It takes a heck of a lot to hack an EVM

share this article

Rather than blindly alleging that an electronic voting machine has been rigged, it pays to find out about its inherent technical and administrative safeguards — so that you ask the right questions.

The EVM and VVPAT make for a strong team that provide each other with the safeguard of corroboration. | R.V. Moorthy

I keep hearing about random conspiracy angles about EVMs and their tamperability. One of them suggests bizarrely that some telecom towers can manipulate them via wireless services. Apparently, this kind of accusation was made initially by Congress leader Arjun Modhvadia, in the wake of the polling in the recently concluded Gujarat Assembly election — he claimed that he could see a bluetooth device called ‘ECO 105’ listed in the connections available on his mobile phone, which, he presumably was suggesting, meant that the EVM was being hacked.

This by itself is so ambiguous it cannot possibly be taken as evidence of hacking. I can set up a bluetooth connection on my PC or my mobile phone and call it ‘Narendra Modi's personal phone’ or ‘Donald Trump's nuclear code button’ and keep it open. Just because anyone in the vicinity can access it if I keep it open doesn't mean that they have access to Modi’s personal phone or Trump’s nuclear codes.

 

The EVM is a single-chip unit with a pre-programmed set of instructions that are readable only by specific devices via a wired connection. One can override these by replacing the chip of the EVMs after dismantling the unit altogether, but the control unit records keystamping if its a Generation 2 machine and doesn’t allow functioning at all if its a Generation 3 machine. So, any tampering, if it has occurred at all, can be easily detected by officials. Besides, all party representatives are allowed a testing of the EVM before it is randomly deployed to booths and the EVMs are also tested at the polling stations. After polling, EVMs are locked with seals signed by the representatives. So, party representatives can also verify if their seals have been broken, during counting.

Then, there was this report by Jantakareporter which was retweeted by AAP leaders and Prashant Bhushan. This report hinted that there could be a quid-pro-quo between the EVM manufacturers and those involved in the alleged scam — there was a similar pattern in the venture-funding investments of the company that supplied chips for the EVMs and a firm that the Gujarat State Petroleum Corporation strangely took on as partner in a loss-making and allegedly fraudulent gas-exploration project. The proposition is that the chip-manufacturer could override the EVM code (supplied by Bharat Electronics Ltd.) by inserting code that is favourable for a particular party.

 

It is impossible to know the order in which the candidates are chosen by a particular party beforehand at the time of manufacturing a chip.

While the ownership matter is something to look into, the rigging at the manufacturer level idea fails by a simple logical test — all EVMs have candidates on the machine arranged by two orders: 1) Prioritising by national/state and other parties/independents and then 2) Arranging by the name of candidates in alphabetical order.

The data on votes are stored on candidate name and not on party, with the ECI simply tabulating it later on party-wise basis. It is impossible to know the order in which the candidates are chosen by a particular party beforehand at the time of manufacturing a chip. And, what’s more, EVMs are randomly allocated to States and constituencies.

What logic can be written at the manufacturing level to suit a particular party beforehand, I wonder?

Instead of outlandish theories about EVMs, it is worthwhile to find out what are the safeguards that the ECI has set for the EVMs to be infallible — both technological and administrative. And then test if those safeguards are being violated in any way to allow rigging in favour of a particular party. There is no gobbledygook here, just simple reasoning.

 

Thus far in Gujarat (and elsewhere), we have heard about complaints about EVMs and not specifically about rigging. Those complaints primarily talk about failure of EVMs to register keys or simply not responding to voting or VVPAT not printing the vote properly. These are cases of EVM malfunctioning which is but expected — when 24,000-odd machines are used, a few might well actually display glitches or fail. The ECI has claimed that it has promptly replaced such machines and used replacements that completed the process of polling. It may be pertinent to wonder whether the ECI really did so or was it lackadaisical in doing so — that is a valid question that needs to be asked to hold the ECI to account.

Secondly, the use of the VVPAT should provide us with a corroboration of the voting results tabulated on the EVMs (as it did in the Uttar Pradesh elections where VVPAT was deployed selectively rather than universally in Gujarat).

Instead, all we hear are outlandish conspiracies that suggest that a) EVMs are flawed as they come with manufacturing defects that favour one party, b) the EVMs cleared by the ECIs and deployed in booths can be manipulated with a bluetooth/wifi connection, and so on. This seems a case of people who have made up their minds without applying any thought to understanding the EVM use process.

Lastly, people talk about the return to paper ballots as if it is a perfect solution. They distrust the ECI’s administrative safeguards for the EVMs and want a return to paper ballots, which have looser safeguards because of the balloting system and even if those are in place, there is the age-old possibility of mass rigging through booth capturing, ballot-box stuffing, all of which was quite a thing in the past in many States and not just fantasy (as many of the complaints against the EVM are).

Why you should pay for quality journalism - Click to know more

share this article
This article is closed for comments.
Please Email the Editor