The challenge for SIR is to improve electoral accuracy without making eligible citizens prove their right to vote – Sabir Nishat
When the verifier gets flagged
There is a profound institutional irony when an administrative system designed to verify voters flags an election commissioner for a mismatch in his own name. Election Commissioner Sukhbir Singh Sandhu was among those served notices during Delhi's Special Intensive Revision (SIR) after his present electoral entry did not match an earlier intensive-revision record. His details were subsequently verified and his name retained. Former Deputy Prime Minister LK Advani and Foreign Secretary Vikram Misri were among other prominent electors caught up in the exercise.
The significance of these cases is not that prominent citizens received notices. It is what they reveal about the scale and architecture of the exercise.
Delhi's draft roll contains about 97.53 lakh electors. Of these, 33.13 lakh were identified for notices - 19.33 lakh for what the Election Commission calls "logical discrepancies" and 13.79 lakh because their entries could not be mapped to the 2002 roll. The Supreme Court, on September 22, questioned whether notices were being issued mechanically and sought greater clarity on the "logical discrepancy" category.
A legitimate objective, a difficult execution
There is a straightforward case for periodically cleaning electoral rolls. Deaths, duplicate registrations, migration and changes of residence can make voter databases inaccurate. The Supreme Court in May upheld the ECI's authority to conduct Bihar's SIR, holding that the exercise was within its constitutional and statutory powers and that electoral-roll integrity was a legitimate objective.
The question, therefore, is less whether electoral rolls should be verified than how verification is conducted and who bears the cost of correcting the database.
Delhi provides a revealing example. Automated checks can flag differences in names, parents' names, ages and family relationships. But historical electoral records are not pristine databases. Names are spelt differently, initials change, addresses shift and family relationships may be recorded inconsistently.
A failed match with a two-decade-old roll can therefore reveal a limitation in the State's data rather than anything about a citizen's present eligibility.
Yet once a discrepancy is generated, the uncertainty becomes a compliance burden for the voter.
The unequal cost of compliance
That burden is not socially neutral.
A notice may be formally identical whether it reaches a senior civil servant or a daily-wage worker. Their capacity to respond is not. Verification requires time, documents, transport, digital access and familiarity with administrative procedures. For one citizen, attending a hearing may be an inconvenience. For another, it can mean losing a day's wages or travelling to another district or state to locate an old document. Research published in the American Political Science Review illustrates the problem. A field experiment involving 2,306 migrants in Delhi and Lucknow found that 98 per cent of eligible respondents wanted local voter registration. Providing assistance at home to complete and submit registration documents increased registration by 24 percentage points and turnout in the next election by 20 percentage points.
The implication is important: bureaucratic friction can affect electoral participation independently of political interest.
Bihar's SIR exposed similar difficulties. Field reports documented workers who possessed Aadhaar, voter cards and MGNREGS job cards but found that these did not necessarily correspond to the documents initially prescribed for verification. Migrant families also faced practical questions over how husbands and sons working in Delhi, Punjab or elsewhere could return to complete the process.
The Supreme Court recognised this structural problem in its Bihar judgment, noting that the claims-and-objections mechanism could disproportionately burden "migrant workers, rural populations, and socio-economically disadvantaged groups". It directed institutional assistance through Booth Level Agents and other mechanisms as a corrective.
The State's own research raises a warning
The Election Commission's training and research institution, IIIDEM, has also highlighted the vulnerability of migrants, the urban poor, first-time voters and remote communities to complex registration procedures, stringent documentation, fixed registration centres and excessive dependence on digital systems. This points towards an important principle: if the State possesses the database and the technology to identify discrepancies, it should also shoulder a substantial part of the burden of resolving them.
A voter should not effectively have to prove that the State's historical record is wrong.
Political storm over the ECI
The SIR debate has now become inseparable from a larger dispute over the functioning of the Election Commission.
The Congress and other Opposition parties have intensified their criticism, citing an Indian Express investigation that reported at least 14 instances over 10 months in which Election Commissioners Sukhbir Singh Sandhu and Vivek Joshi recorded objections to decisions concerning electoral rolls, voter registration and the Commission's IT systems.
The report has raised questions about centralised control of electoral-roll data and the relationship between the CEC and the two Election Commissioners. The Commission has said differing views are a normal part of institutional deliberation and that SIR decisions were ultimately unanimous.
The Opposition has nevertheless demanded action against CEC Gyanesh Kumar, with some leaders calling for his removal and even stronger measures. The Cockroach Janata Party (CJP), led by Abhijit Dipke, separately demanded that Kumar resign within 48 hours and threatened protests.
The BJP, meanwhile, has defended Kumar and the Commission, arguing that differences during deliberations demonstrate that the institution allows debate rather than functioning through unilateral decisions. BJP spokesperson Sambit Patra said the SIR process was ultimately unanimous and pointed to the Supreme Court's validation of the Bihar exercise.
These competing claims should be distinguished from the underlying administrative questions. Internal disagreement, by itself, does not establish that SIR is unlawful. Equally, a final unanimous decision does not by itself resolve questions about how that decision was reached or whether safeguards were adequate.
What SIR must get right
The central test should therefore be neither political approval nor political opposition to SIR.
It should be whether the exercise improves the accuracy of electoral rolls without making legitimate voters carry a disproportionate burden for correcting administrative discrepancies.
Several safeguards follow naturally.
First, every notice should clearly state the precise discrepancy and the documents or evidence required to address it.
Second, machine-generated flags should receive meaningful human scrutiny before a citizen is asked to respond. The Supreme Court has already raised this concern in Delhi, asking whether additional filters are required.
Third, the process should offer multiple modes of response - online, by email or WhatsApp where appropriate, and through accessible physical channels - while ensuring that digital systems do not become another barrier. Fourth, migrants, elderly voters, persons with disabilities and economically vulnerable citizens need proactive assistance rather than merely a notice and a deadline.
Finally, the ECI's internal decision-making must remain transparent enough to sustain public confidence. The Commission is a constitutional institution; its credibility depends not only on the legality of its final decisions but also on the fairness and explainability of the processes that produce them.
SIR can serve the legitimate purpose of keeping electoral rolls accurate. But accuracy has two sides. Preventing an ineligible person from remaining on the roll matters; preventing an eligible citizen from being wrongly excluded matters just as much.
The ultimate measure of an electoral verification exercise should therefore be whether it makes the roll more accurate without making the franchise harder to exercise.
(The writer can be reched at sabirnishat554@gmail.com)