounts are settled with the higher-up contractors only at the end of the period.  The respondents in the field study reported that many of the women suffer sexual exploitation, and that children are used for numerous unpaid services. 

The construction companies in this way save a phenomenal amount of money, by paying much less than market wages on the one hand, and extracting four to six hours overtime without pay on the other.¹⁰  For example, simple calculations reveal that each labourer otherwise entitled to more than Rs. 10,000 for eight months work is given only Rs. 2000, the remaining being shared by the company and the middlemen.¹¹  A maistry takes a gross commission of Rs. 30,000 and a group maistry Rs. 100,000 for one contract.¹²  As these middlemen invest money initially and settle accounts with the company only at the end of the year, they are highly exploitative.  The company has no formal or written relation with the labourers.  It is the maistry's responsibility to supervise, to extract work, and to settle the accounts as well as to settle disputes.  Almost all the labourers are illiterate and rely totally on the maistry in settling accounts (Ramana Murthy, 1991). 

The Palamuuru contract system violates the legal rights of workers.  It contravenes the Bonded Labour Abolition Act 1976, the Interstate Migrant Workmen Act (Regulation of Employment and Conditions of Service) 1979 and the Minimum Wages Act 1974 of the Government of India.¹³  Consider for instance the case of Chenayya, 28 years old, of Palem village: 

Field observations.  We present the observations made during the field investigations (as part of a study on rural migration in the region) in August 1991 in five villages in three revenue divisions (mandals) and during another visit in October 1994 in these villages and one more village.  The information presented here pertains to 49 contract labour households, which had in total 162 migrants.  Some of these households were re-visited in October 1994.  Information was elicited from the respondents using both questionnaires and informal inquiries.  Four out of the six villages have exclusively contract labour migration, and two villages have casual labour migration as well.  Inquiries were made about the socioeconomic background of the migrant labourers, the terms of their contracts, and why they went.  We present a summary of these and some profiles of labourers before moving to the analysis section. 

The villages were essentially poorly irrigated, semi-arid areas with red spongy soils having a low water retention capacity.  Village tanks are the principal source of irrigation, supported by wells, irrigating less than one third of cultivated land (see Table 3).  The main crops grown are paddy, jowar (sorghum), bajra (millet), groundnut and castor bean, as well as other pulses.  Low irrigation in the villages limits commercial agriculture to those households with good soil and/or water supply.  

Six out of the 162 contract labourers were aged above 45 years, and these reported 25 years of contract labour.  These people had worked on the Bhakranangal dam (circa 1951), Hirakud (circa 1959), Nagarjunasagar dam (circa 1961), and various steel plants, railway lines, and road constructions since then (Ramana Murthy, 1991).  Fifty-nine labourers between 35-45 years of age had been working since they were teenagers.  Fifty-eight people aged between 25-35 and 37 people aged between 10-15 years were working in the ongoing Narmada Dam (Madhya Pradesh)-Sardar Sarovar (Gujarat) project funded by the World Bank.  The health of the migrant workers deteriorates during the contract work, and many became unfit for this work by the age of 45 years. 

Contract labourers primarily hail from landless agricultural labour households (22 percent) and marginal and small farm households (67 percent), as shown in Table 4.  We have classified households owning less than 0.5 hectare of land as marginal; 0.5 to 2 hectares as small farms; and 2-4 hectares as semi-middle farms.  With the recent distribution of waste land to landless households, the number of marginal households has risen.  The caste-wise background of households having contract labourers showed 40 percent belonging to the scheduled castes, 15 percent to the scheduled tribes, and 45 percent to the backward castes.  One household belonged to a so-called ‘forward’ caste. 

The gross farm incomes from the subsistence agriculture of these households were estimated in per capita terms to be Rs. 332; Rs. 1866; and Rs. 3970 in 1991 for the marginal, small and semi-middle farms respectively (Table 5).  When the imputed costs of family labour were deducted, some households were shown to have negative net income.¹⁵  Needless to say, most of these people live below the official poverty line. 

Instability and change in migration pattern.  There is a considerable instability in migration patterns seen during our visits.  In two villages, people were migrating as casual labour to Hyderabad city, yet in these villages the Palamuuru contract system had been prevalent.  The transition to casual labour migration has been encouraged by several factors.  These villages had better irrigation facilities, compared to the four contract-labour villages, and even compared with the surrounding villages.  The people’s personal knowledge of Hyderabad city since the early 1980s had enabled them to wriggle out of the grip of the contract system and the middlemen.  However, this has only solved the immediate subsistence problem inasmuch as they continue to sojourn between their village and the city.  They did not avail themselves of any provisions in the city, such as ration cards or education of children.  With their incomplete integration into the urban economy and a faint hope of improvement in the condition of the home village, they shuttled back and forth between rural poverty and urban poverty. 
 

III. Analysis

As expressed by the migrants, the labour relations gain acceptability for the reasons akin to those for accepting attached labour.  

Oppression through reducing bargaining power:  

i) Dyadic: the maistries have a grip if there is an outstanding loan.  The debts incurred by migrant family members restrict their bargaining abilities in seeking either a contract or escape.  Fixing workers into contract labour overtime restricts their access to information on labour markets elsewhere, further weakening their bargaining abilities.  

ii) Structural: Many workers are so bankrupt and destitute that they have few options other than working for the maistry.  However, those with some irrigated land and information on labour markets are in a better position, and more of the casual labour migrants come from such households (Ramana Murthy, 1991: 90-95).  Workers who have relatives in cities are also in a better position to bargain with potential employers, since they can if necessary migrate to cities and look for work.   
  

Oppression through the body (corporeal):  

i) Dyadic: within a contract, the employer can control food intake, water supply, housing conditions, and to some extent the health care of the worker.  Palamuuru labourers, debilitated by the relentless work, depend on their employer for further loans to maintain and improve their health.  Occasional beatings and verbal abuse of those people who disobey the employer or who try to leave without repaying cause other workers to be self-disciplined labourers.  There may be strong gender differences in the nature of corporeal oppression and these require further research.  

ii) Structural: For many women the patriarchal system denies them wages for domestic labour, and then lets their menfolk control the wages they earn as employees.  Women are expected to accept a lack of personal spending money; domestic violence; and coercion from employers.  Sexual harassment of a woman who does physical outdoor work is seen by some people as being the woman’s fault, since she shows herself in public.  The resulting embarassment or illicit sexual liaisons may be seen as oppressive to women even though the people involved might think it perfectly acceptable.  In addition, female workers have to care for and feed children as well as themselves while working (in fields or at construction sites).  Men don’t have this disadvantage, merely because they are men in a patriarchal society.  Most women’s mobility is more restricted than men’s, not only because of the presence of children but also by the understanding that they are responsible for the home/hut/tent and therefore do not have a right to leave it without permission.  These implicit constraints on women's mobility limit their ability to secure information and improve their bargaining power.  Norms about the domestic roles of women cannot be neglected in considering their lives as contract migrant workers.  
  

Oppression through authority:  

i) Dyadic: During the Palamuuru contract, as described above, the maistry has control over the movements and activities of the workers.  Their control derives not only from the debt or contract, but also from class/caste identities and traditions of respect and servility that may operate between a maistry and a worker.  In addition, a female worker may face the authority of her husband, brother, father, or mother-in-law who may force her to work and possibly also deprive her of control over her earnings.  Children themselves are subject to authority of the elders, and in contract labour families, kids became contract labourers in their own right from the age of ten onwards (Ramana Murthy, 1991).  

ii) Structural: Habits and traditions of inter-caste behaviour may not require any explicit enforcement to remain extremely powerful.  Self-regulated submissive behaviour is embedded in the caste system, where workers of lower castes do not eat food in front of an upper caste maistry, may hesitate to touch them, keep a distance from them, do not enter their home/hut or their cooking area, and so on.  Employers hailing from the upper castes may have high political, social and economic status.  Furthermore, when going in public the worker may face ridicule and ostracism if they decide to flout the dictates of the maistry.  The combination of class and caste factors militates against the person’s consciousness in fighting the employer.  By contrast, workers who stay within the system become a part of a cohesive group serving one maistry and are subject to pressure to conform to the behaviour of that group.  
  

Value exploitation:  

i) Dyadic: It is possible to measure the gross profits reeled in by each maistry as the difference between their revenue and their wage bill to get a rough indicator of value-exploitation per worker and its variations.  However this account will not cover the additional indirect exploitation undertaken by the construction company by underpayment and the overtime without pay.  Handling the profits at local and higher levels requires more information and a clear conceptualisation of profit and surplus value.  Firstly, input costs on the employer's side should be allowed for.  Secondly, the profits realised by the contractor’s firm may increase over time after the contract itself is finished.  A dyadic approach to estimating the rate of value-exploitation would suffer from the tendency to focus too heavily on direct relations of production (the labour relation) and to ignore indirect exploitation through exchange.   

Nevertheless, even crude estimates, as given earlier, can have a powerful effect on consciousness since many workers may be unaware of the extent of exploitation.  Another area ignored by the calculation of value exploitation between employer and employee is the extent of unpaid labour that is being done by women and children (and perhaps also men, if they go to collect firewood or to get water).  In addition to cooking, watching children, cleaning, and doing other work aimed at reproduction of the labour force, women also provide sexual services (by choice or duress; for the husband or for the maistry).  It was difficult to gather information on this important area without long periods of residence, but we have begun to develop an agenda for local research and activism.  

ii) Structural: Perhaps more important than the pairwise rate of value-exploitation is the fact that these labourers seem to have a buffer between themselves and capitalist exploitation.  They would have to fight off the maistries and group maistries before they could wage a class struggle with the ultimate employers or even find out who they are.  The fight against parasitic middlemen would perhaps save them from bondage only to become free wage labourer in the usual capitalist sense (Miles, 1987).  New forms of unfreedom may also emerge, as Brass contends, yet at least the current forms might disappear through the growing consciousness of the people.  Such a radical consciousness is severely hindered by caste divisions and kinship loyalties (as Brass himself has pointed out, Brass (1986)).  

The recognition of modes of exploitation in interlocked markets is useful in two ways – both for innovative research, as mentioned above, and for innovative strategies for action.  The legal abolition of bonded labour has not succeeded in freeing bonded workers, though it may have had some effects on their consciousness.  Twice in separate instances described by respondents, Palamuuru contract labour were declared as bonded labour and released—once by the Lathur District collector at the request of a local voluntary agency, and once more recently in Ranga Reddy District.  There is no evidence of their rehabilitation.  These people will, in all probability, have now re-entered contract work.  One can argue for the formation of labour markets with generalised rights, yet continue the proletarian struggle for the reduction of value-exploitation and the elimination of oppression.  One author has gone so far as to include among the definitional characteristics of ‘free wage labourers’ the proviso that they must have the genuine right to withdraw their labour through quitting or going on strike (Grossman, 1997).  Grossman’s study, based in South Africa, illustrates how within capitalism the ‘freedom’ of workers is perpetually constrained.  It also shows that ‘freedom’ is a social construct which can be used in many contexts to represent various situations as acceptable.  At least one author has questioned the free/unfree dichotomy itself, arguing that it sets up a spectrum that lends itself to liberal and not to progressive values (Prakash, 1990).  The debate over unfree labour is not a matter of ascertaining facts, but rather one of spreading an action-oriented awareness of exploitative social relationships.  

One research proposal arising out of the rather disturbing evidence above is to explore modes of resistance as well as modes of oppression along the two dimensions shown in Table 1.  In this way the analysis presented in this paper moves beyond Hodgson’s original analysis and moves toward the action orientation of most political economy authors.  However the model-making marxists, notably Bhaduri and Bardhan, have ignored the need to analyse the change process and have instead appeared to see the system as perpetually reproducing itself.  Local non-governmental organisations (NGOs) are working on watershed development and natural resource management, and since the early 1990s they have begun to involve members in micro-credit groups.  However most of the NGO networks have weak links at the grass roots at present.  They could focus not only on income-generation and substitute credit sources, but also on the cultural, patriarchal, class, and caste bases of oppression in this region.  The case study suggests that strategies of the NGOs could perhaps be expanded to include a wide range of anti-oppressive activities.  Access to seaonal migrants would have to be focused during the farming season, or else visits to the works sites would be needed.  
   

IV. Conclusion

Interlocking of markets has received attention elsewhere in the literature.  It has been argued that landlords’ interlinkages can reduce the accumulation of capital on the part of producers, leading to retarded growth (Bhaduri, 1973, 1983).  The intermediaries in the Palamuuru labour case display a reactionary character by fixing a contract wage less than subsistence level, trapping the labourer in debt bondage.  However we cannot say that this form of bondage is restricting the development of capitalism, when the labourers are creating infrastructure which can be seen as future means of production (e.g. roads; dams both for power and for irrigation).  The rate of exploitation, direct and indirect, is high since intermediaries pay a wage much lower than urban market wages and extract considerable unpaid overtime and child labour.  Yet the exploitation may draw complicity from the labourer whose terms of contract resemble those of co-existing and pre-existing attached labour relationships in the region.  We would argue that in addition to being exploitative and parasitic on the hard work of labourers, the intermediaries are also using traditional caste-based and patriarchal modes of oppression to maintain their exploitative labour relations.   

Underneath the actions of the agents involved, we also have to look at the roles and identities that are internalised by workers.  Maistries and group maistries are seen as those who save the labourer in distress by offering work when otherwise they might starve.  Some lower-caste people who serve the landlords and employers think they will be rewarded with patronage during crises.  This patronage may at present consist only of loans which further bind the worker and the worker’s family.  It would be misleading to judge from some workers’ attitudes such as loyalty, dependence, devotion and submission that their bondage is completely self-chosen and voluntary.  On the contrary, studying local cases of resistance, resentment, and revolt would throw more light on the underlying compulsions as shown by Kapadia (1995) and by Lerche (1995).  This is an area for future research.  

Organising labour in the light of the informal and interlocked nature of labour markets poses an extremely difficult task.  Migrant labourers are scattered and often travelling.  The eradication of bonded labour will need a multi-pronged strategy involving propaganda, legislation, mobilisation, rehabilitation, and some provision of basic amenities.  Any fragmented effort may fail. Isolated attempts to rehabilitate people have been failures.  Low wages, both for men and for women, themselves have to come under scrutiny, since the lack of independent subsistence is one reason why people accept bondage.  The state has tried three means to help such workers: the minimum wage laws, which are hardly ever implemented where local market wages lie below them; the employment guarantee programmes, which have not been enough to meet the needs of the masses; and the law abolishing bonded labour (dated 1976) which is still not effective.  A new strategy is needed.  

The NGOs in the district have started women's education programmes which try to empower people to secure access to specific government relief programmes.  In particular, women are taught to approach authorities directly for seeking access to local employment schemes.  Our analysis has shown that the oppression that is linked to value-exploitation extends into family life, village rituals, traditions and norms of caste hierarchy, and thus is to some extent beyond the reach of the modern state.  The elimination of the parasitic middlemen’s role is part of the answer yet more fundamental changes are also needed.  In the process of struggle, any measures taken must recognise the past structural locations, the beliefs and values, and the embodied capabilities of workers, as well as the dyadic long-term relationships upon which some workers depend.  Nevertheless, there are prospects for change: 

 

Notes

1.  We are grateful for comments received at the Conference of the Indian Political Economy Association, 1995, and at the Conference of the Society for the Advancement of Socio-Economics, 1994, on material presented here.  We also wish to thank D. Narasimha Reddy for his support regarding this work.  Finally, we are grateful to the advice of the anonymous reviewers regarding revision of the paper.   back to text  

2.  "Manu, writing later than Koutilya, outlined only seven categories of dasas.  They are: persons captured in battle; those enslaved in return for food; dasas born in the house of the master; those who are bought; dasas inherited as part of patrimony; dasas who are given away by their parents; and persons enslaved for not paying a fine." See Chakravarthy (1985), 40.   back to text  

3.  An account of various colonial laws is given by Dingwaney (1985) in Patnaik and Dingwaney, eds. (1985), 340-50.   back to text  

4.  One example illustrating the gender bias implicit in ungendered analyses of ‘workers’ is van der Linden’s analysis of employers’ and employees’ strategic choices regarding employment relations.  Van der Linden glosses over gender differences and possible additional sources of control or oppression facing women, relative to men.  Van der Linden presents an idealised model of employers’ motives and employees’ motives.  His use of s/he as the pronoun for employers here does not resolve the problem of having a fundamentally ungendered analysis at this point (1998: 516-519). 
back to text  

5.  Studies such as Kapadia (1995) and Lerche (1995) diverge from the pattern set by Bhaduri (1983) and Miles (1987).   back to text  

6.  Mahabubnagar was earlier known as Palamuuru.  It was renamed after the erstwhile ruler, the 6th Nizam, Mir Mahabubali Khan in 1929, but people still call it informally by its old name, and the migrant labourers are referred to as Palamuuru labour.  The district was under autonomous kingdoms (Samsthanams), and administratively and politically the region remained isolated and insulated from the famous Telangana Peasant armed revolt of 1947.   back to text  

7.  The Report on Indebtedness in Nizam Domain (1934) mentioned that the district was under severe drought, near famine conditions, with loss of cattle and subsequent indebtedness among farmers.  
back to text  

8. "The construction of Nizamsagar dam involved a massive need for labour both skilled and unskilled.  Especially excavation of earth, blasting, blasting rock in foundation, quarrying stone for machinery and miscellaneous.  The labour for these works had been imported from dominions including Raichur, Gulbarga, Mahabubnagar, Nalgonda and Karimnagar...Palmuuries coming from Mahabubnagar did a good work at excavation, especially earth work at ground level...The method of recruiting the labourers is mainly through the department by forwarding advances ranging from Rs. 5 to Rs. 10 per year for the gangs willing to come." The Report on the History of Nizamsagar Project, Hyderabad, Deccan, Public works Dept., H.E.H.Nizam Dominion, 1939 (A.P.Archives, Hyderabad).   back to text  

9.  The use of the male gender here is important empirically because (a) one cannot study female supervisors, so can only comment on male supervisors; and (b) the gendering of supervision may give hints to future researchers on why the system perpetuates itself so repeatedly.  Life histories, family case studies, or critical incident analysis may be needed to examine why men turn to this particular activity and how these men interact with the large contractors in ways that ultimately exploit both men and women.  The present study has more limited aims and did not use in-depth qualitative methods. back to text  

10.  At the rate of 4 hours per day of extra-labour, there are 112 days of unpaid work!  Taking 336 workdays per year multiplied by Rs. 35 daily wage (Hyderabad construction industry wage at that time) the sum is Rs. 11,820.  Assuming an absence of 30 days in that period, the labourer is still entitled to Rs. 10,710.   back to text  

11.  The public works department while calling for tenders on construction work gives a rough cost of production, which includes provision of market wages.  The lowest bidder is given the contract and there is nobody to check on what is actually paid.  The construction companies find it advantageous to rely on the middlemen for various reasons such as easy mobilisation and few direct management problems—it’s the maestry’s headache and the company can save a phenomenal amount of money by underpayment.  The labour supply is not only assured for the present but also for the future.  
back to text  

12.  By contrast the statutory minimum wages for agricultural labour tasks over this period were all less than Rs. 20 per day.   back to text  

13.  According to the Bonded Labour Abolition Act, a person is presumed to have entered into the contract in consideration of i) advance of money by the person/ascendants/descendants from the employer, or ii) in consideration of being born a certain caste, or...for a specified/unspecified period of time, based on oral or written contract.  Bonded Labour Abolition Act 1976, Labour Year Book, Labour Bureau. Shimla. 1-5.  See also Brass (1997b: 42) for a review of what he calls India’s ‘law-as-rhetoric’.   back to text  

14.  From field data collected by Ramana Murthy in 1991 and 1994.    back to text  

15.  Given the subsistence nature of the local agriculture, these gross farm incomes had to be estimated from information on the yield and acreage of respondents’ fields, after deducting the imputed cost of reported purchased inputs.  Net incomes were also calculated by deducting imputed values of family labour, using an estimated 1991 local wage rate of Rs. 10 per day for all.  Actual wages vary according to the sex and age of worker, and by task, but this minimal figure was used as a rough estimate’.    back to text  

16.  In Telugu the word for famine also means drought (karuvu).   back to text  

17.  From field data collected by Ramana Murthy in 1991 and 1994.    back to text  

18.  From field data collected by Ramana Murthy in 1991 and 1994.    back to text  

19.  From field data collected by Ramana Murthy in 1991 and 1994.    back to text  

20.  From field data collected by Ramana Murthy in 1991 and 1994.    back to text  

21.  From field data collected by Ramana Murthy in 1991 and 1994.    back to text  

22.  From field data collected by Ramana Murthy in 1991 and 1994.    back to text  



Sources

Agarwal, Bina. (1998). "Disinherited Peasants, Disadvantaged Workers: A Gender Perspective on Land and Livelihood," in Economic and Political Weekly. 28 March. Vol. 33, No. 13. A2-A28.  
back to text  

Bagchi, A.K. (1975). "Relation of Agriculture to Industry in the Context of South Asia," in Frontier. Autumn. 22-24.  

Balagopal, K. (1988). Probings in the Political Economy of Agrarian Classes and Conflicts. Hyderabad: Perspectives Press.   back to text  

Bandyopadyaya. (1976). "Vettichakiri in M.K.," in National Labour Institute Bulletin. No. 2 (3).  

Bardhan, P. K., ed. (1989). The Economic Theory of Agrarian Institutions. Oxford: Clarendon Press.  

Barucha, S.M. (1937). Agricultural Indebtedness in H.E.H. Nizam Dominion. Hyderabad: Andhra Pradesh Archives.  

Basu, Kaushik. (1983). "The Emergence of Isolation and Interlinkage in Rural Markets," Oxford Economic Papers. No. 35. 262-280.   back to text  

Basu, Kaushik. (1986). "One Kind of Power," Oxford Economic Papers. No. 38. 259-282. 
back to text  

Bhaduri, Amit. (1973). "A Study in Agricultural Backwardness Under Semi-Feudalism," in Economic Journal. March. 120-37.   back to text  

Bhaduri, Amit. (1983). The Economic Structure of Backward Agriculture. Delhi: Macmillan. 
back to text  

Bhaduri, Amit. (1986). "Forced Commerce and Agrarian Growth," in World Development. 14:2. 267-272.  

Bhaduri, A. (1977). "On the Formation of Usurious Interest Rates in Backward Agriculture," in Cambridge Journal of Economics. No. 1. 341-352.   back to text  

Bhalla, A.S., et al, eds. (1993). Images of Rural India in Twentieth Century. Delhi: Oxford University Press.  

Bharadwaj, Krishna. (1985). "A View on Commercialisation in Indian Agriculture and Development of Capitalism," in Journal of Peasant Studies. Vol. 12, No. 4. 494-519.   back to text  

Bharadwaj, Krishna, and P.K. Das. (1975). "Tenurial Conditions and Modes of Exploitation," in Economic and Political Weekly. June. Vol. 10. A49-55.  

Brass, Tom. (1986). "The Elementary Strictures of Kinship: Unfree Relations and the Production of Commodities," in Social Analysis. No. 20.   back to text  

Brass, Tom. (1990). "Class Struggle and Deproletarianisation of Agricultural Labour in Haryana (India)," in Journal of Peasant Studies. Vol. 18, No. 1. 36-67.   back to text  

Brass, Tom. (1991). "Market Essentialism and the Impermissibility of Unfree Labour: A Reply to Shlomowitz," in Slavery and Abolition. Vol. 12, No. 3. 225-244.  

Brass, Tom. (1995). "Unfree Labour and Agrarian Change: A Different View," in Economic and Political Weekly. 1 April. 697-699.   back to text  

Brass, Tom. (1997a). "Some Observations on Unfree Labour, Capitalist Restructuring and Deproletarianisation," in Brass and van der Linden, eds. (1998). 57-76.   back to text  

Brass, Tom. (1997b). "Introduction," in Brass and van der Linden, eds. (1998). 11-44.   back to text  

Brass, Tom, and Henry Bernstein. (1992). "Introduction: Proletarianisation and Deproletarianisation on the Colonial Plantations," in Journal of Peasant Studies. 19. 1-40.   back to text  

Brass, Tom, and Marcel van der Linden, eds. (1998). Free and Unfree Labour: The Debate Continues. Berlin and Paris: Peter Lang AG.   back to text  

Breman, Jan. (1984). Of Peasants, Paupers and Migrants. New Delhi: Oxford University Press.  

Chakravarty, U. (1985). "Of Dasas and Karmakaras: Servile Labour in Ancient India," in Patnaik and Dingwaney, eds. (1985). 35-75   back to text  

Dingwaney, Manjari. (1985). "Unredeemed Promises: The Law and Servitude," in Patnaik and Dingwaney, eds. (1985). 283-347.  

Fay, Brian. (1987). Critical Social Science: Liberation and its Limits. Oxford: Polity.   back to text  

Government of Andhra Pradesh, Bureau of Economics and Statistics (various years). Handbook of Statistics, Mahabuubnagar District, Hyderabad.   back to text  

Grossman, Jonathan. (1998). "The Right to Strike and Worker Freedom in and beyond Apartheid," in Brass and van der Linden, eds. (1998).   back to text  

Harriss, J. (1982). Capitalism and Peasant Farming: Agrarian Structure and Ideology in Northern Tamil Nadu. Bombay.   back to text  

Harriss-White, B. (1996). A Political Economy of Agricultural Markets in South India: Masters of the Countryside. New Delhi: Sage.   back to text  

H.E.H. Nizam Dominion. (1939). The Report on the History of Nizamsagar Project. Hyderabad, Deccan, Public Works Department. (Andhra Pradesh Archives, Hyderabad).   back to text  

Himmelweit, S. (1995). "The Discovery of ‘Unpaid Work’: The Social Consequences of the Expansion of ‘Work’," in Feminist Economics. Vol. 1, No. 2. 1-20.   back to text  

Hjejle, Benedicte. (1967). "Slavery and Agricultural Bondage in South India in the Nineteenth Century," in Scandinavian Economic History Review. Vol. XI, Nos. 1-2.   back to text  

Hodgson, Geoffrey. (1982). Capitalism, Value and Exploitation. London: Martin Robertson. 
back to text  

Hodgson, Geoffrey M. (1989). "Institutional Economic Theory: the Old Versus the New," in Review of Political Economy. Vol. 1, No. 3. 249-269. 

Hodgson, Geoffrey M. (1993). "Institutional Economics: Surveying the 'Old' and the 'New'," in Economics and Evolution: Bringing Life Back into Economics. Cambridge, UK: Polity Press. Originally published in Metroeconomica. Vol. 44, No. 1, 1993. 1-28.   back to text  

Hoff, K. and J. Stiglitz. (1990). "Imperfect Information and Rural Credit Markets: Puzzles and Policy Perspectives," in World Bank Economic Review. Vol. 4, No. 3. 235-250.   back to text  

Jodhka, Surinder Singh. (1994). "Agrarian Change and Attached Labour: Emerging Patterns in Haryana Agriculture," in Economic and Political Weekly. 24 September. Vol. 29, No. 39.  
back to text  

Kabeer, N. (1994). Reversed Realities: Gender Hierarchies in Development Thought. London: Verso, and Delhi: Kali for Women.   back to text  

Kalpagam, U. (1994). Labour & Gender: Survival in Urban India. Delhi: Sage.   back to text  

Kapadia, Karen. (1992). "Every Blade of Green: Landless Women Labourers, Production and Reproduction in South India," in Indian Journal of Labour Economics. 35:3. 266-276.   back to text  

Kapadia, Karen. (1995). "The Profitability of Bonded Labour: The Gem-Cutting Industry in Rural South India," in Journal of Peasant Studies. 22:3, April. 446-483.   back to text  

Kerr, Ian J. (1998). "Free or Unfree? Railway Construction Labour in Nineteenth Century India," in Brass and van der Linden, eds. (1998).   back to text  

Kosambi.D.D. (1956). "On the Development of Feudalism in India," in Journal of Bombay Branch of Royal Asiatic Society. 31/32. 258-69.   back to text  

Krishnaiah, M. (1998). "Regular Farm Labour System in Semi-Arid Areas: Emerging Patterns in Andhra Pradesh," in Economic and Political Weekly. 28 March. A20-A28.  

Kulke, Hermann, ed. (1997). The State in India 1000-1700. Delhi: Oxford University Press.  

Kumar, Dharma. (1967). Labour and Caste in South India: Agriculture in Madras Presidency. London: Cambridge University Press.  

Labour Bureau. "Bonded Labour Abolition Act 1976," in Labour Year Book. Shimla. 1-5.  

Lerche, Jens. (1998). "Agricultural Labourers, the State and Agrarian Transition in Uttar Pradesh," in Economic and Political Weekly. 28 March. A29-A35.   back to text  

Lerche, Jens. (1995). "Is Bonded Labour a Bound Category? Reconceptualising Agrarian Conflict in India," in Journal of Peasant Studies. 22:3, April. 484-515.   back to text  

Miles, Robert. (1987). Capitalism and Unfree Labour: Anomaly or Necessity?. London: Tavistock.  back to text  

Mohan Rao, J. (1999). "Agrarian Power and Unfree Labour," in Journal of Peasant Studies. 26: 2/3. 242-262.  

Olsen, Wendy K. (1996). "Marxist and Neo-Classical Approaches to Unfree Labour in India," in Brass and van der Linden, eds. (1998). 379-404.   back to text  

Olsen, Wendy K. (1998). "Paradigms in the Analysis of Credit Structures and Discourses in South Indian Banking," Working Paper No. 5, September. University of Bradford, Graduate School of Social Sciences and Humanities.   back to text  

Patnaik, Utsa. (1983). Peasant Differetiation. New Delhi: Oxford University Press.   back to text  

Patnaik, Utsa, and Manjari Dingwaney, eds. (1985). Chains of Servitude: Bondage and Slavery In India. New Delhi: Sangam Publishing.   back to text  

Pradan, H.Pradhan. (1979). "Semi-feudalism: The Basic Constraints of Indian Agriculture," in Arvind N. Das and V. Nilkanth, eds. Agrarian Relations in India. New Delhi.  

Prakash, Gyan. (1990). Bonded Histories: Genealogies of Labor Servitude in Colonial India. Cambridge: Cambridge University Press.   back to text  

Radhakrishna, R., et al. (1987). "A Study of Poverty Levels: Inter-District Analysis," Working Paper No. 431. Centre for Economic and Social Studies, Hyderabad.   back to text  

Ramachandran, V.K. (1990). Wage Labour and Unfreedom: An Indian Case Study. Oxford: Oxford University Press.   back to text  

Ramana Murthy, R. V. (1991). "Seasonal Labour Migration In Semi-Arid Areas: A Case Study of Palamuuru Labour," M.Phil. thesis, Department of Economics, University of Hyderabad. 
back to text  

Rammohan, C. (1987). Palamuuru Labour (Telugu). Hyderabad: VIRASAM (Revolutionary Writers Association).   

Rao, R.S. (1995). Towards Understanding Semi-Feudal Semi-Colonial Society. Hyderabad: Perspectives Press. Series title: D. Narasimha Reddy, ed. Studies in Political Economy. Department of Economics, University of Hyderabad.   back to text  

Rao, Usha. (1994). Palamoor Labour: A Study of Migrant Labour in Mahabubnagar District. C.D. Deshmukh Impact Centre, Council For Social Development, Hyderabad.   back to text  

Ravinder, A. (1989). "Labour Migration: A Dimension of Poverty," unpublished M.Phil. thesis, Department of Economics, Osmania University, Hyderabad.   back to text  

Reddy, D. Narasimha. (1990). Rural Migrant Labour in Andhra Pradesh. Report submitted to the National Commission on Rural Labour, Government of India.   back to text  

Saradamoni, Kunjuleks. (1973). "Agrestic Servitude in Kerala in the Nineteenth Century Century," in Indian Economic and Social History Review. 371-385.   back to text  

Saradamoni, Kunjuleks. (1980). Emergence of a Slave Caste: The Pulayyas of Kerala. New Delhi: People’s Publishing House.   back to text  

Sarkar, T. (1985). "Bondage In Colonial Context," in Patnaik and Dingwaney, eds. (1985). 97-126.   back to text  

Sharma, R.S. (1956). "The Origins of Feudalism in India (c.AD 400-650)," in JESHO. 1. 297-328.   

Sinha, Shantha. (1993). "Changing Landlord-Labour Relationships," in Bhalla, A.S., et al, eds. (1993).   

Srinivasan, T.N. (1989). "On Choice Among Creditors and Bonded Labour Contracts," in P. K. Bardhan, ed. (1989). 203-220.   back to text  

Srivastava, Ravi. (1989). "Interlinked Modes of Exploitation in Indian Agriculture During Transition: A Case Study," in Journal of Peasant Studies. Vol. 16, No. 4. 493-522.   back to text  

Standing, Guy. (1981). "Migration and Modes of Exploitation: Social Origins of Immobility and Mobility," in Journal of Peasant Studies. Vol. 8, No. 2. 173-211.   

Stein, Burton. (1980). Peasant State and Society in Medieval in South India. Delhi: Oxford University Press.   

Subrahmaniam, Rama Raju, and Thimma Reddy. (1994). Bonded Labour in Andhra Pradesh (A Study in Medak District). Hyderabad: Centre for Economic and Social Studies.   back to text  

Thapar, Romila. (1960). A History of India - Vol. 1. Harmondsworth: Penguin.   

Thompson, E.P. (1978). The Poverty of Theory and Other Essays. London: Merlin.   back to text  

Unni, Jeemol. (1998). "Non-Agricultural Employment and Poverty in Rural India: A Review of Evidence," in Economic and Political Weekly. 28 March. A36-A44.   

Venkateswarlu, D. (1997). Ph.D. Dissertation, Department of Political Science, University of Hyderabad.   back to text  

Walker, T.S., and J.G. Ryan. (1990). Village and Household Economies in India's Semi-arid Tropics. Baltimore and London: Johns Hopkins University Press.   back to text