A close election is much more likely to be the target of small-scale EVM tampering than a landslide

Reactions to the Gujarat elections largely broke along ideological or party lines. But among journalists who pride themselves on being “neutral”, a common theme emerged: whatever these results mean, it is time to stop complaining about Electronic Voting Machines.

Some suggested that the Bharatiya Janata Party’s narrower than expected-victory constitutes disproof of any EVM tampering:

Others, before the results were announced, used the occasion of election day to remind us that concerns about tampering are risible:

The first claim – that a close election disproves EVM tampering – is a category error. The outcome of an election has no bearing on specific accusations of electoral malpractice. It proves absolutely nothing.

Such an argument rests on a number of assumptions, all of them faulty: that EVM tampering is centrally coordinated; that it is distributed across the state. That those responsible would use it to guarantee a landslide victory. In effect, the suggestion is that most or a vast number of EVMs are tampered, or none at all.

This runs contrary both to seven decades’ worth of electoral malpractice in India, and to the specific concerns raised by most EVM-sceptics. EVM-tampering at a state or nation-wide scale, so as to effectively script an election outcome, is indeed wildly implausible. But “rigging” in India has always tended to be small-scale, at the level of the individual booth or district or seat. Outside Jammu and Kashmir, claims of whole state or national elections being rigged are virtually non-existent.

Learning from experience

Look at the traditional forms of malpractice. Leave aside those that are practiced by most parties and, thanks to our moral cynicism, no longer controversial: spending beyond the limit, bribing voters, using government resources (in another era, Indira Gandhi being found guilty of the last sparked the Emergency). Booth capturing was once common, but it could account for, at the most, individual seats. Probably the most prevalent in contemporary India is electoral roll manipulation. In the years leading up to an election, parties try to have names fraudulently deleted or added, so as to subtly alter the social profile of the electorate for their benefit. This, too, is local and small-scale.

If EVM tampering does occur, it is far more likely to occur at this scale. And specific allegations of tampering generally target individual EVMs or polling stations, not entire elections. A close election is, in fact, much more likely to be the target of small-scale tampering than a landslide because a single machine could swing the result.

What of the argument that the entire business of EVM suspicion is silly and counterproductive? This is, for one thing, contrary to the Supreme Court’s view of the matter. In 2013, the SC ordered the use of Voter-Verified Paper Audit Trail or VVPAT machines, which print a record of each vote to ensure that the voter’s choice has been accurately recorded. According to the Court, voters’ confidence in EVMs requires a paper trail, “an indispensable requirement of free and fair elections”.

Baijayant “Jay” Panda, the Biju Janata Dal MP, has dismissed tampering allegations as a “time-honoured tradition by losing candidates and parties”, and likened the adoption of EVMs to motor vehicles replacing horse-drawn carriages. But if EVMs are inherently superior and more advanced, why do the vast majority of advanced democracies not use them? Germany, a country that can hardly be accused of technophobia, abandoned EVMs in 2009, citing the difficulty of verifying their reliability without specialist technical knowledge.

The most prominent EVM-sceptic among Indian politicians is actually now a BJP spokesperson: GVL Narasimha Rao. He started VeTa, a non-governmental organisation that promoted the withdrawal or reform of EVMs after the 2009 Lok Sabha elections that the BJP lost, and wrote a book, Democracy at Risk! Can we trust our electronic voting machines? with endorsements from LK Advani and Chandrababu Naidu (In case you were wondering, his answer was an emphatic No). Rao and his colleagues were responsible for the petition that led the Supreme Court to order the use of VVPAT.

Being neutral – and fair

A close election result tells us nothing about EVM accuracy; nor can all EVM scepticism be dismissed as partisan whining. Why then do so many journalists attack EVM sceptics? It may be a response to the immense pressure to appear neutral in an environment when, especially on social media, journalists are constantly harassed and abused for their alleged bias. But, in this case, journalists are performing rather than embodying neutrality. Genuine neutrality means judging an issue on the merits – on the basis of the facts, logic, and ethical imperatives – and not ridiculing legitimate questions in favour of a lazy celebration of Indian democracy. In fact, we badly need “neutral” journalists to give such questions a fair hearing.

The legitimacy of our electoral system is always hard-won and provisional. It is not enough to assume that the Election Commission is faultless. It needs to be subjected to continued scrutiny by parties, citizens and the press. Competitive politics is often the best check on electoral malpractice. Gujarat, in which one party is far more mobilised and socially rooted than the other, is more vulnerable than most states.

Examined on the merits, the Gujarat elections actually constitute a step forward for the credibility of EVMs. Paper audit trails, whose introduction had been delayed due to the Centre not providing the funds, was used throughout (in Uttar Pradesh, by contrast, only 20 constituencies had VVPAT). At one station in each constituency, the VVPAT totals were counted, and no discrepancies found. This comparison of VVPAT and EVM totals needs to be repeated, at much larger scale (the Aam Aadmi Party proposes 25% of stations), in every election. Then we may be able to say that our voting system is credible. But the process of testing and improving it must go on.