The dalit contract with India


back to issue

IT was a hot May in Delhi in 2009, and listless crowds, patrolled by the blue-capped Bahujan Volunteer force, made up the Bahujan Samaj Party’s pre-election rally in the Ram Lila Grounds. Right up front was a large raised platform meant for the media. Its scale was ironic as well as inevitable: ironic, because Mayawati has made a career out of ignoring ‘upper caste journalists’ for misrepresenting her politics; inevitable, because in a general election otherwise too complex for sound bytes, the question: ‘Will India have its first Dalit woman prime minister?’ had a clarifying elegance to it. On either side of the media platform, and a good 500 metres from the podium, were her people. The distance between speaker and spoken-to conveyed an imperium that went with Mayawati’s politics of dignity. Her rant against the conniving Congress, past holder and current predator of her Dalit vote bank in Uttar Pradesh, also made sense.

What was puzzling was the speech itself. More than half of it was devoted to details of the Constitution – from B.R. Ambedkar’s tussle with the Congress over its drafting, to the reservations and empowerment it offers to Scheduled Castes six decades later.1 Sentences I had heard in the careful calibrations of law school, the monotonous baritone of courts, and the air-conditioned confines of the India International Centre were now being taught by a mass politician to legions of her unlettered followers. I have since wondered about the relationship between the Indian Constitution and Dalit politics. What is the basis of this link in realpolitik; what is its psychological character? And what does it say about India’s founding document, that sixty years on, the most prominent space it has in mass politics is in the non-liberal articulations of former Untouchables?

First, the obvious. The Constitution is used instrumentally to strengthen Dalit representation in politics. Dalits number 16.2 per cent of India’s population. Since Article 330 of the Constitution guarantees that 15 per cent of the Lok Sabha will be occupied by Dalits only, they are adequately represented in the Lok Sabha. Compare this to Muslims in India, who are roughly 13.4 per cent of the population, but do not benefit from political reservations. As a result, Muslim-centric parties (like the Muslim League) are inconsequential; the percentage of Muslims in politics is far below their national average. What makes Dalit politics even more impressive is that Dalit votes tend to be split. Analyzing Dalit vote trends in the 2009 elections, Rahul Verma found that rich urban Dalits tended to vote for the Congress, while it was left to the poor, rural Dalits to vote for the BSP. By contrast, Muslims – rich or poor, upper caste or low caste – vote tactically for the same party.2 Yet Muslim electoral politics is not able to compete with the Parliamentary quota that the Constitution guarantees to Dalits.


Conversely, reservations alone cannot explain Dalit political power. Take the Scheduled Tribes, who are the only other group allotted political reservations in the Constitution. Their Ambedkar – the Oxford-educated Jaipal Singh Munda – was a forceful voice in the Constituent Assembly debates. But this has not led to tribal politics reaching anywhere near the organizational level of Dalit politics. As Ramachandra Guha points out, 60 years after Independence ‘unlike Dalits, they [tribals] have been unable to effectively articulate their grievances through the democratic and electoral process.’3 One major reason why Dalits are able to organize better is the shared experience of untouchablity, which connects Dalit jatis scattered across the subcontinent. By contrast, there is little to link tribals from central India (like Jaipal Singh) and tribals from the North East. There are no shared social experiences to cause them to vote as a block.

Dalits are also numerous enough in several states to benefit from India’s first-past-the-post politics. In Uttar Pradesh, for instance, where politicians typically need around 25 per cent of the votes to win, the 20 per cent Dalit population begins with a head start over other identity groups fighting for political spoils. But even in areas where tribals are a majority, tribal politics is either fragmented (Chhattisgarh) or mauled (Jharkhand). Besides, reserved seats do not always benefit Dalit-only parties. In Uttar Pradesh, the BJP has consistently won most of the reserved seats. Ajoy Bose, a political biographer of Mayawati, explains it thus: ‘Since all candidates are Dalits, the Dalit vote is divided… the BJP’s Dalit candidate had the extra benefit of the party’s traditional upper caste base.’4 This is not necessarily a negative for Dalits: mainstream parties with Dalit politicians can initiate caste compromises of the kind Mayawati has recently attempted in UP, with her bid to woo that state’s Brahmin community. Political reservations provide a platform to Dalit politicians for negotiating from a position of strength.


Political reservations are only one half of the bridge that leads to Dalit power; reservations in government jobs (Article 15(4)) and employment (Article 16(4)) are the other half. Administrative reservations have helped in two ways. The first is that they have created a Dalit elite whose members have gone on to stand for political office. Mayawati’s father was a (reserved) government employee, and Behenji nurtured ambitions of writing the difficult central administration exam (UPSC) before foraying into politics.5 Since rich Scheduled Castes can legally avail of quotas, the same Dalit families – like Kumari Shailja’s – have, in a couple of generations, become a political elite.

Beyond individual examples like Kanshi Ram, Mayawati and Meira Kumar, lies yet another factor. Dalit government employees have organized themselves within government, and this organizational structure has formed the nucleus of a larger political movement outside the steel frame of the bureaucracy. In 1978, Kanshi Ram formed the Backward and Minority Communities Employees Federation, followed by the Dalit Soshit Samaj Sangharsh Samithi (DS4), which eventually became the political Bahujan Samaj Party.


Dalit government employee organizations and student federations, all beneficiaries of constitutional reservations, act as feeders into Dalit parties or SC/ST cells within national parties – much like the Student Federation of India feeds into the CPI(M), and the ABVP and RSS provide young leaders to the BJP. The relationship between Dalit government employees and political parties has been ill-studied in academia so far. Anecdotal evidence suggests a potent cycle. Perhaps this example captures it best: Ram Vilas Paswan heads a Bihar-based Dalit party called the Lok Janshakti Party. Between 1996 and 2009, he was a regular feature in every union cabinet. He repeatedly won from the reserved constituency of Hajipur in north Bihar, and once held a Guinness record for winning an election there by the largest ever margin. In 2003, he was invited to speak at a conference in Berlin, with the banal title: ‘Dalit politics is here to stay’. This is what the veteran Dalit politician had to say:

‘We could significantly enforce the Presidential Directives to include proportionate numbers of SCs and STs in the Delegations going abroad… We could also appoint Mr Birke Ram, as Director Finance of the big Railway Public Sector Corporation. Today it is one of the highly profitable PSUs in the country… No SC/ST was ever allowed to become the Cabinet Secretary to the Government of India.’

It is telling that his speech was not about electoral politics, but about the nitty-gritty of administrative transfers and postings.6 That was what he saw as the true import of Dalit politics.

While the importance of political participation can’t be understated for any community that faces historical injustice and discrimination, increased Dalit representation in politics does not automatically mean that Dalit interests are better articulated. Dalit parties like the BSP are not necessarily ‘purer’ than mainstream parties like the Congress, where the necessity to woo Dalit voters has to be balanced with the impetus to form a pan-Indian majority. The close relationship between Dalit student, occupational, and political formations also has its drawbacks. Apart from seeing state offices in purely instrumental terms, it also makes other forms of Dalit identity subservient to the political. Yet, for better or worse, this nexus between political and administrative reservations has become the hinge on which contemporary Dalit politics swings.


So far I have described how the Constitution has created a Dalit political and administrative elite who work in tandem, incubating structures in government organizations before placing them in the rough and tumble of electoral politics. But Mayawati’s May 2009 speech hinted at a psychological role of the Constitution, one that goes to the heart of contemporary Dalit politics.

To understand this, it is critical to see constitutional reservations for Dalits not as an idea of equality based on first principles, but as a historic compromise; a result of political power play within India’s freedom movement. By the 1930s, the British faced two major claimants for nationhood, in addition to the Indian National Congress. Muhammad Ali Jinnah’s Muslim League claimed to speak for British India’s 20 per cent Muslims, while B.R. Ambedkar claimed to represent British India’s ‘Depressed Classes’. As Sunil Khilnani points out, it was in the British interest to deny India freedom by claiming that there were too many discrete Indian groups to form a single, integrated nation.7 


Motivated at least in part by this latter argument, the British announced, in 1932, the creation of ‘communal electorates’, i.e. separate seats and voters for Dalits and Muslims. An agitated Mohandas Gandhi went on a fast unto death against separate electorates for Dalits. Faced with intense pressure from popular sympathy for an ailing Gandhi, Ambedkar compromized, giving up on the demand that Dalit voters be kept separate, but gaining reserved constituencies for the ‘depressed classes’. This Poona Pact of 1932 became the basis for providing reservations to the ‘depressed classes’ in the Government of India Act, 1935, which in turn, became the template for the Constitution of India, 1950.

This power-sharing agreement ended up benefiting both Ambedkar and the Indian National Congress. As Sekhar Bandopadhyay points out, from 1916 onwards, Dalit political assertion was propped up by colonial patronage. But as Independence approached, Ambedkar’s party faced annihilation from the Congress’ ability to put up Dalit candidates and win Dalit votes.8 In the 1946 elections, the Congress could accurately claim to represent the largest share of the Dalit vote.9 The inclusion of Ambedkar in the Constituent Assembly, and of the terms of the Poona Pact in the Constitution, were thus welcome steps for Ambedkar. For the Indian National Congress, the immediate benefit of the Poona Pact was to put an end to the idea of a separate Dalit nation. This might seem trifling today, but as late as 1940 Ambedkar harboured dreams of a separate country for Dalits. In his book Pakistan or the Partition of India, he argued that ‘the transfer of minorities is the only lasting remedy for communal peace.’10 


But the most far-reaching implication of the incorporation of the Poona Pact by the Constituent Assembly was that the Constitution became the Dalit contract with the Indian nation. That contract is not in sync with the liberal nationalism that Nehruvian interpreters of India’s Constitution like to extol. It is, instead, a hard-nosed power-sharing agreement between groups, more in the nature of the agreement between Christians and Shi’ites in Lebanon or the Constitution of post-Apartheid South Africa. Using examples from these two countries, Leonard Wantchekon points out that power-sharing agreements are necessary for the transition from a state of conflict to a state of democracy.11 The Poona Pact, that violates notions of formal equality and contributes to what is wryly described as the Constitution’s ‘asymmetric discrimination principle’,12 was perhaps necessary to avoid the alienation of the Scheduled Castes from the Indian mainstream.


Wantchekon’s research also shows that once democracy comes, the majority is tempted to renege on the power-sharing agreement. It speaks of the wisdom of the national movement that in 1950 – when the Congress was not only in power in India, but had trumped Ambedkar for the Dalit vote – it resisted the temptation to renege on the Poona Pact. The move, instead, to appoint Ambedkar law minister and head of the drafting committee had the symbolic value of sealing the Constitution in the eyes of subsequent non-Congress Dalit politicians (even if Ambedkar resigned soon after). Since then, the Indian Constitution has been amended to provide the same reservation benefits to a numerical majority (Other Backward Classes). The logic of the Poona Pact, meant to protect a minority from the vagaries of the majority, has been turned on its head. Yet, reservations for the numerical majority ensure that the logic of the Poona Pact will never be questioned; reservations for Dalits is unlikely to be withdrawn.

Benedict Anderson argues that ‘political symbols play a major part in the way a nation is depicted and fed into the imagination of its citizens.’13 The most famous Dalit totems in modern India are the blue-suited Ambedkar statues.14 They dot entrances to Dalit bastis in Indian villages and demarcate spaces in urban India where Dalit politics has gained a foothold. For a group that has been defined by physical exclusion, the power of these statues comes from their placement. Mayawati’s statue parks in Uttar Pradesh have the same aim – capture physical space, and in doing so create history for those who have been denied it for centuries. The political symbolism of the constitutional reservations for Dalits also lies in their placement. The Constitution is a mere collection of words, but it aims to map out the geography of Indian nationhood. The symbolism of the constitutional provisions for Dalits is that it carves out 15 per cent of this national space for Dalits, and in doing so creates the historic basis for shared nationhood. In that sense, its symbolism is similar to that of the Ambedkar statues.


Our political landscape has altered since the Constitution was enacted. The decline of the Congress and the growth of region- and caste-based parties have ensured that in a first-past-the-post-system, Dalits, a significant numerical minority, have increased bargaining power. The decline of the Congress has also led Dalits to vote for other parties. In the run up to Independence, Ambedkar’s party was roundly defeated in the 1946 Constituent Assembly elections. By contrast, today’s Dalit parties either win power (BSP) or gain significant vote share (LJP, RPI). The fact that the Constitution has allowed and facilitated these electoral and rhetorical shifts speaks volumes for its elasticity.

In the sixty-three years since India’s Independence, diverse ethnic, linguistic or ideological groups – whether championing language chauvinism in Tamil Nadu, separatism in Kashmir, or rebellion in the red corridor – have questioned India’s Constitution. Each of these identity groups benefits from special constitutional provisions. Linguistic groups, around whom the states were reorganized in 1956, benefit from the federal provisions of the Constitution. In addition, Article 30(1) provides ethnic, linguistic and religious groups autonomy in their higher educational institutions. Kashmir enjoys relative autonomy through the controversial Article 370; tribal areas have similar rights. But these ‘group rights’ can still be justified within liberal jurisprudence – they don’t have anything like the slice-of-cake logic that the Poona Pact ensured. This perhaps explains why Dalit politicians have never criticized the Constitution, only interpreted it their way.


Mayawati ended up on the wrong side of the 2009 general elections; her Delhi speech proved to be in vain. Not only did she win fewer seats than expected, she lost out to the Congress’s resurgence in her home state of Uttar Pradesh. The Third Front on which her prime ministerial ambition was tethered, was undone by the Congress and the BJP. Yet while the opportunist in Mayawati has one eye on tomorrow’s elections, her other eye is on history. She continues to build statues of herself and her Dalit pantheon, continues to sprinkle her speeches with references to the Constitution of India. Symbols are for posterity as well as expediency and sixty years on, the Dalit contract with Indian nationhood shows no sign of ageing.



1. ‘Dalit’ is a political term referring to ex-untouchable castes. ‘Scheduled Caste’ is a legal term, which excludes Christian and Muslim Dalits. This essay uses both phrases interchangeably.

2. Rahul Verma, ‘Dalit Voting Patterns’, Economic and Political Weekly 44(39), 2009.

3. Ramachandra Guha, ‘Adivasis, Naxalites and Indian Democracy’, Economic and Political Weekly 45(32), 2007.

4. In conversation with the author, May 2009.

5. Ajoy Bose, Behenji: A Political Biography of Mayawati. Delhi, 2008.

6. Pol.htm

7. Sunil Khilnani, ‘Arguing Democracy: Intellectuals and Politics in Modern India’, CASI Working Paper Series 9(2), 2009.

8. Sekhar Bandyopadhyay, ‘Transfer of Power and the Crisis of Dalit Politics in India, 1945-47’, Modern Asian Studies 34(4), 2000.

9. Id.

10. B.R. Ambedkar, Pakistan, or the Partition of India. Bombay, 1940. c.f. Sunil Khilnani, op cit.

11. Leonard Wantchekon, ‘Credible Power-Sharing Agreements: Theory With Evidence From South Africa and Lebanon’, Constitutional Political Economy 11(4), 2000.

12. Sudhir Krishnaswamy and Madhav Khosla, ‘Reading A.K. Thakur v. Union of India: Legal Effect and Significance’, Economic & Political Weekly 43(29), 2008.

13. Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism. London, 1983.

14. Nicolas Jaoul, ‘Learning the Use of Symbolic Means: Dalits, Ambedkar Statues and the State in Uttar Pradesh’, Contributions to Indian Sociology 40(175), 2006.