‘An elephant in Lucknow is eating up all the NREGA money we are sending from New Delhi.’
(Rahul Gandhi, political rally in Lucknow, August 2009)
‘Na khaunga, na khaane doonga.’
(Neither will I eat, nor will I let others eat.)
(Narendra Modi, a common refrain in election speeches, 2014)
Introduction
The elephant that Gandhi was referring to was the Bahujan Samaj Party (BSP) which carries the election symbol of an elephant.Footnote 1 The BSP was then ruling over Uttar Pradesh from the capital city of Lucknow, and Gandhi was angry at the poor performance of the Mahatma Gandhi National Rural Employment Guarantee Act of 2005 (henceforth NREGA) in the state. Similarly, a common refrain of Narendra Modi during the electoral campaign of 2014 was his assertion that he himself won't eat money and, furthermore, that he will prevent others from doing so. In other words, he was laying forth his own clean/non-corrupt credentials and promising that he would impose these non-corrupt measures on others as well. Given that he had already been elevated to the prime ministerial candidate of the BJP, this was seen as Modi's pithy anti-corruption declaration. The metaphor of eating money is one that is commonly ascribed in the Hindi-speaking belt of India to describe practices that are glossed together as ‘corrupt’. To claim the eating of money by others, especially in the context of developmental programmes such as NREGA, is an ever-present mode through which discontentment with the functioning of the Indian state is voiced. Almost everything that is wrong with India—especially its abysmal social indicators of health, education, poverty, unemployment, and low growth—is traced directly to the predilection of the state, its agents, and other actors to ‘eat’ money. But what exactly is it to eat money? How is it accomplished and by whom? What are the effects of money-eating on welfare measures? And what has the push to make the state transparent and accountable achieved in post-liberalization India?
In this article, I work towards a disaggregation of ‘corruption’, which, as many have noted, covers a confusingly large range of practices.Footnote 2 I do so not to build up a new typology of corruption where one grades certain practices as falling on a spectrum of—for instance—gift to bribe.Footnote 3 Neither am I distinguishing analytically between ‘retail corruption’ and ‘elite/high-level corruption’Footnote 4 or between various sorts of corrupt practices such as ‘bribe’, ‘commission’, ‘fee’, and so on.Footnote 5 I do, however, wish to avoid collapsing of all such actions/practices into one—villainous—category of corruption. To circumvent this commonplace approach to corruption, I prefer to focus on those practices that are discussed as money-eating. In the Hindi language, this corruption talk often takes the form of speculating on how much an official ‘eats’. Or it is expressed by discussing how ‘fat’ the person has become because of the amount they eat. To describe someone as waiting to indulge in the eating of money is similarly done by describing their hunger (bhook). Those who's hunger was unquenchable were described as constantly hungry (hamesha bhooke) or, against the background of intense human–big cat conflict in the Himalaya, as ‘bhooke bagh’ (hungry big cats). A form of corrupt giving is often described as ‘chai-paani’ (tea-water) or supplicants are advised by others to feed the official (khilana padega).Footnote 6 Sometimes they are warned that such and such person eats a lot (bahut khata hai) or—to imply their ‘cleanness’ (safai)—doesn't eat at all so there is no need for khilana-pilana (eating-drinking).
Most anthropologists working in India have commented, in one way or the other, on the ubiquity of corruption talk they encounter in the field. Gupta, for instance, takes these narratives seriously as ethnographic data and shows their discursive effects.Footnote 7 It is important to point out that the discourse of corruption that he and others present is just as rife within different levels of the state—amongst the supposed corrupt agents—as in other spaces. So it is omnipresent in everyday chatter amongst the block offices flung out on the Himalayan borderland that I worked in and it is just as evident in the annual ‘election’ of the most corrupt IAS officers that the Uttar Pradesh IAS association undertook a few times.Footnote 8 I describe both the discourse and practice of eating money by rooting it within the specific context of one development programme. Furthermore, my entry point into the study of corruption is through a focus on precisely that measure—transparency—that has been recently introduced in India in order to stymie it. Transparency is meant to illuminate the functioning of the state, and the assumption is that this new luminosity of a notoriously opaque system will, in and of itself, hinder corruption. Transparency, in its current conceptualization by the developmental Indian state, emerges as the categorical Other of corruption. The vehicles for rendering corruption transparent and effacing it in the Indian state that I am considering here are two legislations passed in 2005: the National Rural Employment Guarantee Act (NREGA) and the Right to Information Act (RTI).
In the language of the state, the NREGA and RTI can both be seen as legal instruments that will stem the outflow of government money—what are euphemistically discussed as ‘leakages’—from the Indian state. In the imagination of the law, higher echelons of the state bureaucracy, development policy and organizations, as well as in much of the Social Sciences literature on corruption in India, the state appears leaky. Like a leaky vessel, it constantly drips out valuable resources that have the ultimate effect of stripping it—quite literally—dry. It is this image of the leaky Indian state that is being targeted by transparency regimes and auditory measures that will, it is believed, stem the (out)flow of government funds.Footnote 9 When I term the Indian state a ‘leaky’ one I am referring to a bureaucratic system comprised of a diversity of actors and institutions that is unable to close in on itself; that perpetually brims over; that cannot quite achieve that much-hankered-after watertight character.
The metaphor of leakages and talk of leaky states are evocative of a generalized belief that corruption is the root cause of underdevelopment in India, brought on by poor or bad implementation of well-designed policies and programmes. This belief is either explicitly assertedFootnote 10 or lies submerged below the surface in studies and discussions of why the developmental Indian state fails to deliver on its well-meaning and beauteous promises. There is, first, the very real problem of obtaining reliable empirical data on corruption in India, as opposed to its discourse and perception.Footnote 11 Indices constructed using rather dubious methodologies by organizations such as Transparency International continue to rely on perceptions of corruption—its talk and discursive presences—rather than on its actual existing practice. All my field spaces—ranging from small offices in the district headquarter and blocks to villages and the archives of vernacular newspapers—were saturated with corruption talk, but it was only rarely that I directly witnessed corruption-as-practice. Of course, it is to be expected that these transactions would not be undertaken in front of me and that, in general, they take place ‘under the table’ as the popular euphemism goes. However, the paucity of such experiences has led me to consider more seriously just how closely talk of corruption reflects actual practice. This is not a question I have an answer to, but I do want to pose it before moving onto my far more central problem with the corruption literature in India. I find an excessive focus on corruption detracts from the core issues of how the Indian state actually functions, what it does, when, how, and why.
A central contention of this article is that the current, widely accepted account of corruption as the malaise that hinders ‘real’ (asli) development in India has inhibited our capacity to better understand the functioning of the Indian state. I argue that we need to eschew treating corruption as an explanatory trope for the failure of development. What is urgently required is, first, a stronger empirical and disaggregated detailing of what corruption is and how it operates in everyday life in contemporary India. Subsequently, and on the basis of this work, we need to reconsider the morality-laden, normative tropes—derived from a disturbing colonial past and finding sustenance in a global development-speak that consistently identifies South Asian states as defined by lack of ‘transparency and accountability’—that continue to structure and define the terms of the hegemonic corruption talk in India. This article, drawn from my longer work,Footnote 12 is an attempt at precisely such an exercise in shifting the frames through which we understand and discuss discontentment with the grandiose and overarching institution of the developmental state in India.
A recent collection of essays on corruption in South Asia has tackled the study of corruption through a focus on the similarity and intersections between words such as venality and veniality,Footnote 13 and by viewing corruption through a focus on corruption of intimate relations, memory, and natural resources, such as a river in this era of climate change.Footnote 14 As Das shows beautifully, through her interweaving of the literary and the ethnographic, the anthropological study of corruption requires a greater appreciation of what she describes as the ‘possibility of life’ itself.Footnote 15 These essays remind us that a shift in the way we understand corruption can only begin when we change or reimagine the pejorative vocabulary through which it is discussed and which has a long stigmatized history on the sub-continent. Let us, then, reimagine corruption and the Indian state through an ethnographic account of the eating of money and its luminous Other, and against the noisy backdrop of the splish-splash of money as it slowly but surely leaks its way out.
Two paradoxes
This article works through two paradoxesFootnote 16 that emerge from over a decade's research on Indian bureaucracy and the implementation of welfare programmes. The first paradox relates to the corruption–transparency dyad. My research shows that the state's push to render its workings transparent in the implementation of NREGA was, in fact, what could be termed a success. However, this successful push to produce transparent functioning had the entirely unanticipated result of making the NREGA very difficult to implement, for reasons that I touch upon below.Footnote 17 In short, transparency in operations and the decrease in money-eating negatively impacted the take-up of NREGA in Uttarakhand as the desire and motivation to implement it was on a wane due to its stringent transparency requirements.
When the question of the linkages between ‘corruption’ and implementation in the Indian state emerges, the discussions moves, broadly speaking, in two directions. The first, is an instrumentalist one that pairs up with neo-Orientalism to argue that, in fact, corruption is good at greasing the wheels of underdeveloped and unequal states such as India and, in fact, the absence of such practices and patronage networks would impede the functioning of such societies.Footnote 18 My ethnography of state process and things below will, I hope, distance this work from the simplistic utilitarianism of such work. In a sophisticated account, Gould traces events, narratives, and publicity in 1930s–1960s Uttar Pradesh to demonstrate how this dominant discourse of institutionalization of corruption in India emerged.Footnote 19 The challenge that Gould's work then presents to an anthropologist is not so much to posit the systemic nature of money-eating but rather to exhibit the complexities of its actually existing practice, and to ask what we might learn from such an exercise. So Shah describes the systematic capture of development project implementation in Jharkhand by local elites, who make their cuts through modes that bear strong resemblances to what I encountered in Uttarakhand.Footnote 20 In contradistinction to the World Bank and other development institutions, Shah contextualizes these practices within a local moral economy. Without in any way condoning corruption, Shah deftly removes the implications of a degenerate or deficient society that are always lurking in international development accounts.
Related to the systematicity of corruption in India and the location of these practices within their social and moral orders approach, there is burgeoning ethnographic literature on what so-called corrupt practices actually do for certain constituencies. Anjaria and Jauregui have shown that everyday corruption in India opens up, if briefly, what Anjaria describes as a ‘space of negotiation’, and Jauregui, by working through the concept of jugaad, considers the practical capacity for getting things done.Footnote 21 Das further complicates the study of corruption by pointing out its simultaneous banalization and condemnation in everyday life, as well as by exploring the proximity of honesty and corruption. Corruption is not normalized, in her work, but neither is it demonized; it rather assumes different avatars within ethnography, literature, the everyday, and politics.Footnote 22 This article, too, follows the ethnographic presences and absences of money-eating in the Himalaya, but with a critical difference. These new ethnographies have arrived at their conclusions through work with those who are seeking certain resources: space on Mumbai streets as vendors (Anjaria), access to land or electricity in a low-income Delhi neighbourhood (Das), and even just the capacity to do one's own routine police work in a scenario where these state agents themselves feel only partially empowered (Jauregui).
I have arrived at a similar place from a different site—the bureaucratic everyday—and through a tracing of the effects of anti-corruption measures not just on bureaucrats but also on non-human actants including paper and ID cards. And herein emerges the second paradox of this article. Once we have a clearer understanding of what the Indian state is in practice, and how it operates and makes these operations visible and manifest, we are able more clearly to understand how the state can then also claim to be transparent or un-corrupt. The central evidentiary proof that attests to both the functioning and transparency of the Indian state is its paperwork.Footnote 23 Thus, the stemming of the leakages from the Indian state can only ever be arrived upon by scrutinizing precisely that paperwork which masks the leakages in the first place. In other words, we can only ascertain anti-corruption measures have been successful through a reliance on those artefacts that allow corruption to occur in the first place. To think through the implication of this quintessential Catch 22 situation we need to understand the paradoxical effect of transparency in the implementation of NREGA in the leaky Indian state. Following on from that, we will be better positioned to understand how the state (is able to claim it) functions. It is my hope that if one makes a greater effort to really understand how—and through what material and affective practices—the developmental state in India is birthed, the hegemonic hold of the literature and discourse on villainous corruption and bad implementation will weaken. Instead of joining the large chorus that spends its time bemoaning what is popularly described as the ‘cancer of corruption’ in India, and dreaming up ever-expanding auditory and punitive measures to somehow control the truant state, we need, I suggest, to concentrate our attention on the much harder task of articulating the banal bureaucratic everyday that produces sarkar in the first place.
Guaranteeing corruption-free employment in a leaky state
Rural public works programmes in India today draw directly from famine relief measures instituted by the colonial British state in the late nineteenth century.Footnote 24 In the late 1970s public works programmes morphed into an answer for the more recalcitrant issue of endemic malnourishment and chronic poverty in rural India. From 1980 onwards a series of public works schemes were designed for rural India. These were all extremely targeted programmes, which screened beneficiaries for eligibility on the basis of measures such as age, gender, caste, class, income, region, and size of landholdings. Large outlays were devoted to this central strategy of rural development by the Indian state in successive five-year plans. Public works scheme followed public works scheme, with minor modifications undertaken to their finer details. The objective of all these schemes was the same: to ‘pull’ or ‘lift’ people out of poverty through the active generation of employment on public works and repayment in the forms of wages or food or, sometimes, both. The success of these schemes was measured against statistics such as total outlays and expenditure, number of people provided with employment, and so on, with a sharp eye kept on overall rural poverty figures.
In 2005, public works programmes for rural India were to witness what is often described as a ‘revolution’, with the enactment of the National Rural Employment Guarantee Act or NREGA. The name Mahatma Gandhi was to be subsequently appended to NREGA, turning it into the more wordy MGNREGA, but I will continue to refer to the statute as NREGA in this work. With the enactment of NREGA, any adult rural resident of India can enforce his or her right to work by demanding, in writing, work to be provided by the state. By the terms of this agreement, the work to be provided shall be manual in nature, will be on public works in his/her vicinity, and shall be compensated with minimum wages. If s/he is not provided with work within 15 days of making the demand then s/he must be provided an unemployment allowance by the state. NREGA is similar in intent to the plethora of employment schemes/public works programmes that preceded it. Further, much like its predecessors, it swallows up extant schemes, thereby allowing for the establishment of a clear genealogy of rural employment programmes. However, NREGA distinguishes itself from its predecessors by virtue of being a legal guarantee of employment provided by the state. Enshrinement into the Indian Constitution is meant to confer legal rights on the entire rural populace—that is 73 per cent of India's 1.2 billion population—who can now demand ‘the right to work’ instead of pleading for state largesse.
The most vociferous opposition to the NREGA at the time of its passage in Parliament and subsequently, right up to the present day, comes from those who worry about ‘leakages’ in the state. Public works programmes in India have traditionally been associated with extremely high levels of leakages. The massive budget that the NREGA would command, critics claimed, would only precipitate corruption in an already faulty delivery system. Thus, an article in the English-language national newspaper, Pioneer, was entitled ‘Rename NREGA as Corruption Guarantee Scheme’, and the magazine Business Standard carried an article entitled ‘Corruption with a human face’—a dig at the ruling coalition's slogan of ‘Development with a human face’.Footnote 25 Advocates of the NREGA responded to the well-founded allegations of corruption by pointing out the very different legal status enjoyed by this programme. Briefly put, the argument goes that the introduction of law will, somehow, ‘create transparency’ and an attendant ‘accountability’, which in turn will slay corruption. It is through the direct tackling of corruption in the delivery of welfare measures that this statute not only somewhat silenced its oppositional voices but, also, can boast a new-ness that has the power to render the working of the state transparent to its citizens. As the prime minister declared at the inaugural speech of the NREGA in 2006:
The NREGA must be a transparent, people's programme. Therefore, methods of estimation and measurement of works and rates of payment for each task should be made transparent. The labour that seeks work must understand what is offered, on what terms and demand its full entitlement. Similarly, there should be complete transparency in maintenance of muster rolls and payment of wages. There should be fairness all around. Footnote 26
Transparency is said to be built into the NREGA through two broad means. The first, in the words of the Act, is through its somewhat incestuous ‘marriage’ to its ‘sister legislation’, the Right to Information Act 2005 (RTI). Passed alongside the NREGA, the RTI is aimed at making the Indian state transparent by overturning the anti-disclosure of state information terms set up under the Officials Secret Act of 1923.Footnote 27 Both these legislations share kinship in their core mission of increasing the ‘accountability’ and ‘transparency’ of the Indian state in order to ‘contain corruption’. The RTI states this ambition upfront: ‘Whereas the Constitution of India has established democratic Republic; and Whereas democracy requires an informed citizenry and transparency of information which are vital to its functioning and also to contain corruption and to hold Governments and their instrumentalities accountable to the governed’ (The Gazette of India 2005).
The second and related mode through which transparency is to be ensured is via an ‘innovative feature’ of the NREGA, which is the ‘central role’ given to ‘social audits’.Footnote 28 Whether through the means of social audits or the marriage to the RTI, the law stipulates that all official documents or statistics, which are created or transacted in the process of undertaking any work under NREGA, should be available to anyone desirous of seeing them. In other words, transparency shall be effected or ‘made’ through documents: their production, circulation, display, control, and auditing. Elsewhere, I have shown the ironical effect of these ‘transparent-making documents’ to be the creation of a crisis in implementation.Footnote 29 This crisis was made visible not just in the near-total absence of NREGA works or labourers in villages in my field sites in Uttarakhand, but also in the millions of rupees that were lying unutilized in government bank accounts. In this earlier work, I focused on the everyday worlds of street-level bureaucrats and the paperwork they laboured on in order to produce evidence of action. For it is through the production of variegated documents and attendant statistics that the state attests to the discharge of its duty of effecting development. The push for transparency in NREGA had not only led to an exponential explosion in the paperwork demanded of an already overworked and creaky bureaucracy, particularly those positioned in villages and Block offices, but also some very astute designing of the scheme had hindered the eating of money.
The transparency clauses built into the NREGA have allowed for a new system of documentation to be introduced into an arrangement (bandobast) that was considered traditional (paramparik). Referred to colloquially as the Contractor Raj and the PC system, this is a mode of operation in which a private contractor oversees the execution of public works and rural employment schemes. Public works schemes, it was openly acknowledged, were primarily run by contractors. For instance, the Sampoorna Grameen Rozgar Yojana (SGRY), which is what the NREGS supplanted, was considered ‘easy’ to implement, as the district faced no problem in exhausting the annual budget allocated to it. The ease of implementation of such schemes was directly related to the paperwork involved. As my office-mates told me, if you gave a million rupees today to the village headperson, s/he would produce a receipt for the scheme the very next day—and that was that. The way the Contractor Raj worked was, crudely, this: The village headperson would be given, say, Rs 50,000 to build a road in his/her village. S/he would immediately hand over Rs 30,000 of this amount to whichever contractor was dominant in the region and maintained close links with the local politicians. From that Rs 30,000 this contractor would build a road using substandard material, and by employing easily exploitable migrant labourers from neighbouring Nepal or from the north-Indian states of Uttar Pradesh and Bihar, who would work for below-minimum wages. By these means the contractor would be able to extract his/her own ‘cut’ from the amount received. As for the remaining Rs 20,000, it would be divided up between the local government functionaries under what is commonly known as the ‘PC system’. PC refers to the percentage that is ‘cut’ from total amounts and reserved for each official depending on his/her position. Thus, a junior engineer can demand a PC of 7 per cent, a Block Development Officer (BD) of 10 per cent and a Gram Panchayat Vikas Adhikari (GPVA) or village level worker of, say, 5 per cent. The general belief in Uttarakhand is that the PC system ‘eats up’ at least 25–30 per cent of total funds available.
In direct recognition of the Contractor Raj, the NREGA makes the employment of contractors in execution of works a legally punishable offence.Footnote 30 Criminalizing contractors has been coupled with the introduction of new documents designed to remedy the ease with which books could be fixed hitherto. The required documents under previous schemes consisted merely of a work order, a muster roll, and receipts, all of which could be handled by a contractor and/or his/her allies such as the local government functionary. The nexus between government functionaries and contractors in Uttarakhand is so strong that it becomes difficult to disentangle them. Thus, in many areas the village headperson was the contractor, or the village-level development worker was closely related to the local contractor. Between them, they controlled the documents that attested to the existence of the scheme. The MR was always a kachcha, i.e., a temporary one that was made pukka (permanent/correct) before being submitted for official inspection. Receipts were easily manufactured and work orders just as easily obtained from the engineer in charge, on payment of the appropriate PC. Thus, on paper, these previous schemes were easily implementable. In contrast to this easily controllable and manipulable system the documentary logic of the NREGA was much more complicated. By law it was ordained that the muster rolls must be pukka and bear their unique registration number, be kept at the worksite, be accessible to any who wished to inspect them, and correspond precisely, in the details of their entries, to the entries made in an ID that workers themselves control. This ID, a tiny but vital document called the job card, is legally supposed to be in the possession of the household to which it has been issued at all times. In effect, then, this law has created a stringent system of double accounting: one involving not only a large number of mutually corresponding documents that different state actors are required to produce, but also, crucially, a document that centrally attests to the truth of the scheme and resides outside the domain of the state, in the hands of the labourer.
In short, the job card played a key role in upsetting long-standing practices of eating money. This little document looks innocuous, but in practice it had allowed for the setting up of a system of double accounting that rendered auditable, perhaps for the first time ever, rural public works programmes in India. The details of wages paid and days worked are entered into the job card and these details are expected to perfectly match onto the official muster roll. It is through a cross-matching of the job card and muster rolls that the auditing of NREGA is undertaken. If and when the entries don't match, it is found to constitute evidence of ‘corruption’ and action against officials could be—theoretically—taken. Though in practice the job card was subjected to a whole range of manipulations, the spectre of the law coupled with the copious and complicated nature of transparent-making documents did dissuade contractors from directly picking up NREGA.
Contractors would tell me straight out that they were wary of the NREGA as it involved ‘too much work’ with all the documents and registers that it has in its wake. To deal with the job card, in particular, required extra ‘effort’ that they were not willing to undertake. It requires exerting control over a much larger dominion than previous schemes required. As there were plenty of other big state construction schemes still running in Uttarakhand, which did not involve similarly complicated documents, contractors decided to focus their attention on them. To put it somewhat crudely, in Uttarakhand it was not the eating of money by contractors and other middlemen that had led to NREGA being, as it was often described, ‘a flop show’ very much unlike its predecessors. Rather it was the fact that middlemen, contractors, and particular officials were so wary of touching a law (kanoon) that had enforced papery mechanisms to obstruct money-eating, that the scheme was facing a crisis of implementation. And yet all the talk of NREGA's poor performance in Uttarakhand—gauged through the exhaustion of allocated funds—centres upon the eating of money. When I would go to villages and ask why there was no work happening under the auspices of this programme, people would say it is because all the money is being eaten up (sab paisa khaya ja rahan hai). Sonia Gandhi, too, criticized the Uttarakhand government for the loot and cheating that was disallowing the NREGA from being implemented. In January, 2012, she began her party's electoral campaign in Uttarakhand by highlighting the corruption of the then-incumbent Bharatiya Janata Party's (BJP) government. In her speech she declared ‘The rampant loot and the cheating are still prevailing. Only 10 per cent of the people got the benefits of the Congress-led UPA Governments Mahatma Gandhi NREGA policy. Only 40 per cent of the funds were utilized by the BJP Government’.Footnote 31 As I argue above and elsewhere, the underutilization of the funds was due not to looting and cheating but, perversely, due to the push for transparency.
It was only within the state bureaucracy that the non-performance of NREGA was sometimes hinted at. In a workshop between village headpersons and state representatives that I attended in Dehradun (the capital city of Uttarakhand) I once heard an official knowingly say that the problem with NREGA implementation lay in the manner in which it hinders the ‘skimming off of the cream from the milk’ (doodh se malai hatana). It was in this workshop and in further conversations with street-level bureaucrats as well as labourers in possession of job cards that made me think more closely about the doing of development work in rural India. At first glance, the Contractor Raj and its partial disruption by the transparency-demanding NREGA would appear to attest to the fundamental ‘corruption’ that bedevils the Indian state. This would contradict my overall claim that corruption is not the answer to the question of why laws/development programmes struggle to be realized in letter and spirit. However, I stand by my original argument and want to take this opportunity to think more deeply about what corruption is and how it is analysed. In Uttarakhand the Contractor Raj and the associated PC system is not considered—if we use the same moralizing language that NREGA activists, development experts, policy wonks, and the media do—a good thing per se. But then neither is the state or state-led development (sarkari vikas) or anthropology. So let us, for a moment, suspend moral judgement and stop thinking of this ‘corrupt’ practice through the dichotomy of virtue and vice, instead focusing only on what the contractors were enabling: the execution of a rural employment programme, albeit at reduced wages for the labourers, and with the very real possibility of the elimination of a range of eligible beneficiaries who did not fit into the contractors’ social networks. Thus, the situation was not ideal, and the drafters of the NREGA recognize this only too well and therefore are keen to eliminate it. But in a region where the state commands extremely low material resources, the Contractor Raj did allow work to take place. My informants felt a direct animosity to the local contractors and described them as goondas (thugs) and badmash (rogues). Yet they, too, thought that the Contractor Raj enabled something to happen which the state on its own was struggling to accomplish. Furthermore, even in places where the contractors were absent, a similar process operates whereby state agents disallow certain communities or individuals from benefiting from the welfare programmes. These state agents, similarly to contractors, ‘eat money’ by cooking figures and keeping kaccha muster rolls. The distinction between the state and the non-state in villages is hard to maintain and, for all practical purposes, doesn't really matter.
Beyond the fact of overworked and underpaid state employees, there is a more crucial aspect of the developmental state that is pushing me to reconsider the effects of the Contractor Raj. This is linked to what is considered the ‘real work’ of the (paper) state by its employees, which is keeping the sarkari zindagi (state life) of the law alive. An ethnographic approach to the law, bureaucracy, and the state highlights how and through what substances the developmental performance of the state is birthed. While the agents of the state are busy making sure that the state life of law keeps going, by working on their papers and files, and by participating in state rituals like meetings, someone has to go out and do the more gritty, hands-on work that a scheme such as the NREGA demands. The contractors were in possession of the time, motivation, savvy, and social relations to handle this. I also find it telling that the most comprehensive analysis of the performance of the NREGA thus far, undertaken by the Comptroller and Auditor General of India (the CAG) points to the difficulties in expending the budget and making the programme implementable. Covering a period from April 2007 to March 2012, the CAG report notes that in 14 States and one Union Territory, only 30 per cent of the annual plan budget was used, thus indicating ‘significant inefficiencies in implementation of annual plans’.Footnote 32 But how exactly—in the language of the CAG, the state, and all welfare agencies—is implementation of NREGA achieved? I argue that it is the play of paper, documents, files, and ID cards, the routine meetings and inspections, and the culture of sarkar that all allows the state to make itself manifest and produce evidence of functioning. I am interested in an elucidation of an ontology—the prosaic realities of the developmental state in the Indian Himalaya. Here the papery worlds of bureaucrats and the existence of the Contractor Raj is a ‘social fact’ in Durkheim's sense of the words.
The sarkari (state-like/official) and the asli (real)
This ethnographic approach to the functioning of the developmental state pushes me to make a further—and a more foundational—point about the eating of money in the leaky Indian state. While it is true that the forms of documentary experiments implemented by the transparency-seeking NREGA did unsettle the Contractor Raj, they did not eliminate it. On the contrary, as Aiyar and Mehta point out from their work in Andhra Pradesh, which is often held up as an exemplar of NREGA implementation, new actors have emerged and other means have been found for the eating of money. They describe a ‘deepening of the corruption syndicate’ and a shift in the manner in which money is pilfered, which allows for the programme to be executed.Footnote 33 Furthermore, I want to caution against the assumption that underpins almost all auditing of NREGA, particularly by the state, that the documents it has produced map onto the real (asli). A clear distinction was made, by my interlocutors, between the sarkari (state-like or governmental) and the asli (real). The sarkari is what we found in official documents (kaghaz) but they bore a complicated relationship to what was really happening, if we take the labourers testimonies as constituting evidence of the actual. The correspondence between job card and muster roll that was required by law was described as a ‘headache’, ‘a pain’, or as one person once put it like an elephant that has gone mad (mast hathi) and is messing up the nice system of yore. They posed hindrances, as controlling and mastering this new form of paperwork requires effort and more work, not because they constitute reliable testaments of what was really taking place. The belief that documents somehow map onto the real is, at best, naïve. In fact, local officials are themselves the first to point out the artifice of documents, their fake (nakli), government-like (sarkari) quality. At the same time as they dismiss the veracity or, as they would put it, the ‘real-ness’ (aslipan) of these official documents, they are acutely aware that it is only through their production and transaction that the implementation of the NREGA would be adjudged. Distant from the everyday life of the state with its dreary paperwork and ritualized performances of state-ness, these very documents are held up by proponents of the NREGA, politicians and senior state officials in New Delhi as empirical evidence of the success of a transparent, corruption-free welfare programme in rural India. No doubt there was an initial reduction in the ‘leakages’ that traditionally accompany public works programmes in India. However, as the experience from other states shows, the new documentary regime could be overcome with some more ‘effort’ or through innovative new means to eat money. In Uttarakhand, too, new alliances were set up that could and did allow for a mastery over those very documents that were troubling the eating of money. Therefore, to argue either that corruption was entirely eliminated or that the state's functioning was rendered visible, would be an utter simplification. It would involve treating transparent-making documents as empirical manifestations of a real (asli) of the state (sarkar), and not, as my informants always claimed, as mere matter that must be moulded and cajoled into producing an appearance of transparency.
An instance of the utter malleability of official records that allow for money-eating came across very clearly in a short comparative study I did in the South Indian state of Odisha with a team of activists in late 2007. Research teams fanned out in different Blocks of a total of three districts in the State. Results were swiftly arrived at after a fortnight of intensive surveying. In Kalahandi's Narla Block, where my group undertook an intensive survey, 48 per cent of the labourers’ statements that we recorded did not match the official muster roll. In the rare cases that we were able to locate the job cards, 24.2 per cent of their entries did not match the muster roll, and 12.5 per cent of the names entered in the muster rolls were fake. Gathering together questionnaires for all the Blocks in three districts, and after a daylong brainstorming session with the entire team, concrete results on the performance of the NREGA were arrived at. They were made public in a meeting with the press and the government at the District Magistrate's office in Boulangir district, where it was announced that only 60 per cent of the entries in the muster rolls that we sampled had been confirmed by the labourers. The figure of 40 per cent irregularities was a clear outcome of this exercise. The implication of this figure was that an equal, if not greater, amount of money was being ‘leaked away’ or ‘eaten up’ from the planned funds outflow of the NREGS.
The 40 per cent figure did generate some shock, particularly in the national Press, which got the results of this survey because it was led by a prominent activist. It was held up as yet another example of why NREGA was not working in Odisha and, indeed, much of India: due to corruption. Coming from Uttarakhand and from the little government offices and mountains of files in which I had been absorbed, what struck me was not so much the methodology used by activists to derive such (alarming) figures but more the manner in which the entire record keeping system of the state of Odisha had been obviated through structural flaws in the design of the job card itself. The proforma issued by GoI mentions the basic features a job card should contain, but it is up to states to design the specific ones themselves. The Odisha job card contained no column for wages paid, which defeated the very purpose of the document. Further, as the interim report of the survey went on to note:
Even the number of days worked is hard to verify, as the names of the labourer and worksite have been replaced by ‘codes’. To decipher a labourer's code, one has to refer to page one of the Job Card, which is often in English! The worksite code, for its part, is plain Greek for the cardholder. As for employment dates, they are supposed to be written in a rectangle divided into four boxes, but these are filled in different ways by different functionaries, further confusing the reader. In this and other ways, the Job Card is virtually unreadable, even for trained investigators—let alone semi-literate labourers.
One of the primary recommendations of our research in Odisha was to re-design the job card and redistribute it. The assumption in the operational guidelines of the NREGA was that by producing a detailed template, the actual Job Card in different states would do the same work—of enabling transparency. Yet, this assumption of sameness all over the country was belied in the way states produce different variants of what should be a basic bureaucratic document. In the re-design of a document the truths that could be derived upon through auditing exercises were lost. In Uttarakhand, while there were no such starkly problematic issues with document design, what the Odisha comparison showed was how easy it is to manipulate and obfuscate the sarkari kagahz and distance it from the asli. And, yet, the assumption of all auditing exercises—whether undertaken by the state itself or the media, activists, and NGOs or any other body—is to concentrate on the official documents as evidence and testaments of reality including of leakages and money-eating. If we are so quick to term the Indian state a paper tiger, then why are actors still holding onto the belief that the state's own documents will allow for a realistic assessment of the measure of its villainous corruption?
Containing money-eating in a state made of paper
This article has taken two lines of argumentation in its questioning of the prominent discontent with the Indian state: the eating of money or leakages, practices that are otherwise glossed together as constituting corruption. In the first place, I have argued that a recent attempt by the Indian state to contain corruption and render its actions transparent met with some success. This containment of money-eating had the ironic effect of rendering the developmental programme ‘unimplementable’ even as public opinion and political leaders would have us believe that the failure of the programme was an outfall of ‘corruption’. Thus, this ethnographic example serves to caution against knee-jerk reactions that pin all failings of the Indian state as arising from its corrupt nature even as it shows the counterintuitive effects of popular reforms.
The second and related intervention this article has made is to narrow its focus onto precisely those artefacts that were hindering the eating of money, in order to question their correspondence with truth. This correspondence of paper/official documents with reality is the sine qua non of all auditing measures in India at the moment. Without this basic belief in the veracity of official documents, this work contends, none of the checking for corruption and accountability that takes place at a massive scale all the time in India, could take place. Distant from the everyday life of the state with its dreary paperwork and ritualized performances of state-ness, malleable documents are held up by proponents of the NREGA, politicians and senior state officials in New Delhi as empirical evidence of the success of a transparent, corruption-free welfare programme in rural India. The implication here is that if we are trying to arrive at a realistic assessment of how money gets eaten up and how much and by whom, we need to distance ourselves from the state's own material testimony. But given the manner in which both the state functions and makes its results known to itself and its citizenry, is this even possible? In other words, how can we arrive at officially provable evidence of both money-eating and true-transparency if the papers that prove either/both serve as the core testimonials? This holds particularly true of the systemic forms of money-eating that creates the leaky welfare state. However, even for the high-profile scams—coal, telecom, fodder etc.—that periodically hit national headlines and involve the highest-level bureaucrats and politicians, the only way convictions can happen is if the files/documents prove culpability and complicity. The never-ending court cases and inquiries are centred upon precisely this establishment of criminal complicity through an examination of meetings, signatures, memos, scribbles, stamps, letters, and other official documentation with oral testimonies only providing leads to go on (rather than concrete actionable evidence).Footnote 34
In an era when corruption is the most reviled feature of the Indian state, and transparency the most deeply desired reform, my work calls for a pause in how we think about—and thus tackle—both. This article has worked through a shift in language from the technocratic to the quotidian vernacular. It goes on to explore the impact of the introduction of transparency on the eating of money in the context of India's largest anti-poverty measure. There is a cruel irony and a deep paradox to the fact that the ending of money-eating resulted in a low take up of the programme in a region of extreme poverty and unemployment. Perhaps what is also surprising there was that the programme failed due to simple changes in paperwork and the introduction of a little ID card. The next step taken by this article, and drawing from the longer work that underpins it, is to interrogate more closely how standard accounts of both the functioning and the dysfunctionality of the state are arrived at. The extreme reliance on material testimony, certainly in the state's own self-critical examination but even amongst independent auditing teams, is not just a question of epistemology: through what methodologies and knowledge practices do we measure and assess corruption in India? It also leaves us with a more profoundly political question: what structural, material, and moral changes are required to escape the paper tiger that is the Indian state and craft a truly transparent regime of government?
