Blurred Boundaries: Discourse of Corruption, Politics, and Imagined State (PDF)

blurred boundaries: the discourse of corruption, the culture of politics, and the imagined state AKHIL GUPTA-Stanford University While doing fieldwork in a smal I vii I age in North India (in 1984-85, and again in...

blurred boundaries: the discourse of corruption, the culture of politics, and the imagined state AKHIL GUPTA-Stanford University While doing fieldwork in a smal I vii I age in North India (in 1984-85, and again in 1989) that I have named Alipur, I was struck by how frequently the theme of corruption cropped up in the everyday conversations of villagers. Most of the stories the men told each other in the evening, when the day's work was done and smal I groups had gathered at habitual places to shoot the breeze, had to do with corruption (bhrashtaachaan and "the state."1 Sometimes the discussion dealt with how someone had managed to outwit an official who wanted to collect a bribe; at other times with "the going price" to get an electrical connection for a new tubewel I or to obtain a loan to buy a buffalo; at still other times with which official had been transferred or who was likely to be appointed to a certain position and who replaced, with who had willingly helped his caste members or relatives without taking a bribe, and so on. Sections of the penal code were cited and discussed in great detail, the legality of certain actions to circumvent normal procedure were hotly debated, the pronouncements of district officials discussed at length. At times it seemed as if I had stumbled in on a specialized discussion with its own esoteric vocabulary, one to which, as a lay person and outsider, I was not privy. What is striking about this situation, in retrospect, is the degree to which the state has become implicated in the minute texture of everyday I ife. Of course north Indian villages are not unique in this respect. It is precisely the unexceptionability of the phenomenon that makes the paucity of analysis on it so puzzling. Does the ubiquity of the state make it invisible? Or is the relative lack of attention to the state in ethnographic work due to a methodology that privileges face-to-face contact and spatial proximity-what one may call a "physics of presence?" In this article I attempt to do an ethnography of the state by examining the discourses of corruption in contemporary India. Studying the state ethnographical ly involves both the analysis of the everyday practices of local bureaucracies and the discursive construction of the state in public culture. Such an approach raises fundamental substantive and methodological questions. Substantively, it allows the state to be disaggregated by focusing on different bureaucracies without prejudging their unity or coherence. It also enables one to problematize the relationship between the translocality of "the state" and the necessarily localized offices, institutions, and In this article I attempt to do an ethnography of the state by examining the discourses of corruption in contemporary India. I focus on the practices of lower levels of the bureaucracy in a small north Indian town as well as on representations of the state in the mass media. Research on translocal institutions such as "the state" enables us to reflect on the limitations of participant-observation as a technique of fieldwork. The analysis leads me to question Eurocentric distinctions between state and civil society and offers a critique of the conceptualization of "the state" as a monolithic and unitary entity. [the state, public culture, fieldwork, discourse, corruption, India] American Ethnologist 22(2):375-402. Copyright© 1995, American Anthropological Association. blurred boundaries 375 practices in which it is instantiated. Methodologically, it raises concerns about how one applies ethnographic methods when the aim is to understand the workings of a translocal institution that is made visible in localized practices. What is the epistemological status of the object of analysis? What is the appropriate mode of gathering data, and what is the relevant scale of analysis?2 An ethnography of the state in a postcolonial context must also come to terms with the legacy of Western scholarship on the state. In this article I argue that the conventional distinction between state and civil society, on which such a large portion of the scholarship on the state is based, needs to be reexamined. ls it the "imperialism of categories" (Nandy 1990:69) that allows the particular cultural configuration of "state/civil society" arising from the specific historical experience of Europe to be naturalized and applied universally? Instead of taking this distinction as a point of departure, I use the analysis of the discourse of corruption to question its utility in the Indian context. The discourse of corruption turns out to be a key arena through which the state, citizens, and other organizations and aggregations come to be imagined. Instead of treating corruption as a dysfunctional aspect of state organizations, I see it as a mechanism through which "the state" itself is discursively constituted.3 In addition to description and analysis, this article also has a programmatic aim: to mark some new trails along which future anthropological research on the state might profitably proceed. The goal is to map out some of the most important connections in a very large picture, thereby providing a set of propositions that can be developed, challenged, and refuted by others working on this topic. In so doing, this article seeks to add to a fast-growing body of creative work that is pointing the way to a richer analysis of "the state" (some examples are Abrams 1988; Anagnost 1994, in press, n.d.; Ashforth 1990; Brow 1988; Cohn 1987a, 1987b; Handelman 1978, 1981; Herzfeld 1992a; Kasaba 1994; Mitchell 1989, 1991; Nugent 1994; Taussig 1992; Urla 1993; Yang 1989). I should point out that much more needs to be done to lay the empirical basis for ethnographies of the state. Very little rich ethnographic evidence documents what lower-level officials actually do in the name of the state.4 Research on the state, with its focus on large-scale structures, epochal events, major policies, and "important" people (Evans et al. 1985; Skocpol 1979), has failed to illuminate the quotidian practices (Bourdieu 1977) of bureaucrats that tell us about the effects of the state on the everyday lives of rural people. Surprisingly little research has been conducted in the small towns (in the Indian case, at the level of the subdistrict [tehsm) where a large number of state officials, constituting the broad base of the bureaucratic pyramid, live and work-the village-level workers, land record keepers, elementary school teachers, agricultural extension agents, the staff of the civil hospital, and others. This is the site where the majority of people in a rural and agricultural country such as India come into contact with "the state," and this is where many of their images of the state are forged. Although research into the practices of local state officials is necessary, it is not by itself sufficient to comprehend how the state comes to be constructed and represented. This necessitates some reflection on the limitations inherent in data collected in "the field." The discourse of corruption, for example, is mediated by local bureaucrats but cannot be understood entirely by staying within the geographically bounded arena of a subdistrict township. Although in this article I stress the role of public culture and transnational phenomena, I do not want to suggest that the face-to-face methods of traditional ethnography are irrelevant. But I do want to question the assumption regarding the natural superiority-the assertion of authenticity-im­ plicit in the knowledge claims generated by the fact of "being there" (what one may call the "ontological imperative"). Such claims to truth gain their force precisely by clinging to bounded notions of "society" and "culture." Once cultures, societies, and nations are no longer seen to map unproblematically onto different spaces (Appadurai 1986; Gupta and Ferguson 1992; Hannerz 1986), one has to rethink the relationship between bodily presence and the generation 376 american ethnologist of ethnographic data. The centrality of fieldwork as rite of passage, as adjudicator of the authenticity of "data," and as the ultimate ground for the judgment of interpretations rests on the rarely interrogated idea that one learns about cultural difference primarily through the phenomenological knowledge gained in "the field." This stress on the experience of being in spatial proximity to "the other," with its concomitant emphasis on sensory perception, is linked to an empiricist epistemology5 that is unable to comprehend how the state is discursively constituted. It is for this reason that I have combined fieldwork with another practice employed by anthropologists, a practice whose importance is often downplayed in discussions of our collective methodological tool kit. This is the analysis of that widely distributed cultural text, the newspaper (for an early example, see Benedict 1946; an exemplary recent discussion can be found in Herzfeld 1992b).6 I have looked at representations of the state and of "the public" in English-language and vernacular newspapers in India. By focusing on the discursive construction of the state, I wish to draw attention to the powerful cultural practices by which the state is symbolically represented to its employees and to citizens of the nation.7 These public cultural practices are enacted in a contested space that cannot be conceptualized as a closed domain circumscribed by national boundaries. Folk, regional, and national ideologies compete for hegemony with each other and with transnational flows of information, tastes, and styles embodied in commodities marketed by multinational capital.8 Exploring the discursive construction of the state therefore necessarily requires attention to transnational processes in the interstate system (Calhoun 1989). The interstate system, in turn, is not a fixed order but is subject to transformations that arise from the actions of nation-states and from changes taking place in international political economy, in this period that has been variously designated "late capitalism" (Mandel 1975) or the era of "flexible accumulation" (Harvey 1989). For instance, the new liberalization policies being followed by the Congress government in India since the 1990 elections can only be understood in the context of a transnational discourse of "efficiency" being promoted by the International Monetary Fund (IMF) and the collapse of the former Soviet Union, one of India's most important strategic and economic partners. Similarly, intense discussions of corruption in India in 1989,9 centering on a transaction in the international arms economy, bring home the complex intermingling of local discourses and international practices. What is the theoretical importance of these observations? Briefly, it is that any theory of the state needs to take into account its constitution through a complex set of spatially intersecting representations and practices. This is not to argue that every episode of grassroots interaction between villagers and state officials can be shown to have transparent transnational linkages; it is merely to note that such I inkages have structuring effects that may overdetermine the contexts in which daily practices are carried out. Instead of attempting to search for the local-level or grassroots conception of the state as if it encapsulated its own reality and treating "the local" as an unproblematic and coherent spatial unit, we must pay attention to the "multiply mediated"10 contexts through which the state comes to be constructed. In developing my analysis I have drawn substantially on other ethnographers of South Asia who have paid attention to the state. In her analysis of the rituals of development performed at the inauguration of a large water project in Sri Lanka, Serena Tenekoon (1988) demonstrates that the symbolic distribution of water in all directions across the landscape of the country becomes a means by which the reach of the state is represented. In this case, the literal enactment of traversing the space of the nation comes to signify the ubiquity and translocality of the state. Conversely, James Brow (1988) shows how a government housing project in Sri Lanka makes the state concretely visible in the eyes of villagers. Here, the emphasis is on the possibilities of imagining the translocal that are enabled by the embodiment of the state through spatial markers such as houses.11 blurred boundaries 377  Since the ethnography of the state developed in this article focuses on the discourse of corruption, and since corruption lends itself rather easily to barely concealed stereotypes of the Third World,12 it might be worthwhile to say something about how I proceed to develop a perspective on the state that is explicitly anti-orientalist. When notions of corrupt "underdevel­ oped" countries are combined with a developmentalist perspective, in which "state-society relations" in the Third World are seen as reflecting a prior position in the development of the "advanced" industrial nations, the temptation to compare "them" to "our own past" proves irresistible to many Western scholars.13 Instead, one needs to ask how one can use the comparative study of Third World political formations to confront the "naturalness" of concepts that have arisen from the historical experience and cultural context of the West. Focusing on the discursive construction of states and social groups allows one to see that the legacy of Western scholarship on the state has been to universalize a particular cultural construction of "state-society relations" in which specific notions of "statehood" and "civil society" are conjoined.14 Instead of building on these notions, this article asks if one can demonstrate their provincialism in the face of incommensurable cultural and historical contexts. 15 I begin with a series of vignettes that give a sense of the local level functioning of "the state" and the relationship that rural people have to state institutions. Everyday interactions with state bureaucracies are to my way of thinking the most important ingredient in constructions of "the state" forged by villagers and state officials. I then look at the broader field of representations of "the state" in public culture. Finally, I attempt to demonstrate how local level encounters with the state come together with representations in the mass media. This is followed by the conclusion, which systematically draws out the larger theoretical issues raised in the article. encountering "the state" at the local level For the majority of Indian citizens, the most immediate context for encountering the state is provided by their relationships with government bureaucracies at the local level. In addition to being promulgated by the mass media, representations of the state are effected through the public practices of different government institutions and agents. In Mandi, the administrative center closest to Alipur, the offices of the various government bureaucracies themselves served as sites where important information about the state was exchanged and opinions about policies or officials forged. Typically, large numbers of people clustered in small groups on the grounds of the local courts, the district magistrate's office, the hospital, or the police station, animatedly discussing and debating the latest news. It was in places such as these, where villagers interacted with each other and with residents of the nearby towns, as much as in the mass media, that corruption was discussed and debated. Therefore, looking closely at these settings allows us to obtain a sense of the texture ofrelations between state officials and clients at the local level. In this section I draw on three cases that together present a range of relationships between state officials and rural peoples. The first concerns a pair of state officials, occupying lowly but important rungs in the bureaucratic hierarchy, who successfully exploit the inexperience of two rural men. The second case concerns a lower-caste man's partially successful actions to protect himself from the threats of a powerful headman 16 who has allies in the bureaucracy by appealing to a higher official. The third example draws on a series of actions conducted by the powerful Bharatiya Kisan Union (literally, Indian Peasant Union), a grassroots farmers' movement that often strikes terror in the hearts of local state officials. Because they give a concrete shape and form to what would otherwise be an abstraction ("the state"), these everyday encounters provide one of the critical components through which the state comes to be constructed. Small but prosperous, Mandi 17 houses the lowest ends of the enormous state and federal · bureaucracy. 1 8 Most of the important officials of the district, including those whose offices are 378 american ethnologist in Mandi, prefer to live in another, bigger town that serves as the district headquarters. Part of the reason is that rental accommodation is hard to come by in Mandi (as I discovered to my frustration); equally important, it enables them to stay in closer touch with their superior officers. Sharmaji was a patwari, an official who keeps the land records of approximately five to six villages, or about five thousand plots, lying on the outskirts of Mandi. The patwari is responsible for registering land records, for physically measuring land areas to enter them in the records, and for evaluating the quality of land. The patwari also keeps a record of deaths in a family in the event of a dispute among the heirs about property, or the need to divide it up at some point. There are a number of officials above the patwari whose main-if not sole-duty is to deal with land records. On average, the total comes to about two officials for each village. Astonishing as this kind of bureaucratic sprawl might appear, it must not be forgotten that land is the principal means of production in this setting. Sharmaji lived in a small, inconspicuous house deep in the old part of town. Although I was confused at first, I eventually identified which turns in the narrow, winding lanes would lead me there. The lower part of the house consisted of two rooms and a small enclosed courtyard. One of those rooms had a large door that opened onto the street. This room functioned as Sharmaji's "office." That is where he was usually to be found, surrounded by clients, sycophants, and colleagues. Two men in particular were almost always by his side. One of them, Verma, himself a patwari of Sharmaji's natal vii lage (and therefore a colleague) was clearly in an inferior position. He functioned as Sharmaji's alter ego, filling in his ledgers for him, sometimes acting as a front and sometimes as a mediator in complex negotiations over how much money it would take to "get a job done," and generally behaving as a confidant and consultant who helped Sharmaji identify the best strategy for circumventing the administrative and legal qmstraints on the transfer of land titles. The other person worked as a full-time Man Friday who did various odd jobs and chores for Sharmaji's "official" tasks as well as for his household. Two of the side walls of the office were lined with benches; facing the entrance toward the inner part of the room was a raised platform, barely big enough for three people. It was here that Sharmaji sat and held court, 19 and it was here that he kept the land registers for the villages that he administered. All those who had business to conduct came to this "office." At any given time there were usually two or three different groups, interested in different transactions, assembled in the tiny room. Sharmaji conversed with all of them at the same time, often switching from one addressee to another in the middle of a single sentence. Everyone present joined in the discussion of matters p rtaining to others. Sharmaji often punctuated his statements by turning to the others and rhetofically asking, "Have I said anything wrong?" or, "Is what I have said true or not?" Most of the transactions conducted in this "office" were relatively straightforward: adding or deleting a name on a land title; dividing up a plot among brothers; settling a fight over disputed farmland. Since plots were separated from each other by small embankments made by farmers themselves and not by fences or other physical barriers, one established a claim to a piece of land by plowing it. Farmers with predatory intentions slowly started plowing just a few inches beyond their boundary each season so that in a short while they could effectively capture a few feet of their neighbors' territory. If a neighbor wanted to fight back and reclaim his land, he went to the patwari who settled the dispute by physically measuring the area with a tape measure. Of course, these things "cost money," but in most cases the "rates" were well-known and fixed. But however open the process of giving bribes and however public the transaction, there was nevertheless a performative aspect that had to be mastered. I will illustrate this with a story of a botched bribe. One day, when I reached Sharmaj i's house in the middle of the afternoon, two young men whose village fell in the jurisdiction of Verma were attempting to add a name to the title of their plot. They were sitting on the near left on one of the side benches. Both were blurred boundaries 379 probably in their late teens. Their rubber slippers and unkempt hair clearly marked them to be villagers, an impression reinforced by clothes that had obviously not been stitched by a tailor who normally catered to the "smart" set of town-dwelling young men. They appeared ill at ease and somewhat nervous in Sharmaji's room, an impression they tried hard to dispel by adopting an overconfident tone in their conversation. Although I never did find out why they wanted to add a name to the land records, I was told that it was in connection with their efforts to obtain fertilizer on a loan for which the land was to serve as collateral. When I arrived on the scene, negotiations seemed to have broken down already: the men had decided that they were not going to rely on Verma's help in getting the paperwork through the various branches of the bureaucracy but would instead do it themselves. Sharmaji and the others present (some of whom were farmers anxious to get their own work done) first convinced the young men that they would never be able to do it themselves. This was accomplished by aggressively telling them to go ahead and first try to get the job done on their own and that, if all else failed, they could always come back to Sharmaji. "If you don't succeed, I will always be willing to help you," he said. Thereupon one of the farmers present told the young men that Sharmaji was a very well-connected person. Without appearing to brag, Sharmaji himself said that when big farmers and important leaders needed to get their work done, it was to him that they came. Perhaps because they had been previously unaware of his reputation, the nervous clients seemed to lose all their bravado. They soon started begging for help, saying "Tau [father's elder brother], you know what's best, why should we go running around when you are here?" Sharmaji then requested Verma to "help" the young men. "Help them get their work done," he kept urging, to which Verma would reply, "I never refused to help them." The two patwaris then went into an adjoining room, where they had a short whispered conference. Sharmaji reappeared and announced loudly that they would have to "pay for it." The young men immediately wanted to know how much would be required, to which Sharmaji responded, "You should ask him [Verma] that." Shortly thereafter, Verma made a perfectly timed reen­ trance. The young men repeated the question to him. He said, "Give as much as you like." When they asked the question again, he said, "It is not for me to say. Give whatever amount you want to give." The two clients then whispered to each other. Finally, one of them broke the impasse by reaching into his shirt pocket and carefully taking out a few folded bills. He handed Rs. 10 to Verma.20 Sharmaji responded by bursting into raucous laughter and Verma smiled. Sharmaji told him, "You were right," laughing all the while. Verma said to the young men, "I'll be happy to do your work even for Rs. 10, but first you'll need the signature of the headman of your village, that's the law." Sharmaji told them that they didn't know anything about the law, that it took more than Rs. 14 just for the cost of the application because in order to add a name to a plot, the application would have to be backdated by a few months. At the mention of the headman, the young men became dismayed. They explained that relations were not good between them and the headman and that they were in opposite camps. I sensed that Verma had known this all along. Sharmaji then told the young men that they should have first found out "what it cost" to "get a name added to the register" these days. "Go and find out the cost of putting your name in the land register," he told them, "and then give Verma exactly half of that." He immediately turned to one of the farmers present and asked him how much he had paid ten years ago. The man said it had been something like Rs. 150. Then both Sharinaji and Verma got up abruptly and left for lunch. The young men turned to the other people and asked them if they knew what the appropriate sum was. All of them gave figures ranging from Rs. 130-150 but said that their information was dated because that is how much it had cost ten or more years ago. The young men tried to put 380 american ethnologist a good face on the bungled negotiation by suggesting that it would not be a big loss if they did not succeed in their efforts. If they did not get the loan, they would continue to farm as they usually did-that is, without fertilizer. No one could tell them what the current figure was. Even Man Friday, who was still sitting there, refused to answer, saying it was not for him to intervene, and that it was all up to Sharmaji and Verma. The "practice" of bribe giving was not, as the young men learned, simply an economic transaction but a cultural practice that required a great degree of performative competence. When vii lagers complained about the corruption of state officials, therefore, they were not just voicing their exclusion from government services because these were costly, although that was no small factor. More importantly, they were expressing frustration because they lacked the cultural capital required to negotiate deftly for those services.21 The entire episode was skiIlfully managed by Sharmaji and Verma. Although they came away empty-handed from this particular round of negotiations, they knew that the young men would eventually be back and would then have to pay even more than the going rate to get the same job done. Sharmaji appeared in turns as the benefactor and the supplicant pleading with his colleague on behalf of the clients. Verma managed to appear to be willing to do the work. The act of giving the bribe became entirely a gesture of goodwill on the part of the customers rather than a conscious mechanism to grease the wheels. Interestingly, a great deal of importance was attached to not naming a sum. In this case, state officials got the better of a couple of inexperienced clients. Petty officials, however, do not always have their way. In the implementation of development programs, for example, local officials often have to seek out beneficiaries in order to meet targets set by higher authorities. The beneficiaries of these programs can then employ the authority of the upper levels of the bureaucracy to exert some pressure on local officials. Several houses have been constructed in Alipur under two government programs, the Indira Awaas Yojana and the Nirbal Varg Awaas Yojana (literally, the Indira Housing Program and the Weaker Sections Housing Program, respectively). Both programs are intended to benefit poor people who do not have a brick (pucca) house. The Indira Awaas Yojana was meant for landless harijans (untouchables), whereas the Nirbal Varg Awaas Yojana was for all those who owned less than one acre of land, lacked a brick house, and had an income below a specified limit. 22 I was told that one of the "beneficiaries" was Sripal, so I spoke to him outside his new house. Sripal was a thin, small-boned man, not more than 25 years old, who lived in a cluster of low-caste (jatav) homes in the village. When I saw the brick one-room dwelling constructed next to his mother's house, I could not help remarking that it looked quite solid. But Sripal immediately dismissed that notion. Sripal was selected for this program by the village headman, Sher Singh. When his name was approved, the village development worker 23 took him to the town, had his photograph taken, and then opened an account in his name in a bank. For the paperwork he was charged Rs. 200. After that he was given a slip (parch,) that entitled him to pick up predetermined quantities of building material from a store designated by the village development worker. The money required to get the material transported to the construction site came out of his own pocket. The village development worker asked him to pay an additional Rs. 500 to get the bricks. Sripal pleaded that he did not have any money. "Take Rs. 1,000 if you want from the cost of the material [from the portion of the house grant reserved for purchasing materials], but don't ask me to pay you anything." Sripal claimed that this was exactly what the vi IIage development worker had done, providing him with material worth only Rs. 6,000 out of the Rs. 7,000 allocated to him.24 Once again he had to fork out the transportation expense to have the bricks delivered from a kiln near the village. Sripal claimed that the bricks given to him were inferior yellow bricks (peelay eenth) that had been improperly baked. He also discovered that the cost of labor was supposed to be blurred boundaries 381  reimbursed to him. Although he had built the house himself because he was an expert mason, he never received the Rs. 300 allocated for labor costs in the program. As if this were not enough, Sripal did not receive any material for a door and a window, so it was impossible to live in the new house. No official had come to inspect the work to see if there was anything missing. Sripal complained that those whose job it was to inspect the buildings just sat in their offices and approved the construction because they were the ones who had the authority to create the official record ("They are the ones who have pen and paper [kaagaz-kalam unhee kay paas hail"). Sripal himself is illiterate. Frustrated about his doorless house, he lodged complaints at the Block office and at the bank that lent him the money for construction. Meanwhile, Sher Singh, who had been employing Sripal as a daily laborer on his farm, became angry at Sripal for refusing to come to work one day. Sripal explained that he could not possibly have gone because his relatives had come over that day and that to leave them would have been construed as inhospitable. In any case, Sripal said, he could not do any heavy work because he had broken his arm some time ago. When Sher Singh found out that Sripal had complained about him and the village develop­ ment worker at the Block office, he threatened to beat him up so badly that he would never enter the village again. Fearing the worst, Sripal fled from the village and went to live with his in-laws. Despite the threat to his life, Sripal was not daunted in his efforts to seek justice. When he saw that his complaints elicited no response, he approached a lawyer to draft a letter to the District Magistrate, the highest administrative authority in the area. This strategy paid off in that a police contingent was sent to the village to investigate. When I asked Sripal to tell me what the letter said, he produced a copy of it for me. "What can I tell you?" he asked. "Read it yourself." The letter alleged that the village development worker had failed to supply the necessary material and that because the headman had threatened to beat him up he had been forced to flee the vi IIage. After the police visit, Sher Singh made peace with Sripal. He even hired Sripal to construct a home for another person under the same program. In addition, Sher Singh stopped asking Sripal to come to.labor on his farm. But the village development worker threatened Sripal with imprisonment unless he paid back Rs. 3,000 toward the cost of completing the house.25 "One of my relatives is a jail warden [thanedaarj," he reportedly told Sripal. "If you don't pay up, I'll have you put away in jail." Sitting in front of the empty space that was to be the door to his house, Sripal told me that he was resigned to going to jail. "What difference does it make?" he asked. "Living like this is as good as being dead." Even though he was ultimately unsuccessful in his appeals for justice, Sripal's case demon-. strates that even members of the subaltern classes have a practical knowledge of the multiple levels of state authority. Faced with the depredations of the headman and vii I age development worker, Sripal had appealed to the authority of a person three rungs higher in the bureaucratic hierarchy. Because the central and state governments are theoretically committed to protecting scheduled caste people such as Sripal, his complaint regarding the threat to his life was taken quite seriously. Sending the police to the village was a clear warning to Sher Singh that if he dared to harm Sripal physically, he wou Id risk retaliation from the repressive arm of the state. Before leaving this episode with Sripal, I want to address explicitly what it tells us about transnational linkages. Clearly, one cannot expect to find visible transnational dimensions to every grassroots encounter; that would require a kind of immediate determination that is empirically untrue and analytically indefensible. For example, IMF conditionalities do not directly explain this particular episode in the house-building program. But by forcing the Indian government to curtail domestic expenditure, the conditionalities do have budgetary implica­ tions for such programs. These influence which programs are funded, how they are imple­ mented and at what levels, who is targeted, and for how many years such programs continue. Similarly, if one wants to understand why development programs such as building houses for 382 american ethnologist the poor exist in the first place and why they are initiated and managed by the state, one must place them in the context of a regime of "development" that came into being in the postwar international order of decolonized nation-states (Escobar 1984, 1988; Ferguson 1990). What happens at the grassroots is thus complexly mediated, sometimes through multiple relays, sometimes more directly, by such linkages.26 Sripal's experience of pitting one organization of the state against others and of employing the multiple layers of state organizations to his advantage no doubt shaped his construction of the state. At the same time, he appeared defeated in the end by the procedures of a bureaucracy whose rules he could not comprehend. Sripal was among those beneficiaries of "development" assistance who regretted ever accepting help. He became deeply alienated by the very programs that the state employed to legitimate its rule. The implementation of development programs therefore forms a key arena where representations of the state are constituted and where its legitimacy is contested. One can also find contrasting instances where local officials are on the receiving end of villagers' disaffection with state institutions. Some examples are provided by several actions of the Bharatiya Kisan Union (BKU). One of the most frequent complaints of farmers is that they have to pay bribes to officials of the Hydel Department to replace burned-out transformers. Each such transformer typically serves five to ten tubewells. A young farmer related a common incident to me. The transformer supplying electricity to his tubewell and those of 11 of his neighbors blew out. So they contributed Rs. 150 each (approximately $10 at exchange rates prevailing then) and took the money to the assistant engineer of the Hydel Department. They told him that their crops were dying for a lack of water and that they were in deep trouble. He reportedly said, "What can I do? We don't have the replacement equipment at the present time." So they gave him the Rs. 1,800 they had pooled and requested that the transformer be replaced as soon as possible. He took the money and promised them that the job would be done in a few days, as soon as the equipment was in. Being an "honest" man (that is, one true to his word), he had the transformer installed three days later. When the same situation recurred shortly thereafter, the young man went to the Kisan Union people and requested that they help him get a new transformer. So about 50 of them climbed on tractors, went straight to the executive engineer's house and camped on his lawn (a common form of civil disobedience in India is to gherao [encircle and prevent movement of) a high official). They refused to move until a new transformer had been installed in the village. The executive engineer promised them that he "would send men at once." Sure enough, the I inemen came the following day and replaced it. Not all such incidents ended amicably. The quick response of these officials was due to the fact that the Kisan Union had already established itself as a powerful force in that particular area, as will be evident from a few examples. In one incident, a crowd walked off with six transformers from an electricity station in broad daylight (Aaj 1989f). The farmers no longer feared the police and revenue officials, on occasion "arresting" the officials, tying them to trees, and making them do "sit-ups." They refused to pay electricity dues (up to 60 percent of agricultural sector dues remain unpaid in a nearby district) and forced "corrupt" officials to return money allegedly taken as bribes. I also heard about an incident in an adjacent village where employees of the electricity board were caught stealing some copper wire from a transformer by irate villagers who proceeded to beat them up and "jail" them in a village house. It should be clear from al I the incidents described above that lower-level officials play a crucial role in citizens' encounters with "the state." Obviously, no singular characterization of the nature and content of the interaction of villagers and bureaucrats is possible. In contrast to Sharmaji and Verma, who manipulate their gullible clients, stand the officials who are manhandled by the peasant activists of the BKU. Similarly, just as local officials employ their familiarity with bureaucratic procedures to carry out or obstruct a transaction by maneuvering blurred boundaries 383 between different levels of the administrative hierarchy, so too do subaltern people such as Sripal demonstrate a practical competence in using the hierarchical nature of state institutions to their own ends. At the local level it becomes difficult to experience the state as an ontically coherent entity: what one confronts instead is much more discrete and fragmentary-land records officials, vii lage development workers, the Electricity Board, headmen, the police, and the Block Development Office. Yet (and it is this seemingly contradictory fact that we must always keep in mind) it is precisely through the practices of such local institutions that a translocal institution such as the state comes to be imagined. The local-level encounters with the state described in this section help us discern another significant point. Officials such as Sharmaji, who may very well constitute a majority of state employees occupying positions at the bottom of the bureaucratic pyramid, pose an interesting challenge to Western notions of the boundary between "state" and "society" in some obvious ways. The Western historical experience has been built on states that put people in locations distinct from their homes-in offices, cantonments, and courts-to mark their "rationalized" activity as office holders in a bureaucratic apparatus. People such as Sharmaji collapse this distinction not only between their roles as public servants and as private citizens at the site of their activity, but also in their styles of operation.27 Almost all other similarly placed officials in different branches of the state operate in an analogous manner. One has a better chance of finding them at the roadside tea stalls and in their homes than in their offices. Whereas modernization theorists would invariably interpret this as further evidence of the failure of efficient institutions to take root in a Third World context, one might just as easily turn the question around and inquire into the theoretical adequacy (and judgmental character) of the concepts through which such actions are described. In other words, if officials like Sharmaji and the village development worker are seen as thoroughly blurring the boundaries between "state" and "civil society," it is perhaps because those categories are descriptively inadequate to the I ived realities that they purport to represent. Finally, it may be useful to draw out the implications of the ethnographic material presented in this section for what it tells us about corruption and the implementation of policy. First, the people described here-Sharmaji, the village development worker, the Electricity Board officials-are not unusual or exceptional in the manner in which they conduct their official duties, in their wiIIingness to take bribes, for example, or in their conduct toward different classes of villagers. Second, despite the fact that lower-level officials' earnings from bribes are substantial, it is important to locate them in a larger "system" of corruption in which their superior officers are firmly implicated. In fact, Sharmaji's bosses depend on his considerable ability to maneuver land records for their own transactions, which are several orders of magnitude larger than his. His is a "volume business," theirs a "high margin" one. He helps them satisfy their clients and, in the process, buys protection and insurance for his own activities. This latter aspect calls for elaboration. It is often claimed that even well-designed government programs fail in their implementation, and that the best of plans founder due to widespread corruption at the lower levels of the bureaucracy. If this is intended to explain why government programs fail, it is patently inaccurate (as well as being class-biased). For it is clear that lower-level officials are only one link in a chain of corrupt practices that extends to the apex of state organizations and reaches far beyond them to electoral politics (Wade 1982, 1984, 1985). Politicians raise funds through senior bureaucrats for electoral purposes, senior bureaucrats squeeze this money from their subordinates as well as directly from projects that they oversee, and subordinates follow suit. The difference is that whereas higher-level state officials raise large sums from the relatively few people who can afford to pay it to them, lower-level officials collect it in small figures and on a daily basis from a very large number of people. It is for this reason that corruption is so much more visible at the lower levels. 384 american ethnologist  The "system" of corruption is of course not just a brute collection of practices whose most widespread execution occurs at the local level. It is also a discursive field that enables the phenomenon to be labeled, discussed, practiced, decried, and denounced. The next section is devoted to the analysis of the discourse of corruption, and especially to its historically and regionally situated character. blurred boundaries 385  the imagined state Banwari, a scheduled caste resident of Ashanwad hamlet, 25 kms. from Jaipur said, "I haven't seen the vidhan sabha or the Lok Sabha.51 The only part of the government I see is the police station four kms. from my house. And that is corrupt. The police demand bribes and don't register complaints of scheduled caste people like me." [Times of India 1989:7] So far, this article has dealt with the practices of local levels of the bureaucracy and the discourses of corruption in public culture, respectively. Together, they enable a certain construction of the state that meshes the imagined translocal institution with its localized blurred boundaries 389 embodiments. The government, in other words, is being constructed here in the imagination and everyday practices of ordinary people. Of course, this is exactly what "corporate culture" and nationalism do: they make possible and then naturalize the construction of such nonlo­ calizable institutions. It then becomes very important to understand the mechanisms, or modalities, that make it possible to imagine the state. What is the process whereby the "reality" of translocal entities comes to be experienced? To answer this question, one must grasp the pivotal role of public culture, which represents one of the most important modalities for the discursive construction of "the state." Obviously, not everyone imagines the state in quite the same manner. So far, very little research has been done on the relationship between diversely located groups of people and their employment of the different media of representation and of varying resources of cultural capital in imagining "the state." For example, Ram Singh and his sons are relatively prosperous men from one of the lowest castes (jatav) in Alipur. They had recently acquired a television set as part of the dowry received in the marriage of one of the sons. Ram Singh told me, in a confession born of a mixture of pride and embarrassment, that since the television had arrived their farm work had suffered because, instead of irrigating the crop, they would all sit down and watch television. (Both the pumpsets used for irrigation and the television set were dependent on erratic and occasional supplies of electricity.) Television was a constant point of reference in Ram Singh's conversation. I interviewed Ram Singh in the context of the impending elections (the elections took place in December 1989; the conversation dates from late July). He said: The public is singing the praises of Rajiv [Gandhi]. 52 He is paying really close attention to the needs of poor people [Bahut gaur kar raha hain]. Rajiv has been traveling extensively in the rural areas and personally finding outthe problems faced by the poor. For th is reason, lwill definitely support the Congress (I). We consider the government which supports us small people as if it were our mother and father [Usi ko ham maa-baap key samaan maantey hain]. If it weren't for the Congress, no one would pay any attention to the smaller castes [chotee jaat]. Not even god looks after us, only the Congress. At this point, his son intervened: The Congress is for all the poor, not just for the lower castes. It is exerting itself to the utmost, trying to draw people into [government] jobs [Bahut jor laga rahen hain, naukri mein khichai kar rahen hain]. Ram Singh returned to the discussion: Although the government has many good schemes, the officials in the middle eat it all [beech mey sab khaa jaate hain]. The government is making full efforts to help the poor, but the officials don't allow any of the schemes to reach the poor. "Doesn't the government knows that officials are corrupt?" I asked. "Why doesn't it do anything?" Ram Singh replied: It does know a little bit but not everything. The reason is that the voice of the poor doesn't reach people at the top [Garibon ki awaaz vahaan tak pahuchti nahin]. If, for example, the government sets aside four lakhs for a scheme, only one lakh will actually reach us-the rest will be taken out in the middle.53 Ram Singh's position here displays some continuity with an older, hierarchical vision of the state.54 Typically, in such views, the ruler appears as benevolent and charitable whereas the local official is seen as corrupt. While this may very well be the case, I think that one can adequately explain Ram Singh's outlook by examining contemporary practices rather than the sedimentation of beliefs.55 One should look at practices of the state that reinforce this outlook. When a complaint of corruption is lodged against a local official, the investigation is always conducted by an official of a higher rank. Higher officials are thus seen as providing redressals for grievances and punishing local officials for corrupt behavior. Ram Singh's case reminds us that all constructions of the state have to be situated with respect to the location of the speaker. Ram Singh's particular position helps us understand why he imagines the state as he does. He is an older, scheduled-caste man whose household now owns 390 american ethnologist one of the five television sets in the village, a key symbol of upward mobility. Several of his sons are educated, and two of them have obtained relatively good government jobs as a consequence.56 The scheduled castes of this area in general, and the jatavs in particular, have historically supported successive Congress regimes. The first thing that impresses one about Ram Singh's interpretation of "the state" is how clearly he understands its composition as an entity with multiple layers and diverse locales and centers. Although the word for regime and state is the same in Hindi (sarkaan, 57 Ram Singh maintains a distinction between the regime and the bureaucracy. He sees the regime's good intentions toward the lower castes being frustrated by venal state officials. Clearly, Ram Singh has a sense that there are several layers of "government" above the one that he has always dealt with (the very top personified by then-Prime Minister Rajiv Gandhi), and that the different levels can exert opposing pulls on policy (specifically, those that affect a scheduled-caste person like him). Interestingly, Ram Singh reproduces an apologetics for the failure of policy (the formulation is all right, it is the implementors that are to blame) pervasively found in India's "middle classes," delivered by politicians belonging to the regime in power, and reproduced in the work of academics, higher bureaucrats, and sympathetic officials of international agencies. The second striking fact about Ram Singh's testimony is that apart from his nuanced description of the state as a disaggregated and multilayered institution, his analysis closely parallels a discourse on the state that is disseminated by the mass media and is therefore translocal. Ram Singh's example demonstrates the importance of public culture in the discursive construction of the state: he talks knowledgeably about "the public's" perception of Raj iv and of Rajiv's itinerary. His son's perception of the Congress as being "for all the poor'' clearly also owes a great deal to mass-mediated sources. My suspicion that the close association with Rajiv Gandhi and the explanation about the corrupt middle levels of the state was influenced by the impact of television gained force when one of his sons explained: 58 We are illiterate people whose knowledge would be confined to the village. This way [i.e., by watching television], we learn a I ittle bit about the outside world, aboutthe different parts of India, about how other people live, we get a little more worldly [Kuch duniyaadaari seekh /aayten hain]. 59 In the buildup to the elections, the government-controlled television network, Doordarshan, spent most of the nightly newscast following Rajiv Gandhi on his campaign tours. Obviously, it was not just the country that was being imagined on television through the representation of its different parts but also the national state through the image of "its" leader. Popular understandings of the state therefore are constituted in a discursive field where the mass media play a critical role. Ram Singh's words reveal the important part that national media play in "local" discourses on the state. Clearly, it is not possible to deduce Ram Singh's understanding of "the state" entirely from his personal interactions with the bureaucracy; conversely, it is apparent that he is not merely parroting the reports he obtains from television and newspapers.6() Rather, what we see from this example is the articulation between (necessarily fractured) hegemonic discourses and the inevitably situated and interested interpretations of subaltern subjects. Ram Singh's everyday experiences lead him to believe that there must be government officials and agencies (whose presence, motives, and actions are represented to him through the mass media) interested in helping people like him. Only that could explain why his sons have succeeded in obtaining highly prized government jobs despite their neglect by local schoolteachers and their ill-treatment by local officials. Yet when he talks about "the public," and with a first-person familiarity about Raj iv's efforts on behalf of the poor, he is clearly drawing on a mass-mediated knowledge of what that upper-level of government comprises, who the agents responsible for its actions are, and what kinds of policies and programs they are promoting.61 blurred boundaries 391  There is obviously no Archimedean point from which to visualize "the state," only numerous situated knowledges (Haraway 1988). Bureaucrats, for example, imagine it through statistics (Hacking 1982), official reports, and tours, whereas citizens do so through newspaper stories, dealings with particular government agencies, the pronouncements of politicians, and so forth. Constructions of the state clearly vary according to the manner in which different actors are positioned. It is therefore important to situate a certain symbolic construction of the state with respect to the particular context in which it is realized. The importance of the mass media should not blind us to the differences that exist in the way that diversely situated people imagine the state.62 For instance, Ram Singh's position as a relatively well-to-do lower-caste person, whose family has benefited from rules regarding employment quotas for scheduled castes, explains his support for the higher echelons of government. At the same time, his interaction with local officials has taught him that they, like the powerful men in the villages, have little or no sympathy for lower-caste people like him. Therefore, he has a keen sense of the differences among different levels of the state. On the other hand, if he seems to share with the middle-class a particular view of the failure of government programs, it is the result of the convergence of what he has learned from his everyday encounters with the "state" with what he has discerned, as his son indicates, from the mass media. Congress rhetoric about being the party of the poor obviously resonates with Ram Singh's experience; that is why he calls the Congress government his guardians (maa-baap) and blames the officials in the middle for not following through with government programs. Ram Singh's view of the state thus is shaped both by his own encounters with local officials and by the translocal imagining of the state made possible by viewing television. 392 american ethnologist