Colonial rule and the contradictions of post-colonial democracy

In an essay on the post-colonial state in India, Sudipta Kaviraj argues that the prefix ‘post-colonial’ indicates ‘not the trivial fact that this state emerged after the colonial regime departed, but … that some of its characteristic features could not have arisen without the particular colonial history that went before’.Footnote ⁵ He further asserts that ‘[t]he entire story of the state for the half-century after Independence can be seen in terms of two apparently contradictory trends … that can be schematically regarded as the logic of bureaucracy and the logic of democracy’, and that ‘[t]he antecedents of both these trends can be found in the history of colonial rule’.Footnote ⁶ This conflict between bureaucracy and democracy is precisely what came to mark the rule of the RJD in Bihar (discussed later) and what accompanied, more generally, the political ascendancy of lower and backward caste parties in North India.

While simultaneously continuing to be an instrument of repression, the colonial state in British India underwent some degree of democratization when representative institutions began to be slowly introduced from the early twentieth century onwards, spurred by a series of constitutional shifts that eventually led to a transfer of power to Indians in 1947.Footnote ⁷ The pressure exerted on the colonial administration by Indian nationalist elites proved instrumental in triggering this chain of events, with these elites being aided in no small part by the contradictions inherent in Britain's ‘liberal imperialism’. As Kaviraj puts it:

In the last election conducted in British India, the electorate had comprised just 14 per cent of the adult population. As such, when the Constitution adopted by the new Indian republic embraced universal adult suffrage, the extension of representation that resulted was radical and dramatic. In a country where the vast majority of people were illiterate, this move, moreover, had the potential to engender conflict between the masses and the elites ensconced within representative institutions, who had very little in common with ordinary voters. As it happened, North India did not witness any such conflict in the early post-Independence period, partly on account of the long-standing habits of deference that a caste society had bred among subaltern groups.Footnote ⁹ It was not until the 1970s that backward castes began to mobilize against upper caste dominance, and only in the 1990s that lower and backward caste parties came to form governments on a regular basis in UP and Bihar.

The state that emerged following the attainment of Independence in 1947, however, bore the imprint not only of the Indian nationalist movement but also of the colonial state that preceded it; it was precisely in this sense that its inheritance was contradictory. The coercive apparatuses of the colonial state—the police and the bureaucracy—survived more or less intact in post-Independence India, despite the Congress (which led the nationalist agitation and formed independent India's first government) having promised to reform them. This was mainly on account of the bloodshed and mayhem that accompanied the partition of British India and the birth of the new states of India and Pakistan. In such a political climate, clamping down on any centrifugal tendencies that could further emerge, and zealously protecting what was, at the time, a fragile unity, became a matter of utmost priority for the Congress leadership. This helped to buttress the logic for a strong and centralized state with considerable repressive capacity. Speaking of the bureaucracy specifically, however, it was the rationale of development more than any other that justified its remarkable expansion in the first few decades of the post-Independence period.Footnote ¹⁰

Jawaharlal Nehru, India's first prime minister, and chief architect of the developmentalist ideology of the early years, firmly wedded development to redistributive concerns and identified the state as the best instrument for realizing this vision. State-managed development led to a massive extension of the bureaucracy and the public sector, which came to command vast economic resources in Nehru's time. However, the public sector enjoyed a relative decisional and managerial autonomy, and remained more or less insulated from the political process in this phase. Its influence, furthermore, was justified with reference to the arguments of distributive justice. ‘If the state managed heavy industries, the argument went, existing inequalities of income would not increase; and it would also act against the concentration of resources in a few private hands.’Footnote ¹¹

The insulation from politics was, however, to wane rapidly following Nehru's demise in 1964. Under Indira Gandhi, Nehru's daughter and successor, who took over the leadership of the Congress and the central government, ministers started interfering in the operation of public enterprises, which made for unproductive activity and inefficiency. Anxiety over inefficiency made managements dependent for their survival on political support; for providing this, the political class cynically extracted a price, namely, access to development funds, which they channelled for political ends.Footnote ¹² In other instances, the spoils were divided between politicians and bureaucrats, with the latter being rewarded for facilitating illicit transfers. Over time, this quid pro quo that developed between bureaucrats in charge of the developmental state and elected representatives acquired a persistent and systemic character. Thus, ‘the developmental state increasingly had little to do with realistic redistributive objectives, but became utterly dependent on a disingenuous use of that rhetoric’,Footnote ¹³ which was invoked to spirit away public money into the pockets of state employees and their clients outside the state.

While economic liberalization provided some impetus for the dismantling of the development bureaucracy from the 1990s onwards, the restructuring of the state proceeded slowly and unevenly across sectors and regions in India. Thus, Akhil Gupta and Aradhana Sharma argue that liberalization had very little impact on sectors other than industry, finance, and high-end services. The small-scale sector, which employs a large number of workers, was hardly touched, and governments, without exception, continued to adopt welfare policies that targeted the agricultural sector and the rural poor. That liberalization did not bring about a sharp disruption in policy, particularly from the point of view of welfare, was not very surprising in a country where vast numbers of the electorate remained poor and were likely to vote out governments that accelerated reforms to an extent where they hurt those at the bottom of the pile. In a nutshell, then, the development bureaucracy continued to be relevant in some politically significant sectors of the economy and quite powerful in the North Indian states which, being poorer and more rural than those of the south, provided opportunities for it to loom much larger.Footnote ¹⁴

It will be clear, then, how the logic of bureaucracy came to collide with the logic of democracy; the size, ethos, and culture of the development administration all militated against its serving ordinary citizens well. But there was one other factor—the sociology of the administration, whose contours took shape in the period of colonial rule—that facilitated this collision further. Much has been written about the efforts of the colonial state in British India to enumerate and classify colonial subjects. At the heart of the curiosity driving these efforts lay the institution of caste, identified early on as the singular marker of this society's ‘otherness’ and Oriental character.Footnote ¹⁵ Logistical considerations (that is, the need for enumerators to be familiar with local social structures and languages) meant that the enumerative exercise could proceed only with assistance from Indian officials; what was remarkable, however, was the colonial administration's specific reliance on Brahmans, who presided atop the caste hierarchy.

The reliance on Brahmans could, in part, be explained by the bias prevailing among some high-ranking British officials who privileged a Brahmanical mapping of the Indian social world. Foremost among them was H. H. Risley, census commissioner of India for the 1901 Census, who was also appointed director of ethnography in 1901.Footnote ¹⁶ As Nicholas Dirks argues, ‘In retrospect, Risley's reliance on a Brahmanical sociology of knowledge is astounding … [H]e organized his entire understanding of caste structure and rank according to Brahmanical indices … [and came to rely on] the clerical Brahmanical opinion that permeated the middle echelons of colonial administration in the localities’.Footnote ¹⁷ But it was not just the worldview of officials like Risley that accounted for the preponderance of Brahmans and other upper castes within the colonial bureaucracy. A broader, structural logic was also at work. Brahmans were priests and had access to scriptures, and were the most educated of all caste groups at the time of the advent of the British. This advantage allowed for an early mastery of the English language when it was introduced by the colonial administration, which facilitated an easy socialization into the values of colonial rule. Thus, Brahmans were recruited to clerical and administrative positions in the colonial state in large numbers. As Donald L. Horowitz puts it, it was ‘the felt necessity of the Europeans to govern their vast territories with the aid of local people they found to be particularly capable. Ruling as they did with little manpower, all the colonial powers from time to time designated members of particular groups as their agents.’Footnote ¹⁸ This policy had severe repercussions. It bred resentment among the groups left out and, as the privileged community used its advantage of occupation to steadily accumulate socio-economic gains over time, an ‘advanced-backward’ dichotomy crystallized, which spilled over from the colonial phase into the post-colonial period, often triggering an aggressive politics of ‘catching up’ on the part of those groups that saw themselves as ‘backward’.Footnote ¹⁹

It was not merely the colonial, but also the post-colonial, state that saw upper castes entrenched in important administrative positions. Studies of the Indian Administrative Service (IAS),Footnote ²⁰ which replaced the Indian Civil Service of British India, and from which post-Independence India draws its top-ranking bureaucrats, amply demonstrate this upper caste predominance. As per one study, which obtained data pertaining to 3,675 of the 4,284 officers who made up the IAS on 1 January 1985, 2,205 (60 per cent) of these officers belonged to the upper castes.Footnote ²¹ There were 1,218 Brahman officers alone, accounting for 33 per cent of the sample; the next largest upper caste group was the Kayasths, who numbered 431, accounting for 12 per cent.Footnote ²² The study further found that the affirmative action provisions of the Constitution, which granted Scheduled Castes (SCs or the formerly Untouchable castes) 15 per cent reservations in public sector jobs, had fallen far short of serious implementation in the IAS. ‘[T]hrough the early 1970s, vacancies were being carried over from one year to the next, that is, prescribed quotas were not being met’, with data for 1978–9 indicating ‘that only one-third of the posts reserved for Scheduled Castes … were being filled annually’.Footnote ²³

Studies of the IAS cadres of UP and Bihar, the two North Indian states of concern here, show these cadres to be sociologically skewed in much the same way as the IAS overall, which is not surprising, as state-level cadres are drawn from the national pool. Thus, Jeffrey Witsoe, on examining the Bihar cadre of the IAS in 2002, writes that of the 224 officers about whom he had information (there were 244 IAS officers in Bihar at the time), 143 (64 per cent) belonged to the upper castes. There were only 25 SC and 16 Other Backward Class (OBC) officers, making up 11 and 7 per cent of the pool respectively.Footnote ²⁴ Likewise, Sebastiaan Maria van Gool, in his study of the bureaucracy in UP, found that of the 541 IAS officers making up the UP cadre in 1999, as many as 353 (65 per cent) came from upper caste backgrounds.Footnote ²⁵

State-level administrations in India do not consist only of IAS personnel; several posts therein are occupied by Provincial Civil Service (PCS) officials, who are recruited and employed by their respective state governments.Footnote ²⁶ Nevertheless, IAS officers do end up manning the most important positions in India's civil bureaucracy. Available evidence further suggests that in UP, the top tiers of the PCS tend to be upper caste-heavy as well.Footnote ²⁷ So, given this predominance of upper castes in the bureaucracy overall, it is not surprising that the development administration in the post-Independence years functioned to the dictates of a distinct caste bias. In North India, this bias remained unchallenged all through the time that the Congress remained electorally dominant. This electoral dominance translated into the political dominance of landed, upper caste elites, who made up the party's leadership in the North Indian states. With upper castes commanding influence in both the political and administrative domains, collusion between upper caste representatives and bureaucrats followed, and development funds came to be siphoned off and steered to co-caste clients located outside the state.

The lower and backward caste parties that captured power in UP and Bihar in the 1990s disrupted this long-standing harmony between the political and administrative spheres. The bureaucratic obstruction these parties faced while in government forced them to resort to extra-legal means to push through redistributive outcomes, in a sporadic and uneven way. This had immense implications for the tone and tenor of India's post-colonial democracy and its political culture. It is telling that it was precisely at the moment when political representation deepened significantly in North India after decades of elite dominance that a severe setback occurred on the procedural front. This was a fallout of the imbalance that prevailed between the bureaucratic and the representative arms of the state vis-à-vis the extent of democratization they had each undergone.

In the last phase of colonial rule, the colonial state had been marked by an imbalance of a somewhat parallel nature. The nationalist leadership, breaking the hold of race as a factor that determined eligibility to govern, had forced the colonial administration to allow Indians a say, albeit limited, in the matter of their own rule. But the bureaucratic apparatus of the colonial state had continued to be governed by an ethos that was much less egalitarian and which saw in caste status a reliable measure of worth. This tension, spilling over to the post-colonial state, ripened fully and came to a head in the 1990s in North India, when the politics of lower caste assertion and its electoral manifestations meant that it could be contained no further.

Lower and backward caste parties and the politics of ‘social justice’

The idea of ‘social justice’ is an integral component of the vocabulary of democratization in India. It is associated with an ethical-political project that took shape in the run-up to the achievement of Independence and in the course of drafting the Indian Constitution, several of whose provisions gave it tangible form. This project sees communities (rather than individuals) as units between which parity must be achieved and defines these communities in terms of caste. The initial thrust of the project was to address the disparities arising from historically accumulated injustices and from the practice of ‘Untouchability’, in particular. It was this rationale that underpinned the affirmative action provisions for SCs, or former Untouchables, that the Constitution laid down, and which came into effect at the founding moment of the republic.Footnote ²⁸ Much later, in 1990, the project of social justice broadened to address the wider problem of ‘backwardness’; social justice provisions now came to target a plethora of lower castes who were located above SCs in the caste hierarchy and who had, despite not enduring the practice of Untouchability, lagged behind—economically, socially, and educationally—and who came to be referred to as the ‘Other Backward Classes’ (OBCs) or, simply, ‘backward castes’. Unlike SCs, OBCs were not identified purely on the basis of caste criteria, but socio-economic criteria as well; however, it was well-recognized that caste served, in most cases, as an effective measure of backwardness.

It is worth noting that the OBC category has long had a constitutional life in India. Using the term ‘other backward classes’ to refer to these groups, the Constitution that came into force on 26 January 1950 recommended in Article 340 that the president of India appoint a commission that would both investigate their condition and propose measures to ameliorate their backwardness.Footnote ²⁹ However, it was only following the implementation of the recommendations of the second Backward Classes Commission (also known as the Mandal Commission) by V. P. Singh's National Front government in 1990 that the OBC category came alive politically and administratively nationwide.Footnote ³⁰ The Mandal Commission recommended that, over and above the 22.5 per cent reservations accorded to SCs and Scheduled Tribes (STs)Footnote ³¹ taken together (15 per cent to SCs and 7.5 per cent to STs), OBCs should be accorded an additional quota of 27 per cent.

The implementation of this recommendation marked a watershed moment in Indian politics. It made for a remarkable broadening of the scope of state intervention in favour of social justice and unleashed a dynamic that paved the way for the electoral dominance of lower and backward caste parties in the critical North Indian states of UP and Bihar.Footnote ³² In UP, this dominance lasted from the early 1990s up to 2017, whereas Bihar saw backward caste parties rule up to 2005 and dominate alliances with upper caste parties for a good ten years after that. This spell proved critical to the empowerment of subaltern constituencies in both North Indian states. By provoking vehement opposition from upper caste groups, the central government's implementation of the OBC quota set the ground for a counter-mobilization of all communities benefiting from reservations, leading to an unprecedented polarization around forward caste identities, on the one hand, and backward and lower caste identities, on the other.

Within the umbrella category of the OBC, there was considerable socio-economic differentiation and hierarchy, with some backward caste groups (Yadavs and Kurmis) oppressing others. These divisions were further exacerbated by the politics of the Samajwadi Party (SP) in UP and the RJD in Bihar. These parties, which came to power several times over the next two decades, while seeking to mobilize all OBCs, ended up privileging Yadavs. Nevertheless, coming before this phase, the moment of the early 1990s saw a consolidated and overarching OBC identity crystallize. Likewise, the fractures among SCs—particularly the divide between Chamars (who were the most economically advanced and politically conscious) and other lower castes—were subsumed within a solidarity that developed at this time and which bound together SCs in reaction to the backlash against quotas. With consolidated SC and OBC identities emerging, and both groups mobilizing against a common upper caste enemy, it became impossible for any party to hold together an alliance embracing backward and forward castes in the early 1990s. This severely affected the fortunes of the Congress, which had depended, right up to the 1980s, on a coalition of SCs, Muslims, and upper castes to win elections in UP and Bihar.Footnote ³³

In UP, the BJP now emerged as the representative of upper castes, the constituency threatened by reservations, while the SP and the BSP emerged as the champions of quotas, and as electoral platforms for OBCs and SCs respectively, with the Muslim vote divided between them. With its base taken away and split between these three parties, the Congress was now set for a decimation, which turned out to be so thorough that it has not once succeeded in governing since.Footnote ³⁴ The SP and the BSP emerged as the two principal ruling parties, though the BSP shared power with the BJP three times over the period 1995–2003, exercising considerable influence within the coalition. Indeed, the run of the social justice parties remained uninterrupted in UP until 2017, with the BJP capturing a decisive majority in the state assembly election held that year.

In Bihar, 1990 marked the beginning of 15 years of rule by a backward caste platform whose core base comprised Yadavs (a numerically and economically predominant backward caste), but which also attracted Muslims, and had at its helm a charismatic politician, Lalu Prasad Yadav. Lalu Yadav initially headed a Janata Dal (JD) government, but in 1997, in response to a challenge from senior party leaders, split the JD, forming his own party, the RJD, which ruled Bihar up to 2005.

Both UP and Bihar had seen backward caste parties challenge the dominance of the Congress and rule briefly in the 1960s and 1970s.Footnote ³⁵ But the backward caste politics of that phase failed to disrupt the networks of vertical mobilization deployed by the Congress, which returned to power soon after. It was precisely its vertical mobilization capabilities that had allowed the Congress to project itself as an umbrella party while catering to upper castes for the first four decades of the post-Independence period. It had integrated lower castes (and, initially, backward castes as well) into political structures dominated by upper castes by making clever use of the feudal relations still prevailing in agriculture, and of the material dependence of low caste groups (most of whom were landless agricultural labour) on the high caste ex-tenants of the zamindars and talukdars Footnote ³⁶—a landed class firmly affiliated to the party. By leveraging agrarian relations of domination and subordination in this way, the Congress was able to mobilize the low caste poor without offering them political representation or leadership benefits. As Paul R. Brass argues in his 1960s study of the UP Congress, the party's electoral support in the state was far wider than its leadership, with SCs being its most reliable constituency.Footnote ³⁷

The politics of vertical mobilization thus drew sustenance from a skewed distribution of material power in the countryside, and the disproportionate political influence this allowed landed upper castes to exercise. But in the 1960s and 1970s, economic power began shifting down the ladder of caste and class, and flowing into the ranks of the middle peasantry, most of whom were backward castes. Freed from the shackles that had tied them to the landowning strata from which the Congress drew its leadership, these newly empowered groups abandoned the party now. Land reforms and the Green Revolution worked in tandem to effect this democratization. In UP and Bihar, despite doing little for those at the bottom of the agrarian class structure, land reforms enabled a large class of former tenants to become peasant-proprietors. Since most tenants came from backward caste communities, this undermined the power of upper castes, who made up the majority of the old landowning class. The Green Revolution was associated with ‘the introduction of high-yielding seeds between 1965 and 1966’ and ‘the development of irrigation and the use of chemical fertilizers’. It thus ‘served the interests of those among the landowners who had some investment capacity’ and helped the relatively better-off sections of the middle peasantry achieve economic ascendancy in the countryside.Footnote ³⁸

For the landless SC poor though, it was the processes of industrialization and urbanization that—by bringing them alternative employment opportunities—allowed them to sever economic ties with landed groups and bypass their political control.Footnote ³⁹ But the critical factor driving the breakdown of vertical networks and the forging of horizontal solidarities, both for SCs and OBCs, was the implementation of the OBC quota in 1990. The upper caste hostility to reservations that erupted immediately afterwards helped forge a radical oppositional consciousness among subalterns that refused any truck with vote-bank politics. This consciousness was shaped in significant ways by lower and backward caste parties, for whom this moment presented a much-awaited opening. It would be pertinent to mention here a slogan coined by the early BSP: vote hamara, raj tumhara/nahi chalega, nahi chalega (our votes, your rule/this will not do, this will not do).Footnote ⁴⁰ This was a pithy variation of the speech—constructed around the trope of a ballpoint pen—that Kanshi Ram, the BSP's founder, was known to deliver often. Producing the pen from his pocket with a flourish, atop rally platforms and before massive crowds, Kanshi Ram would argue that the cap represented the metaphorical 15 per cent—the upper castes—who, despite being in a minority, ruled the country, while the body of the pen represented the remaining 85 per cent—the subaltern masses—who had yet to become aware of their numerical strength and what it could accomplish for them.Footnote ⁴¹ The mobilizational discourses that began circulating in the 1990s exhorted voters to support only those parties that offered them representation, that is, awarded tickets to their co-ethnics, and that actually served the interests of their caste groups, rather than only promising to do so. The point of reference here was the vote-bank politics of the Congress variety, which had been committed to exactly the reverse.

I end this section by discussing how the electoral ascendancy of lower and backward caste parties transformed the grammar of politics in North India and affected popular understandings of ‘power’ and ‘the state’. As Kanchan Chandra points out, these parties explicitly represented themselves as champions of the cause of particular ethnic groups to the exclusion of others, that is, they mobilized ethnic ‘insider’ groups by identifying and excluding ethnic ‘outsiders’.Footnote ⁴² This refusal to accommodate the ethnic ‘other’—specifically, the caste ‘other’—within one's party translated into a refusal to share the state with that ‘other’ as well when these parties captured power. The state thus came to be invested with a militant particularity; indeed, the whole point of capturing the state was to make it one's ‘own’ and stamp out its civic potential. This was not wholly different from how state power was exercised at the time of Congress rule. With upper castes then controlling the legislature, having long since established a grip over the bureaucracy, the state machinery was effectively under elite control. When Congress governments undertook to distribute welfare to the poor, it was with a view to putting a human face on a state committed to maintaining elite dominance over the long term. Even during the Congress's populist phase under Indira Gandhi, who bypassed local elites and established direct contact with the masses, and which saw welfare provisions for the rural poor increase, it was ‘dominant landowning castes [who] remained in control of the institutions charged with implementing these [new welfare] measures, and the poor received whatever development resources were not siphoned off on the way down’.Footnote ⁴³

But the image of itself that the Congress sought to project made its politics quite different from those of the caste-based parties. The Congress took care to portray itself as a catch-all party that catered to constituencies across caste, class, and religion. Critical to this portrayal was a discourse of neutrality; party leaders were careful never to be seen as catering to any one constituency at the cost of another.Footnote ⁴⁴ This discourse was at sharp variance with the Congress's actual functioning—its use of the strategy of vertical mobilization and discriminating between constituencies while in government. When the caste-based parties took the stage in the 1990s, they sought to supplant the bias that the Congress had injected into state institutions with their own. But in a marked departure from the Congress, these parties abandoned the pretence of neutrality; they were unapologetic about their claims to represent specific caste groups exclusively and their designs to deploy state power to damage the caste enemy. This change in the language of politics was remarkable and indicated a new normal. The divergence between discourse and practice that prevailed during the time of the Congress served, in a sense, to establish the discourse as the norm and the practice as deviation. Thus, neutrality and a civic ethos continued to be regarded as valuable, notwithstanding their betrayal in lived public life. But the caste-based parties, by vehemently disparaging these norms, legitimized the narrow and sectarian use of both power and the state to large sections of the public.

The shift in popular thinking that resulted left its mark on everyday political discourse. The SP and the BSP had alternated in power in UP over the period 1995–2017, which had fostered a rivalry between them. This rivalry spurred animosity between Yadavs, the backward caste that provided the SP with its core support, and Chamars, the SC community that formed the electoral base of the BSP. This animosity, apart from being driven by party competition, also had a material foundation. Chamars possessed no land and worked as agricultural labour on fields owned by upper castes (Brahmans and Thakurs) and OBCs (Yadavs and Kurmis, who owned the most land among backward groups). Thakurs and Yadavs, in particular, frequently harassed Chamars, and conflict erupted between them and Chamar labour over the issues of low wages, late payments, and molestation of Chamar women working on the fields. Much of this abuse was, however, curtailed by the BSP when it implemented the Scheduled Castes/Scheduled Tribes (Prevention of Atrocities) Act (in short, the SC/ST Act) upon coming to power. The Act, which made the harassment of SCs an offence punishable by law, had been passed by the parliament in 1989. But it had existed only on paper in UP up to the time Mayawati, the BSP chief minister, enforced its provisions—for the first time in 1997 and whenever the BSP formed the government subsequently.

The BSP's firm stance on the Act created a great deal of controversy, as it led to the imprisonment of Thakurs and Yadavs, who committed the most atrocities. However, there were also allegations that SCs were fabricating charges to harass other castes and benefiting from the financial compensation the Act provided to victims.Footnote ⁴⁵ Consequently, one of the first changes to follow from the SP's assuming power after a spell of BSP rule was that SCs no longer succeeded in lodging complaints at will at police stations across UP. With the state changing hands and not willing to ‘tolerate attempts to land Yadavs in trouble’ (‘Yadavon ko phansane ka kaushish bardasht nahi karenge’),Footnote ⁴⁶ SCs now found themselves at the receiving end—with charges of harassment being reversed, those earlier imprisoned being set free, and the latter keen to demonstrate, to Chamars in particular, ‘who was now boss’ (‘ab kaun malik hai’).Footnote ⁴⁷ The SP's lax enforcement of the SC/ST Act was, indeed, one reason why Thakurs preferred the SP to the BSP. It was also in the expectation that the BJP (which they took to be their party) would keep Mayawati from implementing the Act too zealously that Thakurs backed a BJP-BSP coalition in UP.Footnote ⁴⁸

There were several other instances where a decision taken by one government was immediately overturned by the other when the latter acceded to power. In each case, the decision in question was understood to have benefited a narrow constituency, at the cost of another, and the act of overturning served to injure the constituency earlier advantaged and benefit its enemy. There were two factors at work here. One was the decision of the party in power to align the state with its own supporters and make it work for them alone. The other was the public perception that this was good and right, and the expectation nurtured by each constituency that the state, when captured by the party that represented them, would be ‘theirs’—a particularistic instrument shorn of all civic purpose. In such a climate, it was entirely fitting for Chamars in Dehri, my main fieldwork village in Jaunpur, to taunt Thakurs with the question, ‘ab kya karoge?’ (‘What will you do now?’) when the BSP won a majority in the 2007 state assembly election.Footnote ⁴⁹

Reversing the upper caste bias of state institutions was, however, no easy task. In attempting to make the state serve subaltern groups, the biggest obstacles that lower caste parties faced were posed by the bureaucracy and, in the BSP's case, additionally by the BJP, its coalition partner. In the following two sections, with a focus on BSP rule in UP and JD/RJD rule in Bihar, I show how the resistance emanating from the bureaucratic-political nexus was sought to be countered through strategies that, while they succeeded in chipping away at elite privilege, jeopardized the procedural foundations of India's democracy.

Bureaucratic resistance and the bahubali politician

The BSP's formation of a government in UP was, in each instance, followed by a large-scale transfer of senior bureaucrats and police officers. Mayawati—the Chamar leader who heads the BSP and was chief minister all four times the party ruled the state—transferred 62 IAS officers and 105 Indian Police Service (IPS)Footnote ⁵⁰ officers during her first term as chief minister,Footnote ⁵¹ and 1,350 officers, drawn from across the IAS, the IPS, and the PCS, during her second.Footnote ⁵² She was widely understood to have undertaken these reshuffles, which placed SC officials in key bureaucratic posts, for symbolic reasons, in the expectation that it would make her SC base proud to have lower caste officers head prestigious departments. But interviews with BSP officials revealed that these administrative transfers were driven not merely by a symbolic logic, but also by a pragmatic one. Mayawati had correctly predicted that the policy efforts of her government were unlikely to succeed in the face of bureaucratic obstruction and had thus placed departments she considered key to her effective performance in the hands of lower caste officers, relying on transfers at the top to discipline the middle and lower rungs of the administration.Footnote ⁵³ Apart from carrying out actual transfers, she also resorted to the threat of transfers and made unannounced visits to the districts to tame a recalcitrant bureaucracy.Footnote ⁵⁴

That the administration in UP was indeed recalcitrant is borne out by the account, drawn from my fieldwork, that I provide here, which illustrates the attempts it made to sabotage government policy in three distinct spheres. The first pertained to the Ambedkar Village Programme (AVP),Footnote ⁵⁵ the BSP's flagship rural development effort that sought to develop infrastructure (build primary schools and link-roads connecting SC hamlets to main roads) and provide basic amenities (potable water, toilets, drains, and electricity) to villages with large SC populations, with the SC segments being the primary beneficiaries. The AVP met with fierce disapproval from upper castes, as it tended to shift power relations in the selected villages to their disadvantage. With the construction of toilets and hand-pumps in SC quarters, SCs no longer needed to fetch water from upper caste hamlets or defecate on upper caste fields, which reduced opportunities for harassment and humiliation. The building of schools in SC neighbourhoods, again, saved SC children from having to walk through upper caste localities and, still more significantly, allowed for voting booths to be located in SC areas (primary schools in rural UP often serve as voting booths during elections), which lowered chances of ‘booth capturing’Footnote ⁵⁶ by dominant strata. The construction of link-roads, again, would often require individuals, usually from landed backgrounds, to vacate plots they had forcibly occupied and were cultivating illegally, which led to standoffs between SCs and upper castes in the countryside.Footnote ⁵⁷

In order to resist the AVP successfully, upper castes needed to enlist the support of officials within the district administration. This they succeeded in doing quite often; Thakurs in Jaunpur district would frequently be warned by their co-caste bureaucratic contacts that their village was about to be selected for the coming fiscal year, with this advance knowledge helping them deflect the programme elsewhere. In a telling instance that illuminated the networks binding upper castes located inside the state to those outside, the district magistrateFootnote ⁵⁸ of Jaunpur, a Brahman by caste, when confronted by an irate Thakur who refused to vacate a plot to make way for a link-road, persuaded the man to cooperate, with the guarantee that he could resume cultivation elsewhere.Footnote ⁵⁹

The second BSP policy effort that the bureaucracy attempted to thwart related to the issue of land. On forming the government in 1995, and again in 1997, Mayawati sought to enforce the possession of gaon sabha Footnote ⁶⁰ plots that had been allotted to landless SCs by previous governments, but that they had failed to claim due to forcible occupation by landed castes. This effort faced overwhelming resistance, with dominant strata tapping their co-caste contacts in the land administration to evade being identified as transgressors. A close collusion between patwaris (officials responsible for maintaining village land records) and landed strata ensued, with the former falsifying records to protect the latter. On occasion, the middle tiers of the administration were involved as well. Tehsildars (who oversaw all revenue-related functions in the tehsils under their charge and who supervised the functioning of patwaris) were at times party to these conspiracies and guilty of doctoring data pertaining to enforcement.Footnote ⁶¹

Nothing, however, compared to the uproar that ensued following the BSP's implementation of the SC/ST Act. This was the third programmatic component of BSP rule that evinced hostility, this time not just from the bureaucracy but also from the BJP, the BSP's three-time partner in government. The BJP's hostility had to do with the adverse consequences that the Act's enforcement had for upper castes. The unyielding stance Mayawati adopted on the issue of prosecution did away with the impunity with which dominant strata, and Thakurs in particular, had committed atrocities against SCs for so long. This, and the enhanced opportunities for strike action (to raise wages and secure better terms of work) that the Act's enforcement secured for low caste agricultural labour, made for a great deal of unhappiness among upper castes, who formed the BJP's base.Footnote ⁶²

Upper castes threw their weight behind a BSP-BJP coalition in 1997, calculating that the BJP would succeed in tempering Mayawati's zeal once she needed its support to govern. But this calculation failed and upper caste disenchantment with the BJP followed. It was in a desperate attempt to salvage upper caste support (particularly Thakur support, which showed signs of shifting to the SP) that Kalyan Singh—the BJP politician who succeeded Mayawati as leader of the governing coalitionFootnote ⁶³—sought to amend the working of the SC/ST Act upon assuming office. Singh issued a government order stipulating that an accused was not to be jailed while charges were being investigated and a plaintiff was not to receive money to help him fight the case in court. The BSP responded immediately, withdrawing support from the government.Footnote ⁶⁴

It will be evident from this account that the state in UP was badly fragmented during BSP rule. The fragmentation went beyond a simple political/administrative binary; with a BSP-BJP alliance in place three times, the political arm of the state was sharply fractured within. This divided state played a critical role in preventing a lower caste-dominated legislature from delivering substantive and lasting policy gains to lower caste constituencies. The bureaucracy in UP demonstrated a remarkable capacity to defy emerging representational logics and to resist the pressures that these generated. This defiance was aided by the biases prevailing in both the political establishment and society at large. The systemic resistance to the political ascendancy of SCs and OBCs, and the successful obstruction of policy initiatives aimed at empowering these groups, provides an important context against which to understand the phenomenon of the bahubali.

The Hindi term bahubali connotes physical strength and aggressive masculinity, and is used in North India to refer to a brand of politicians who are established criminals, with proven records of murders, kidnappings, extortion, smuggling, and the like. These figures, despite their several run-ins with the law, have no fear of the police and the administration, for two reasons. The first is the easy recourse to violence for which they are notorious, and their capacity to unleash it even as they remain lodged behind bars, by activating the criminal networks that they control, which has the effect of terrorizing government servants no less than ordinary citizens.Footnote ⁶⁵ The second is their money power, drawn from the largely illegal businesses in which they remain involved, which they use to buy off a wide gamut of state agents—police officers, bureaucrats, judges, jail wardens, and officials overseeing the conduct of elections. It is precisely this hold that bahubalis exercise over state employees, through a combination of bribes and terror, that accounts for why parties representing SCs and OBCs are so eager to have them aboard their platforms. It allows these parties to circumvent, in part, the bureaucratic obstruction they typically encounter when channelling resources to subaltern groups. The BSP thus offered parliamentary and state assembly tickets to an entire spectrum of hardened criminals, with most of them emerging winners, often several times in a row. A few of these criminals were: Nafeez Kalia, Kadir Rana, and Yogesh Verma, operating out of Meerut, Muzaffarnagar and Saharanpur districts in western UP; and the brothers Afzal and Mukhtar Ansari, and Ramakant and Umakant Yadav, operating out of Azamgarh, Ghazipur, Mau and Jaunpur districts in the east.Footnote ⁶⁶ The RJD in Bihar likewise inducted the following: Pappu Yadav, operating out of Purnea and Madhepura districts; Vinod Yadav, operating out of Bhagalpur district; Surendra Yadav, operating out of Gaya district; Mohammed Shahabuddin, operating out of Siwan district; Makhi Paswan, operating out of Khagaria district; and Mohammed Suleiman, operating out of Kishanganj district.Footnote ⁶⁷

Bahubalis from SC and OBC backgrounds typically leveraged their grip over the local state to empower lower castes and the poor, and damage the powerful. They invariably had a social justice component to their politics and took on a Robin Hood persona in the constituencies they represented; it was precisely this element that aligned their functioning with the programmatic agenda of lower caste parties. These bahubalis empowered subaltern constituents by getting the bureaucracy to award them public sector jobs and lucrative government contracts, involving the construction of public infrastructure and housing. They also instructed the police to expressly pursue cases that involved upper caste offenders and SC and OBC victims, and bring the former to justice. But the bahubali idiom was not in any way circumscribed by the political parties to which bahubalis remained affiliated. Thus, bahubalis frequently held jan durbars (mass assemblies) where they addressed the concerns of constituents directly, settled local disputes through informal mechanisms that turned on their exercise of charismatic authority, and helped poor voters by distributing largesse from their own pockets.Footnote ⁶⁸ In this sense, bahubali politics constituted an autonomous domain, marked by practices not expressly sanctioned by political parties and which exceeded the legal and institutional limits usually ordering the conduct of elected representatives in liberal democracies. It was symptomatic of this autonomy that the bahubali did not derive legitimacy from the political party; rather, it was to harness the legitimacy that adhered to his figure, prior to and independently of party affiliation, that the party wanted him on board. It was also this autonomy that allowed bahubalis to contest elections as independent candidates—and win.Footnote ⁶⁹

While we are concerned here with bahubalis from the lower and backward castes, and the logic underlying their affiliation with the social justice parties, it needs to be pointed out that not all bahubalis came from subaltern backgrounds and it was not SC and OBC parties alone that awarded them tickets. There were several bahubalis from the dominant castes who were courted and accommodated by parties across the spectrum, the Congress and the BJP included. Thus, in the last state assembly election held in UP in 2017, 137 out of 383 (36 per cent) BJP candidates and 36 out of 114 (32 per cent) Congress candidates had criminal cases pending against them. The figures for the BSP, the SP, and the Rashtriya Lok Dal (a state-based party with a base in western UP) were 150 out of 400 (38 per cent), 113 out of 307 (37 per cent), and 56 out of 276 (20 per cent) respectively.Footnote ⁷⁰ Likewise, in the last state assembly election held in Bihar in 2015, 95 out of 157 (61 per cent) BJP candidates and 23 out of 41 (56 per cent) Congress candidates had criminal cases pending. The figures for the RJD and the Janata Dal (United)Footnote ⁷¹ were 61 out of 101 (60 per cent) and 58 out of 101 (57 per cent) respectively.Footnote ⁷²

The question thus arises as to why it made sense for parties across the board to grant such a considerable number of nominations to criminal-politicians. Bahubalis generically, irrespective of caste background, contributed handsomely to party coffers and also financed their own campaigns. They further commanded resources—money and muscle, and hold over the local state—that bestowed on them a definite electoral edge. All of this gave them a clear advantage, across parties, in the competition for tickets. But when, upon being elected, bahubalis from dominant strata used their grip over the administration to channel benefits to co-caste constituents, they merely reinforced the ties that were already in place, connecting upper castes within the state to those outside it. The political logic driving the functioning of SC and OBC bahubalis was exactly the reverse. These figures sought to weaken the bureaucratic-societal nexus that had serviced dominant strata for decades; their success in this regard, consequently, had politically subversive effects.

I end this section by briefly describing the efforts made by the ECI (which administers parliamentary and state assembly elections and is responsible for upholding the sanctity of the electoral process) to address the entanglement between bahubalis and electoral politics. In what amounts to a substantial handicap, the ECI does not have the power to ban criminals from contesting elections; it can only make recommendations in this regard. Constrained thus, the ECI has sought to ensure that voters remain informed about the criminal antecedents of candidates; it has also sought to keep bahubalis from attempting to influence the outcomes of electoral contests. As I argue in the concluding section, the effort to keep voters informed has failed to achieve much in a context where many constituents vote for bahubalis precisely because they are criminals and in the full knowledge that they are so. On the other hand, the measures undertaken to keep bahubalis from bribing officials, intimidating voters, and capturing booths have had some impact, although it is only during the ECI's more proactive phases that these have been implemented with sufficient force. One such phase was in 1990–96, when the ECI, headed by T. N. Seshan,Footnote ⁷³ took the following steps: staggered polling across several phases in every state, thereby allowing paramilitary forces from outside the state (who were available only in limited numbers) to be deployed for election duty, which made it difficult for candidates to buy off the law and order machinery and resort to booth capturing; placed people with criminal records in preventive detention on the day of the polls, which limited the number of associates bahubalis could mobilize; issued identity cards to voters, which sharply reduced the scope for ‘bogus voting’;Footnote ⁷⁴ and transferred bureaucrats and police officers whom the ECI had reason to think were of doubtful integrity and in the pay of specific politicians and parties.Footnote ⁷⁵ These measures produced results in the 1993 state assembly election in UP. The booth capturing count in this election fell to 255, from 873 in the 1991 state assembly election. The number of constituencies in which polling had to be suspended or deferred also came down—from 17 in 1991 to three in 1993, with the number for murders committed on polling day improving as well—down from 36 to three.Footnote ⁷⁶ M. S. Gill, who succeeded Seshan as chief election commissioner (CEC), continued with several of his predecessor's practices and took the further step of ‘transferring all district-level civil servants and police officers … who had been in their posts for four or more years’ at the time of the announcement of polls.Footnote ⁷⁷ It has to be said, however, that the ECI did not always succeed in countering the operations of criminal-politicians. This was not very surprising, given an environment in which almost all political parties found it convenient to award criminals tickets, and where many voters were attracted by those very attributes in candidates that derived directly from their criminality.

To tease out the implications that the operation of bahubalis from subaltern backgrounds had for the process of democratic transformation in India, it is important to recall the context in which the bahubali emerged as an instrument of levelling. This was one in which democratically elected governments with a social justice mandate had immense difficulty in getting regular institutional channels to translate that mandate into concrete policy outcomes. It was this institutional obstruction that compelled lower and backward caste parties to rely on gangster-politicians to force through a redistribution of resources, in those pockets where their criminal operations were based and their writ over the state ran. It was precisely because they were able to bring the local state to its knees, and force state officials to work for those subaltern groups they otherwise remained reluctant to serve, that bahubalis commanded so much legitimacy among lower caste voters. Remarkably, this legitimacy accrued to them despite their unabashed flouting of the law and practically every procedural norm governing democratic politics. This pointed to the growth of a public culture in North India that valued right consequences, or outcomes, over correct procedures, or means—that was impatient, in other words, with a procedural conception of democracy that could not accommodate the modalities that went to secure justice in real life. But this attrition of procedural sanctity could not be simplistically traced to the onset of SC and OBC assertion in the 1990s. Its triggers went further back in history, to India's colonial past, from which the post-colonial state and its institutions inherited prejudices that affected their functioning in independent India.Footnote ⁷⁸

I have argued in this section that governments representing subaltern groups deliberately overlooked, and sometimes actively encouraged, the infringement of procedural norms when this helped them overcome bureaucratic resistance and arrive at bounded egalitarian outcomes. In the next section, I illustrate how the unfolding logic of backward caste assertion made way for the decimation of institutions, the attrition of bureaucratic capability, and the shelving of the development agenda.

OBC assertion and the animus against development

The 1990 state assembly election in Bihar, held in the aftermath of V. P. Singh's implementation of the OBC quota, decimated the Congress and brought the JD to power. It also transformed the caste make-up of the Bihar legislature, with the proportion of backward caste MLAs (members of the Legislative Assembly) being higher than that of upper caste MLAs for the first time (36.11 per cent and 32.40 per cent respectively). With Lalu Yadav becoming chief minister, the ministry also came to bear the marks of this reversal, with Yadav selecting ministers from OBC ranks. This spurt in backward caste representation gained further momentum in 1995. The assembly election held that year brought the JD back to power; Yadav became chief minister again, and the proportion of backward caste legislators rose dramatically to 49.69 per cent, with that of upper caste legislators falling to 17.28 per cent.Footnote ⁷⁹ This set the stage for a confrontation between a bureaucracy dominated by upper castes and a legislature dominated by OBCs, triggering intra-state fragmentation much like the one marking BSP rule under Mayawati in UP.

This fragmentation manifested itself, first and foremost, in a steep hike in corruption as upper caste bureaucrats exploited the opportunities presented by the rise of a new generation of politicians who had no knowledge of administrative procedures. Further, some officials ‘used this knowledge advantage to stealthily subvert and sabotage through bureaucratic foot-dragging the few constructive policy interventions that the RJD government attempted’.Footnote ⁸⁰ This obstruction aimed at protecting the interests, not only of the bureaucrats in question, but of the socio-economic strata they represented—a landed-cum-administrative class composed of Brahmans, Thakurs, Kayasths, and Bhumihars,Footnote ⁸¹ the upper castes in Bihar. Lalu Yadav's regime defined the curtailing of their dominance as its prime objective. The politics of backward caste empowerment thus threw up a juxtaposition, pithily described in common parlance as ‘democracy versus bureaucracy’,Footnote ⁸² that saw the contradictions of India's post-colonial politics come to a head. The bureaucratic apparatus, and in particular its higher echelons, stood determined to resist the project of caste-based social justice, and defy the government of the day in order to reinforce the status quo. Consequently, this apparatus came to be thoroughly undermined by the government in question, which had been voted to power on a platform of OBC empowerment, and which had every intention of delivering to its base, even if this meant having to bear the costs that a defunct administration imposed.

It was not simply the administration, but also the judiciary (whose social composition mirrored that of the administration) that came into conflict with Lalu Yadav's government. The year 1996 saw a major scandal (popularly known as the ‘fodder scam’) erupt, in which Yadav stood implicated, which involved the embezzlement of nine and a half billion rupees from the Bihar Animal Husbandry Department. The fodder scam was initially being investigated by the Bihar Police, who report to the state government. But after hearing a public interest litigation, the Supreme Court of IndiaFootnote ⁸³ turned the case over to the Central Bureau of Investigation (CBI),Footnote ⁸⁴ India's premier law enforcement agency, stipulating that powers of oversight be retained by the Patna High Court, the apex court in Bihar. Thus began a phase of protracted conflict between the government, on the one hand, and the CBI and the courts, on the other, which culminated in Lalu Yadav's arrest. It is telling that Yadav portrayed the entire investigation as an elite, upper caste conspiracy that sought to rob him of the mandate he had received from the masses. He further alleged that it was because he remained invincible in the arena of ‘democracy’, that is, electoral politics, that his enemies had to use the bureaucracy (referring to the CBI) and the judiciary to subjugate him. This contrast, once again, pitted the representative realm of politics, where he held sway, against the unrepresentative institutions of the state, where elites reigned supreme.

Yadav, indeed, went so far as to accuse the ECI of also being part of the upper caste conspiracy against him. The JD under his leadership had led a campaign in 1993 against the electoral reforms introduced by T. N. Seshan, the chief election commissioner, and had sought to have him impeached. Yadav again crossed swords with Seshan in the run-up to the 1995 state assembly election in Bihar. He refused to meet the December 1994 deadline for issuing identity cards to voters, following which Seshan threatened to postpone the polls. Yadav then took the issue to the Supreme Court, which reminded Seshan that his powers did not allow for such postponement.Footnote ⁸⁵ Seshan did, however, succeed in staggering the polls beyond the constitutional deadline of 15 March 1995, when the term of the incumbent Bihar assembly expired. He did so by pointing to a severe law-and-order breakdown that had occurred during the first round of voting on 11 March, when 14 people had been killed in poll-related violence, and rigging and booth capturing had been rampant. Alleging that Yadav's promise to clean up Bihar's electoral process had proved completely insincere, Seshan, armed with inputs from senior bureaucrats and police officers, argued that 1,050 companies of central security personnel would be required to enforce the rule of law during the upcoming rounds of polling. Recognizing that the centre would need time to mobilize these forces, and that the latter would also have to be moved after every phase, he made the case to P. V. Narasimha Rao, the Congress prime minister, for rescheduling the voting to follow in Bihar—from 15 and 19 March to 15, 21, and 25 March.Footnote ⁸⁶ Seshan's recommendation was accepted by Rao. Yadav (who did not want polling to occur after 15 March, as he would have ceased to be chief minister by then, and lost control of the electoral process) went on to denounce the ‘tyranny of the Brahminical forces’,Footnote ⁸⁷ and accused the CEC of plotting with the Congress and the BJP to thwart his return to power.Footnote ⁸⁸

The fodder scam investigation led to Lalu Yadav's leadership of the JD being challenged by senior politicians in the party, following which he split the JD in 1997 and formed a new party, the RJD, taking with him the majority of JD legislators from Bihar. His consequent arrest and resignation as chief minister, furthermore, did not allow his detractors to take charge. In an unprecedented move, Yadav placed his wife, Rabri Devi, in the post of chief minister and proceeded to govern by proxy from jail.Footnote ⁸⁹ While falling short of a majority, the RJD emerged as the single largest party in the 2000 state assembly election and, with Congress support, went on to form the government, again with Rabri Devi as chief minister.

The period 1990–2005—during which the JD and the RJD ruled Bihar—saw several measures being adopted to rein in a recalcitrant administration. The first was the transfer of bureaucrats, with critical administrative positions in the districts, especially the powerful office of the district magistrate, being handed over to OBC, Muslim, and SC officers. ‘By 1995, twenty-six out of fifty district magistrates … in Bihar were lower castes.’ Transfers took place in the upper tiers of the bureaucracy as well. In 1993, the JD ‘government replaced both the chief secretary and the director general of police, both of whom were Brahman, with lower-caste officers, overlooking two IAS and three IPS upper-caste officers who were in line for the posts’.Footnote ⁹⁰ But the strategy of transfers had its limits, as most government departments were staffed almost entirely by upper castes, and had only a very small pool of lower caste employees. This paucity resulted from the inadequate filling of job quotas by earlier governments. The RJD government was keen to do justice to quotas and appoint officers from subaltern backgrounds, but the impact of neoliberal reforms and the precariousness of Bihar's finances severely restrained its capacity to hire. Unable to recruit lower castes, and determined not to appoint upper castes to important positions, the RJD took to keeping posts deliberately vacant, creating ‘state incapacity by design’. Thus:

Another strategy of containment consisted in shifting power away from the bureaucracy to the political class, comprised not only of elected representatives, but also party activists, political brokers, and musclemen. This led to a spike in political interference in bureaucratic and police functioning at all levels.Footnote ⁹² Further, since upper caste bureaucrats had siphoned off development funds over the past several decades and channelled these to co-caste elites located outside the state, with these funds enhancing the patronage resources at their disposal, the RJD perceived these funds to have fed the concentration of power in upper caste hands and now proceeded to cut off their flow. At this juncture, it was not just bureaucracy, but also development that came to be juxtaposed with democracy and accorded low priority in a populist environment bent on effecting ‘just’ outcomes that were immediately visible.Footnote ⁹³

The cooperative movement had been a major component of planned agricultural development in Bihar. But since landowning castes had exercised tight control over the cooperative sector and drawn subsidized agricultural credit from cooperative banks, the RJD government targeted these banks, effectively making them insolvent.Footnote ⁹⁴ The RJD also refused to provide agricultural subsidies, control prices of agricultural inputs, or ensure the steady supply of electricity that was necessary to keep irrigation costs low. Having the RJD in power also encouraged labour unrest in the countryside, with daily wage workers (most of them SCs and poor backwards) resorting to strike action to improve their terms of work. All of this made direct cultivation an increasingly unprofitable enterprise for the landowning class. Even sharecropping became unviable, as landowners found it impossible to enforce the oppressive sharecropping arrangements that had earlier prevailed, with their bureaucratic and police contacts (who had helped them control tenants) having been rendered toothless by the RJD government. RJD rule, thus, saw landowning households begin to disengage from agriculture, lease out their plots to tenants, and rely on fixed cash rents, of which there was a sharp increase in the countryside.Footnote ⁹⁵

A ‘democratization of agriculture’Footnote ⁹⁶ consequently occurred, with SCs and OBCs, particularly Yadavs, taking over cultivation in the capacity of tenants. This tenant-cultivator class, moreover, had easy access to the state, which was earlier a privilege enjoyed only by landed groups. But RJD rule did not succeed in institutionalizing and making permanent the effects of this downward shift of power in the countryside. For tenants to be granted ownership rights, decisive policy interventions, in particular land reforms, were required, which could not be undertaken in face of the widespread bureaucratic incapacity the RJD had helped to engender. While the bureaucracy would have resisted land reforms vehemently, the important point is that Lalu Yadav, unlike Mayawati, chose to decimate the bureaucracy, rather than discipline it. Having made the bureaucracy altogether ineffective in an effort to empower his subaltern base, he could not render stable the gains he had secured for that base without bureaucratic action.Footnote ⁹⁷ Furthermore, the RJD's backward caste constituency remained differentiated by class: while some better-off OBCs possessed land, an array of poorer OBCs possessed none. The former staunchly opposed redistributive agrarian policies, which exposed the contradictions of the RJD's politics and the inherent limits of any programme of social justice that pivoted around caste.Footnote ⁹⁸

Unlike other chief ministers in India, who chased corporate investment energetically and competed aggressively to keep their respective states ahead in the industry, services, and information technology sectors, Lalu Yadav remained hostile to private and corporate actors, whom he associated with the vile ‘upper caste system’. As Witsoe points out, a politician with such views could reasonably have been expected to bolster the state sector and ensure a steady provision of public goods and services to offset the impact of neoliberal policies. But Yadav did precisely the opposite, shifting power away from the state to a subaltern political class. He had determined that the bureaucracy had to be crippled if the dominance of upper castes was to be dented. It was for this reason that he defined his agenda of lower caste empowerment in symbolic terms, which made empowerment out to be a matter of dignity, rather than in terms of substantive policy outcomes.Footnote ⁹⁹ This last would have required the RJD to deliver tangible benefits to its base, which would have been impossible without a functioning administration in Bihar.