A functioning democracy depends on public confidence in the integrity of the voting process. In India, the introduction of Electronic Voting Machines (EVMs) replaced paper ballots, ending issues like booth capturing and invalid votes. However, as digital systems become more common, concerns regarding transparency, source code security, and physical verification have made voting technology a subject of ongoing debate.
Historical Context: From Paper Ballots to the VVPAT Era
For decades, India used paper ballots, which were vulnerable to manual tampering, invalid markings, and physical seizure of ballot boxes. To address these issues, the Election Commission of India (ECI) partnered with state-run electronics manufacturers (BEL and ECIL) to develop Electronic Voting Machines. First used experimentally in Kerala in 1982, EVMs were rolled out nationwide for the 2004 general elections.
To address concerns that voters could not verify how their digital votes were recorded, the ECI introduced the Voter Verifiable Paper Audit Trail (VVPAT) in 2013, following a Supreme Court directive in Subramanian Swamy vs. ECI. The VVPAT prints a paper slip showing the voter's choice, which is visible behind a glass window for 7 seconds before dropping into a sealed storage box, providing a physical backup for verification.
What is Right vs. What is Wrong
| What is Right (EVM Design Strengths) | What is Wrong (Verification Concerns) |
|---|---|
|
• EVMs are stand-alone machines, meaning they lack internet or network connectivity, preventing remote hacking. • The microprocessors are One-Time Programmable (OTP), preventing software modifications after manufacture. |
• The ECI's low audit rate: only 5 random booths per assembly segment are counted to verify paper slips against digital tallies. • The proprietary nature of the source code, which prevents independent security audits of the software. |
| • Incorporating the VVPAT system, creating a physical paper record for auditing. | • The lack of public transparency in the design and distribution of the Symbol Loading Units (SLU) used in elections. |
🗳️ Statistical Sampling Math
Statistical sampling shows that auditing 5 random booths per assembly segment provides a high probability of detecting systematic software tampering. However, increasing this to 10-15% would address public concerns without causing major counting delays.
The Audit Debate and the Supreme Court Ruling
In April 2024, the Supreme Court of India ruled on petitions seeking a 100% manual count of all VVPAT paper slips. The Court declined to mandate a complete paper recount, stating it would slow down the counting process, delay results by several days, and potentially introduce human counting errors. However, to increase transparency, the Court ordered that the Symbol Loading Units (SLU) be sealed and stored for 45 days after the election, and allowed candidates who finish second or third to request a technical audit of the EVM microchips in up to 5% of the polling stations per constituency.
Table 5.1: Comparative Electoral Verification Standards
| Country | Primary Voting Method | Paper Audit Trail Status | Audit / Counting Policy |
|---|---|---|---|
| India | EVM (Direct Recording Electronic) | Mandatory VVPAT (Since 2017) | Manual count of 5 random booths per assembly segment |
| Germany | Paper Ballots (Electronic banned in 2009) | 100% Physical Paper | 100% manual hand count on election night |
| United States | Varies (Optical Scan / DRE) | ~80% have Voter-Verified Paper Trail | Post-election audits of varying percentages by state |
| Brazil | Electronic Voting Machines (DRE) | No Paper Trail (Fully Digital) | Parallel testing with dummy votes on election day |
Figure 5.1: Counting Time Scenarios: EVM vs Paper Recounts (Estimated Hours)
Visualizing why 100% manual paper recounts delay results compared to stand-alone EVMs.
