Story

India: Misreading the Elections

Bahujan Samaj Party supporters celebrate at a rally in Lucknow in December 2011. It was organised to woo the crucial Muslim and Thakur communities ahead of upcoming assembly election in Uttar Pradesh. Image by Ashok Dutta/Hindustan Times/ Getty Images.

Less than a month before the 2009 Lok Sabha elections, the wedding of a top Uttar Pradesh bureaucrat’s daughter at the Taj Hotel in Lucknow presented senior journalists invited from Delhi with an opportunity to interact with the state’s leading bureaucrats—who are, in Chief Minister Mayawati’s reign, more important than politicians. For a select few celebrity editors, there was even a rare durbar with Mayawati herself, who carefully arrived after the governor had left, presented flowers to the newly married, and proceeded to a barricaded enclosure to meet India’s opinionmakers. I don’t know what the conversation was like, but I saw the journalists’ lips move more than hers.

After the meeting was over, I asked one celebrity TV anchor what he thought the election results were going to look like. He said the Mayawati-led Bahujan Samaj Party (BSP) was very strong, and predicted she could win 50 of the 80 seats the party was contesting. The Delhi media’s awe of Mayawati was at a historic peak; they had taken her prime ministerial ambitions seriously. I told this studio journalist that the buzz in Lucknow was that the Congress could spring a surprise. “No chance,” he said. “They don’t have any organisation. Azharuddin is my friend and he called me to say he needs my help. Even a celebrity like Azharuddin is going to lose!”

Mohammad Azharuddin, once a cricketer, won the Moradabad seat by 50,000 votes, and the Congress increased its vote share from 12.5 percent to 18.25 percent, more than doubling its seats from nine to 21. The BSP won only 20 seats, and appeared to have surrendered the gains it made in the 2007 state assembly elections, when it had won an outright majority with 206 out of 403 assembly seats. The predictions that preceded that election—even those backed by expensive surveys—had done an equally woeful job of projecting the outcome: at her first press conference after returning to the chief minister’s office that year, Mayawati noted that journalists had been complaining about her not giving interviews, and quipped that since they were making their own predictions she did not want to “disturb” them.

India’s second-most important elections—for the Uttar Pradesh state assembly—are upon us again. The Delhi media will again play astrologer, and after the results are announced the channel that got it least wrong will declare victory. But the cracked crystal balls of the studio psephologists are the least of our problems: by the time you read this, the season of wall-to-wall election coverage will be in full swing, and—as with anything that attracts the attention of more than a few dozen broadcast vans—whatever matters of substance are at stake will have been buried under an avalanche of personalities and predictions. You’ll be reading about the “clash of the titans”—Mayawati against Rahul Gandhi; you’ll be told by analysts that Mayawati has her core Dalit base intact, or that Brahmins are moving this way or that, or that the Muslim voter’s inclination towards the Congress is helping the Bharatiya Janata Party polarise Hindu upper-caste votes in urban areas, and that the Rashtriya Lok Dal leader Ajit Singh’s handshake with the Congress is bad news for the Samajwadi Party in western UP, and so on.

But elections are not cricket matches where scores can be recorded ball by ball, nor a stock exchange with definite rise and falls every day. The score is known only once, right at the end. Predicting Indian elections is foolish, but not impossible: it could perhaps be done if election surveys were held in every constituency, but that would be a very expensive exercise. If only the media could eschew its penchant for prediction, which tends to skew all other coverage towards the same superficial register, it would have far more time to examine the realities on the ground, and to unearth the specific local issues and conflicts that truly determine results. The media gives us the impression that little else matters beyond the promises of reservations, “vote-bank politics,” populist grants and loan waivers, and “anti-incumbency.” Actual voters, meanwhile, are already preoccupied with things like the September 2011 floods in eastern UP and the unavailability of fertilisers for farmers in central UP—and politicians are already talking about them as they campaign; but the national media doesn’t seem to have devoted a single story to these subjects.

The problem is not merely that the media does not like stories that don’t fit a statewide generalisation, but also that the minutiae are considered boring. The issues lie in the local, and they don’t always lend themselves to aggregations that could help predict who the next chief minister will be.

Ironically, political parties have a much better idea of what the result is going to look like, because most of them spend heavily on massive surveys conducted in every constituency, which they use to determine details such as which party leaders are most popular with the electorate. At the same time, the political parties are well aware of the media’s propensity for big-picture stories that mesh with the need to proffer predictions, and party leaders have an easy time manipulating the press to spread their preferred narratives in the run-up to elections. In 2007, the otherwise media-shy Mayawati devoted considerable energy to promoting her new "Dalit-Brahmin" alliance, which gave many people the impression that Brahmins were shifting to the BSP en masse. It was a sexy story, and one that we still believe. Yet a post-poll survey showed that just 17 of every 100 Brahmin voters voted for Mayawati. For the most part, the Brahmin voter turned up for the BSP in those constituencies where the party gave the ticket to a Brahmin. But the widespread message that Brahmins were joining the BSP allowed Mayawati to communicate to voters that if Brahmins could join the party, everyone would—which helped sway at least some undecided voters to her side. One of the many nuances that got lost along the way, however, was that the vaunted Dalit-Brahmin alliance only ever involved one of two main Brahmin subcastes—the Kanyakubja Brahmins of central UP, who joined hands with the BSP; the Saryupareen Brahmins of eastern UP were reluctant. The questions that won’t be asked this election linger: what did the Brahmins get by allying with the BSP? And after five years of it, what do Dalit voters make of the alliance?

The general recognition that “caste politics” is crucial in Indian elections—which is not incorrect—nevertheless leads to another interpretative failure. The mechanics of caste are always described by reference to enormous blocs of voters from the same caste shifting from one party to another, which is inaccurate to begin with, and also lends weight to the vocal dismay among liberal political elites that parties are “using” caste in some nefarious manner to win elections. The reality is far more complex; it is equally true that it is, in fact, caste groups that are using political parties, in an effort to increase their representation in power, which in turn provides concrete material returns for their communities on an everyday basis. Furthermore, since Vidhan Sabha constituencies are small, and every vote matters, the shifting preferences of small caste groups take on considerable importance—and these local developments don’t necessarily have any connection to statewide narratives. As a result, almost no attention is paid to groups like the Chaurasias, the peasant OBC community that claims to have four million voters across UP, whose caste profession is the betel trade. When a Chaurasia leader wrote to Rahul Gandhi, his letter was taken seriously enough that the Congress leader called them for a meeting and asked what aid and promises they were seeking. What do the Chaurasias want? Why do they have to use political parties to get access to the government?

An election inaugurates the reign of a new government, but it also marks the end of an old one—whose record is likely to figure rather prominently among voters as they go to the polls. But you will not see many stories that review the performance of the outgoing government, or the perceptions among the electorate about its achievements and failures. It is unlikely that too many journalists will ask citizens whether they feel that law and order in UP has improved in the past five years, in spite of the fact that this was a key reason for the Samajwadi Party’s loss in 2007. Specific issues like these are, instead, gathered up under the broad umbrella of “anti-incumbency”, which, along with “swing”, turns our newsrooms into fortunetellers that employ statistics in lieu of tarot cards, as though math alone can predict the future. In the event that the incumbent party retains power, the resident psephologist—who may have been harping on the swelling signs of anti-incumbency right up until election day—will declare with a straight face that “this time” anti-incumbency did not come into play.

A leading newspaper calls elections “the dance of democracy.” Democracy is about voters, but they play a strangely minor role in the reporting of our elections. The relentless focus on numbers and netas, meanwhile, reinforces the impression that parties are manipulating voters, while the overlooked ground-level details show the various means by which voters attempt to exercise leverage over the parties competing for their affection.

To capture the complexity of all these intricate negotiations would require an entire book, but consider one example: the dacoit Dadua, second only to Veerappan in his fame and exploits, was a Kurmi Robin Hood in the forests of Chitrakoot who spread terror in several districts of Bundelkhand and eastern UP. Until 2002, the Samajwadi Party used to ally with the Kurmis in the region; adding the Kurmi vote to the Yadav and Muslim vote was a surefire formula for victory. But then the BSP won over the Kurmis by allying with Dadua—in exchange for the assurance that the dacoit would be protected from the long arm of the law. For the Kurmis, this arrangement also provided protection from the Brahmins, whose feudal domination had created the conditions for Dadua’s rise, in the first place. But, in 2007, the BSP could not ally with the Kurmis because it was attempting a statewide alliance with the Brahmins. This was just as well: Dalit voters were unhappy with the violence and political dominance of the Kurmis. In 2007, the coalition of extremes between Dalits and (some) Brahmins delivered a victory to the BSP, which also had a spectacular performance in Bundelkhand. But that arrangement may not hold in 2012: many Dalit voters, and not just in Bundelkhand, are unhappy with their Brahmin BSP legislators—and so Mayawati has, in turn, replaced them with new ones.

As I write this, having travelled in UP for the past week, I haven’t met a single person who is not saying it’s too early to tell what the result will be like, because many candidates haven’t been declared, and the choice of candidates (caste, religion, personal popularity, past performance) is more important than party affiliation. During the same week, the Delhi media came up with yet another survey that predicted party-wise seats.