These are the reflections of a Rip Van Winkle in agrarian studies, someone who was active in agrarian research, mainly in India, in the 1970s through to the early 1990s, but who then, so to speak, went to sleep, taking up other avenues of research until recently. Rip Van Winkle, in Washington Irving’s story, succeeded in sleeping through the American Revolution and was shocked, when he woke up, to find that the portrait of King George III had been replaced by that of George Washington. While I don’t think that agrarian studies has undergone quite such a dramatic shift over the last two decades, there is little doubt but that the field has seen some significant changes. By the early 1990s, the controversy between Chayanovian and Leninist theories of peasant economy and agrarian change (represented in the pages of a book that I edited: Harriss 1982) was no longer pursued as vigorously as it once had been, and by this time the mode of production debate had more or less run its course. Questions about the impact of the green revolution, however, were still being debated. Meliorists and critics continued to lock horns over whether or not “modern agriculture” had brought the increased impoverishment of many in the rural economy or rather had been instrumental in improving living standards. The argument that Indian agriculture continued to be characterised by semi-feudalism was still alive, and the critical question of the relationships between agrarian power structures and agricultural productivity had not quite been settled by the contributions to the book on the subject that had been edited by Meghnad Desai, Susanne Rudolph, and Ashok Rudra (1984; see especially the chapter by Ronald Herring). Politically, agrarian power was certainly very much in evidence in the actions of the so-called “farmers’ movements” that had mobilised in different parts of the country, and it remained so into the 1990s.

What of the present, and of the trends of the last 20 years or so? As Ramakumar says in his magisterial review of “Economic Reforms and Agricultural Policy in India,” presented at the conference:

Agricultural development in post-independence India is marked by a failure of the State to resolve the agrarian question, i.e., ending the extreme concentration of land ownership and use and weakening the factors that fostered disincentives in investment and technology adoption, tied workers to a social system with considerable pre-modern features, and compressed purchasing power. While this failure shaped the pattern and nature of agricultural growth in India after 1947, the implementation of economic “reforms” after 1991 introduced new dimensions to the contradictions of the earlier regime. The Washington Consensus-inspired policies in agriculture have had acute adverse effects on the conditions of life and work in rural India (2014, p. 28).

The Indian state has indeed largely bypassed the agrarian question. The failures of agriculture from about the time of the “reforms” and continuing well into the new century, shown up in Ramakumar’s calculations of rates of growth of agricultural production — the Index of Agricultural Production grew at only 0.5 per cent per annum between 1992-3 and 2003-4, and Gross Value of Output for all crops at less than 2 per cent — did not apparently affect economic growth as a whole in the way that was formerly the case in India’s economic history, when a poor performance in agriculture would be reflected throughout the economy. And though agricultural growth picked up significantly after 2003-4, according to Ramakumar’s calculations, it was still not by so much as to have “raised the growth rates of agricultural production of the combined phase of the 1990s and 2000s over the 1980s” (2014, p. 4). These findings must be set against Abhijit Sen’s positive claims in his presentation that Indian agriculture is now growing at an unprecedented rate, and they cast some doubt on the optimism he expressed about India’s future food security. That optimism was flatly contradicted in her presentation by Sheila Bhalla, who reported on the loss of net sown area and on the evidence of the negative impact of declining land/person ratios on productivity per agricultural worker. She thus was led to see an impending long-term threat to food security, contra Sen’s claims.

Sheila Bhalla also presented data to back her claim that “the hard fact is that the typical household operating a holding of less than 4 ha [which she shows to be the case for 96 per cent of farm households] cannot cover actual expenses out of the combined income from crop cultivation and animal husbandry...” (2014, p. 15). The survival of very many depends upon hiring out their labour and/or on engaging in some kind of non-agricultural activity, typically involving self-employment. The 45 per cent or so of farm households operating 0.4 ha or less rely very largely on wages; only in those households with more land than 0.4 ha is cultivation the largest single source of income. Bakshi, Das, and Swaminathan, in studies of household incomes in 13 villages across six States, found that on average households reported incomes from at least three sources. As Bhalla concluded, the failures of post-1991 economic and social policy “to generate acceptable alternative incomes for farmers whose holdings are too small to provide incomes over actual consumption expenditures has condemned very many people to wretched lives, even if some of them, at least, are better off than they were.” Jobs outside agriculture have raised the income levels of some of the poorer households, but only by dint of people taking on work that is a long way from the idea of “decent work” of the International Labour Organisation. And Dhar, Dixit, and Sivamurugan, in a paper on manual employment in rural India based on studies in 13 villages in six States, reported that these poorer households constitute the largest single class in most villages; they are disproportionately Dalit, and increasingly feminised; underemployment is extensive amongst these workers, especially the women; and the great majority live below the (conventionally defined) poverty line. Only increased wage levels can possibly change this situation.

These broad findings are generally borne out in the village studies that were presented. Himanshu, for example, describing findings from the most recent rounds of research in the Uttar Pradesh village of Palanpur, reported that cultivation accounts for only a minor share of all income accruing to village people, and that though access to non-farm employment has become more broad-based and has brought about upward mobility for significant numbers of poor households, increased incomes from non-farm sources are also associated with increased inequality. Interestingly, too, non-farm jobs are increasingly in self-employment and various forms of casual labour, rather than involving regular or semi-regular employment as was more dominantly the case 20 years ago. Gerry and Janine Rodgers reported similar findings from villages they have studied over more than 30 years in Bihar. There, too, increasing reliance on non-farm activity, accessed through migration or by commuting, accounts for a general rise in incomes, though the gains are very unevenly distributed, and the Gini coefficients of household income have risen. A good many instances of similar findings from elsewhere can be found (see, for example, Harriss and Jeyaranjan, forthcoming, in regard to village studies from Tamil Nadu), while the findings of the Project on Agrarian Relations in India (PARI) of the FAS, from 13 villages across six major States, show how very low household incomes are in most cases — only in one village were mean incomes per person above the US$2 per day standard — and how unequally they are distributed: “In every village there was a small class of landlords, capitalists, and rich peasants whose mean incomes were more than ten times those of manual workers” (Bakshi, Das, and Swaminathan 2014, p. 29). Similarly, Vikas Rawal’s study of “Cost of Cultivation and Farm Business Incomes,” drawing on PARI data for 10 villages in five States, shows that the vast majority of cultivator households obtain very low incomes from cultivation, and that many of them, in the year of the survey, incurred losses. But he reports, too, that there is a top fraction of households — landlords and others — obtaining substantial incomes (Rawal 2014).

The PARI studies presented at the conference and elsewhere (Ramachandran, Rawal, and Swaminathan 2010) usually refer to the presence, in the villages that have been studied, of landlords and/or capitalist farmers and rich peasants. At the same time, the findings of the village studies by Himanshu, the Rodgers, and a good many others quite strongly suggest that the process of class differentiation amongst rural households has been moderated, even if income inequality has increased. As the authors of the paper on Bangladesh presented at the conference put it, what is observed is a process of “pauperisation rather than differentiation” (Hossain and Bayes 2014). Vikas Rawal, in a paper published some years ago (2008), shows that there is good reason for thinking both that landlessness has increased amongst rural households, and that inequality of landholding has increased. This reflects in part the proliferation of marginal landholdings that has taken place — holdings that are reproduced, exactly as the evidence referred to above shows, through reliance on off-farm and non-agricultural sources of income. In these circumstances there are reasons for thinking, as Gerry and Janine Rodgers suggested in their presentation, that “agrarian classes can no longer be conceptualised independently of urban structures.” Many of those who attended their session at the Conference appeared to agree, and to recognise with them that “households’ economic networks cut across different production relations in different locations.” These connections differ across caste, community and class, and it is necessary to consider urban class structure, urban labour market segmentation and the disparities in access to decent employment, in any analysis of “agrarian change” (quotes here are from the Rodgers’ presentation at the conference).

The Rodgers’ argument recalls that of Henry Bernstein who, in several publications, has argued that in many rural economies, rather than the peasant classes distinguished in classical agrarian studies, we find “classes of labour.” As he has put it, classes of labour in the conditions of today’s “South”

have to pursue their reproduction through insecure and oppressive — and typically increasingly scarce — wage employment and/or a range of likewise precarious small-scale and insecure “informal sector” (“survival”) activity, including farming; in effect, various and complex combinations of employment and self-employment.

As Jeyaranjan, Nagaraj and I wrote in our study of the village of Iruvelpattu in Tamil Nadu, this description fits very well “the increasingly complex, highly diversified ways in which the very mobile people [of that village] secure their livelihoods across a geographically wide range of locations” (this and the preceding quote, Harriss, Jeyaranjan, and Nagaraj 2010, p. 61; see also Lerche 2010). The description also fits very well the conditions of life of many people in rural India, including those of Palanpur and the Bihar villages of the Rodgers’ research, in States in which the non-agricultural economy is less well-developed and less extensive than it is in Tamil Nadu.

These observations are not in conflict with the findings of the researchers of the FAS concerning the presence, in the PARI villages, of landlords as well as capitalist farmers and rich peasants. It appears, for example, from Rawal’s paper that there are significant numbers of such households in all the villages surveyed. It is not that other scholars do not recognise their presence, of course, but the observation of the Rodgers’ that “between 1981 and 2009 [there was a] sharp decrease in tied labouring households and in landlords” — elsewhere they have spoken of “semi-feudalism meeting the market” (2001) — is one that corresponds with the findings of village studies in other States as well. One example is in a study from Eastern Uttar Pradesh by Ravi Srivastava (presented at a seminar held at the LSE, May 2014). The observations in village studies from Tamil Nadu that the erstwhile landlords are tending to move out of agriculture, some of them altogether, may not be specific to this particular State (Harriss and Jeyaranjan, forthcoming). It will be important, therefore, moving forward, to undertake further analysis of class relationships in the PARI villages, and their “sociological contexts” (as Sripad Motiram put it in an intervention) — thinking, for example, of past research that has shown up the significance of the particular character of the present or previous dominant castes in a village, for the organisation of agriculture, for labour relations, and for interlinked transactions more generally (e.g. Gidwani 2000). In this regard, it may be that a stronger focus than at present on explaining differences between the villages will be fruitful. As Jens Lerche argued in his ambitious paper on “Regional Patterns of Agrarian Accumulation in India”:

There are significant differences in agrarian growth and accumulation trajectories across Indian States and...this is best understood as an outcome of major differences not only in agro-ecological conditions but also in class constellations and accumulation strategies of the dominant rural classes (Lerche 2014, Abstract).

There are a number of questions about the role and positions of landlords, and the dominant classes more generally, in rural India today. One question goes back to an earlier body of literature and has to do with the implications of differences in agrarian power structures for agricultural productivity — to which Lerche refers, albeit indirectly. Another has to do with agrarian politics. At the time that this Rip Van Winkle went to sleep, farmers’ movements were politically significant in India. It was generally considered that they articulated the interests in particular of the most commercialised cultivators, though they were more or less successful in drawing support from across rural society. These movements appear to be much less significant today, as Jonathan Pattenden (2005) has shown with regard to the Karnataka Rajya Raitha Sangha. But how then are the interests of the dominant classes of rural India now articulated? The conference papers had rather little to say on this question.

The papers did, however, have a good deal to say about rural labour relations. Praveen Jha summed up evidence on labour conditions and wage employment — among those who depend on the sale of their labour power for their livelihood (see also remarks above on the contribution from the PARI studies by Dhar, Dixit, and Sivamurugan). Rural labour has been increasingly casualised — and feminised — and there has been a sharp increase in self-employment in non-agriculture, which Jha referred to as a kind of a “parking lot” for labour. Increased migration for work has commonly brought about the tightening of rural labour markets; and real wages have risen in the more recent past after a long period in which they stagnated or declined (following a trend of increase in the 1970s and 1980s). “Labour shortages” in agriculture are widely reported, but it seems that such reports reflect the greater assertiveness of those who have been historically oppressed as much as, or more than, do movements of supply and demand. As Jha said, the historically oppressed castes have gained much more “elbow room.” Yet, as he also said, evidence on the expansion of the formal freedoms of labour is mixed, as is that on the bargaining power of labour households. There is considerable diversity, with respect to the evidence on formal freedoms and bargaining power, including amongst Dalits, as is shown in studies from Tamil Nadu (e.g. Carswell 2013), and the widely observed shift to piece-rate contracts benefits some, but not all.

The trajectories of change in rural India, as they were brought out in papers and presentations at the conference, are not quite those that were anticipated forty years ago. The bypassing of the agrarian question has left behind huge problems of marginalised, and in some sense “excluded”, labour (as the late Kalyan Sanyal argued: 2007), in an agricultural economy which is much less productive than it might be. As was argued (notably by Sheila Bhalla) at the Conference, “agricultural crisis” and “agrarian crisis” should not be treated as being synonymous. Even if the “agricultural crisis” is no longer as acute as it appeared to be ten years ago, its resolution — probably only temporary, in any case — does not solve the crisis of agrarian lives and livelihoods.