Popular political wisdom has it that the battle for New Delhi passes through Lucknow. Uttar Pradesh — with 80 Members of Parliament (MPs) — is the greatest purveyor of representatives in the Lok Sabha. This State has provided most of India’s Prime Ministers and has been home to the Nehru family for five generations. The State also holds crucial importance to the Bharatiya Janata Party (BJP), which, being largely absent in the south, east and north-east of the country, needs a strong performance in Uttar Pradesh to hope to form a government at the centre.
With 18 per cent of the population, Muslims in Uttar Pradesh constitute an important segment of the electorate, decisive in a large number of constituencies where they are demographically dominant. A former support base of the Congress, they are key to the electoral success of the two regional parties, the Samajwadi Party (SP) and the Bahujan Samaj Party (BSP).
However, despite their demographic importance, Muslims in Uttar Pradesh have been traditionally under-represented in the State Assembly until the 2012 Assembly elections in which they obtained for the first time a near-proportionate representation (17 per cent of the seats).
A series of factors are usually invoked to explain the chronic under-representation of Muslims and some of them need to be re-assessed in view of the recent improvement of their situation.
The first factor is a demographic one. The uneven geographic distribution across the State confines the political strength of Muslims to specific sub-regions – Upper Doab, Rohilkhand, Awadh and the North of Poorvanchal. Twenty districts and around 100 tehsils have a Muslim population over 20 per cent. The Muslim vote is a determining factor in 34 out of 80 Lok Sabha constituencies and in 130 out of 403 Vidhan Sabha [State Legislative Assembly] constituencies. Without surprise, the geographical distribution of elected Muslims roughly matches their demographic distribution, though does not entirely equate it. Combined with the effect of the first-past-the-post (FPTP) electoral system, their chance of winning seats is usually strongly determined by their demographic strength at the constituency level.
A second factor of under-representation of Muslims is vote splitting. The presence of several Muslim candidates in a constituency can lead to a split in the Muslim vote and to the loss of seats, often to the benefit of BJP candidates.
Ten seats were lost this way in the 1996 State elections, to the sole benefit of the BJP, the only mainstream party that does not field Muslim candidates. In 2002, the BJP won 13 seats the same way, while six other seats were also lost due to the dispersion of the Muslim vote, in constituencies where sometimes Muslims are in absolute majority.
This phenomenon was much more circumscribed in 2007, where only five seats were lost due to the split of the Muslim vote. The consolidation at the local level of the Muslim vote behind SP, Congress and BSP candidates, added to the limited number of direct confrontations between SP and BSP Muslim candidates may provide an explanation for this trend reversal, which was confirmed again in 2012, the year in which Muslims achieved a proportional representation.
A third factor often evoked is the number of “lost” seats due to reservations of constituencies for candidates from the Scheduled Castes (SCs). The Sachar Committee Report identified a number of such constituencies, having a high proportion of Muslims and a low proportion of SCs. However, a recent study by Francesca Jensenius demonstrated that Muslims are statistically not over-represented in reserved seats (across 14 States), tempering thus this assertion.
The fourth factor has to do with party politics. The perception of Muslim voters splitting their votes locally can discourage parties to distribute tickets to Muslims candidates. In recent years however, mainstream parties — barring the BJP — have made more explicit appeals to Muslims voters, with the hope to add their support to their core electorates. However, a quick examination of ticket distribution shows that the number of tickets distributed by mainstream parties has indeed recently increased but not very substantially, and at a current level lower than in the early 1990s, when the representation of Muslims was at its lowest. In other terms, a large number of Muslim candidates does not necessarily translate into a large number of Muslim representatives.
Many Muslims contest on other tickets or as Independents, but most of the time lose their deposit. Uttar Pradesh has a long history of small Muslim parties. From the Majlis-e-Mushawarat or the Muslim Majlis of Dr. A.J. Faridiin the 1970s, to the Ulema Council, the United Democratic Front and more recently the Peace Party of Dr. Mohammad Ayub, these formations have fared poorly in the polls and in many instances have brought a net loss for Muslim political representation. The reasons for this lack of popularity among Muslim voters are varied. Identified as Muslim outfits, they fail to gather support among non-Muslim voters. Besides, their leadership is usually drawn from elite Muslim groups, disconnected from their base, or led by clerics who do not have the political legitimacy to carry the representation of Muslims. Often, their strategy of distributing tickets to non-Muslim candidates, to avoid the minority tag, is counterproductive. In an FPTP electoral system, there are doomed to fail.
But as we say, the attitude of parties towards Muslim voters has barely changed. And since all institutional explanations for the low representation of Muslims (party bias, effect of reservations, electoral system) fall short of explaining the recent rise of the representation of Muslims, one has to turn to another explanation, connected with their voting behaviour.
It is not easy to assess exactly “how Muslims vote” as, contrary to popular belief, they disperse their votes across parties. One can, however, sketch the following broad trajectory.
In the first three decades post-Independence, Muslims in Uttar Pradesh supported the Congress, first staunchly then reluctantly. They progressively abandoned it to the benefit of the growing secular opposition, led by the various avatars of the Janata Party. In the early 1990s, they followed en masse the newly formed SP, in the aftermath of the Babri Masjid demolition. The main driver for electoral choice was the defeat of the local BJP candidates and for a time, “Maulana” Mulayam [Mulayam Singh Yadav] seemed to be the better placed to do that.
However, through the 1990s and even more so during the 2000s, the support for the SP among Muslim voters eroded, from 75 per cent to 30 per cent in 2009, according to CSDS National Election Surveys. The SP’s decision in 2009 to tie-up with Kalyan Singh, who was the BJP Chief Minister at the time of the Ram Janmabhoomi movement, and the rejection of several prominent incumbent Muslim representatives, alienated large parts of their support among Muslims. Besides, the ascent of the BSP offered an alternative to counter the BJP.
In the recent years, the representation of Muslims increased despite a wide distribution of support among the SP, BSP and the Congress. When one looks at constituency data and available surveys, one sees that vote splitting at the constituency level decreases over time, enabling Muslim voters to use their demographic advantage. This would indicate a growing cohesiveness of the Muslim vote at the constituency level or a greater coordination of their electoral choice. One should not conclude that Muslims organise themselves at the State or even the district level. The electoral map shows that Muslims vote differently from constituencies that are close to each other. The electoral map also shows that the representation of Muslims has dramatically increased in some sub-regions such has Rohilkhand (40 per cent of the MLAs), Western Uttar Pradesh, Doab and Eastern Uttar Pradesh, each at 20 per cent of the MLAs. Their representation in central Uttar Pradesh, or Avadh, has been low and stable, at 10 per cent.
There are some internal reasons explaining why Muslims do not vote en bloc . Beyond the strategic calculation leading Muslim voters to align with the non-BJP candidate most likely to win, they are themselves divided.
An examination of the profile of Muslims MLAs over the last 20 years reveals a striking trend. While Ashrâfs , or upper caste, Muslims make for roughly 20 per cent of the total Muslim population, they occupy more than 70 per cent of the seats won by Muslim candidates over the period. Interestingly, the ratio increases as the representation of Muslims improves.
The surge in the number of Muslim candidates elected to the Assembly in 2007 and 2012 has largely benefited upper caste Muslims, as compared to other caste groups.
The domination of Ashrâfs has a historical rooting. They tend to be wealthier, richer, more educated, belonging to former ruler families, possessing land and, sometimes, nobility titles. More importantly, they are seen by mainstream parties as the legitimate “rulers” of their own community, more apt to exert patronage on their constituents. This party bias and a phenomenon of elite capture of the political space allotted to Muslims by a small number of Ashrâf families account for their disproportionate representation.
This state of affairs has provoked reactions among the millat . New organisations have emerged in the recent years, which strongly criticise the Muslim traditional leadership, advocate an empowerment of Dalit and OBC Muslims and call for a “power shift” from Ashrâfs to Ajlâfs . They accuse high-caste Muslim representatives of focusing on symbolic and religious issues in order to downplay the socio-economic inequalities persisting within the Muslim community. Among these organisations, the All India Backward Muslim Morcha (AIBMM) of Dr. Ejaz Ali and the Pasmanda Muslim Mahaz of Ali Anwar, both hailing from Bihar, have established branches in Uttar Pradesh. Absent from the electoral competition, these umbrella organisations try to act as pressure groups but their influence remains limited, as they haven’t (yet) been endorsed by any mainstream parties.
In western Uttar Pradesh, however, groups of backward Muslims, whose socio-economic condition has recently improved due to the urbanisation of the region, have been asserting themselves socially and economically, and are now able to convert their newly acquired economic capital into political capital or influence. Backward Muslims in this area tend to support the BSP, which remained ahead of the SP in the last 2012 elections.
Thus, the idea of a Muslim vote, purported by a shared predicaments or a shared religion is a myth.
Finally, there is one other possible explanation for the rise of Muslim representatives, which is that Muslim candidates get greater support from non-Muslim voters than they used to. It is often assumed that majority voters tend to discriminate against minority candidates. Studies conducted in the U.K., continental Europe and the U.S. have demonstrated the penalty of contesting as a visible minority in constituencies dominated by majority voters.
In India however, the “majority voters” i.e. Hindu voters, essentially, do not form a cohesive group, divided as they are socially, economically, and politically. Electoral strategies in Uttar Pradesh, as in other States, are guided by the constitution of local alliances between groups, who club their vote in order to build winning local coalitions. Thus, the BSP grew politically as its core supporters — Jatav Dalits — transferred their support to non-Dalit candidates in general seats. Most parties, who seek to create a winning coalition in each seat where they field a candidate, have since emulated the BSP formula of transferability of vote bank. On the other hand, voters admit that in order to support one’s party, one might vote for a candidate from a different group than their own. Identity politics does not necessarily require symmetry between the identity of voters, candidates and parties. There is a strong possibility that more non-Muslim voters vote for Muslim candidates in order to make their party of preference win. In other words, there is no necessary penalty effect for parties fielding Muslim candidates or “Hindu backlash” as it is sometimes called in the media and on the field.
It is possible that the current BJP campaign in the State, aiming at polarising voters on religious lines, may alter this state of affairs. It is very likely however that the rising path of the BJP may push Muslims to vote even more strategically so, creating difficulties for the saffron party in a significant number of constituencies. Whether the BJP will succeed to consolidate Hindu voters, on one issue or another, remains to be seen.
The lesson of this story is that Muslim voters in Uttar Pradesh are not necessary tributary to institutional reforms to improve their representation. The more cohesive their local vote will be, the more mainstream parties will have to take their demographic weight into consideration and will have incentives to distribute large number of tickets to Muslim candidates. This is not to quash the legitimacy of the claim for specific provisions, i.e. reservations, for Muslims on other grounds than political representation. Descriptive representation would not be enough to improve the social and economic deprivation of Muslims in the State. But strategic voting and local coordination of Muslim voters can, on their own, improve their representation.