What Bengal Just Showed

This article stands on its own. My six-part series on the invisible machines of Indian democracy will continue on May 17, when I’ll look more closely at regional limits. Bengal’s verdict deserves more than a passing mention, so I’m breaking the series’ rhythm and sharing these thoughts now.

The main results are clear. The BJP won 206 out of 293 declared seats in the 294-seat assembly, while the TMC dropped to 81. Mamata Banerjee lost Bhabanipur to Suvendu Adhikari by 15,105 votes, ending her hold on the seat since 2011. Forty-three of her ministers and MLAs also lost. The Governor has dissolved the outgoing assembly. The new BJP government will be sworn in on Saturday, May 9, which is Tagore’s birthday.

The date chosen for the oath is meaningful. The new government wants to be seen not as outsiders, but as bringing back a version of Bengal that the TMC is said to have neglected. The ceremony itself will send a stronger message than any speech. This is how political infrastructure works when it is confident and deliberate.

A bigger question is how the TMC, which won 215 seats in 2021 and stayed in power until 2024, became so vulnerable that it faced a 25-point swing in just five years.

I have three main observations to share, plus a final thought about the present moment.

What the SIR did and did not do

The Special Intensive Revision of electoral rolls in Bengal has sparked more analysis in the past month than any recent administrative process. The numbers are striking: about 89 lakh names were removed in total. Of these, around 64 lakh were marked as Absent, Shifted, Dead, or Duplicate. The other 27 lakh were put Under Adjudication due to what the Election Commission called logical discrepancies. By polling day, only 1,607 names were restored out of 34.35 lakh appeals, a restoration rate of just 0.05 per cent. In Kerala, the rate during routine roll revision was over 98 per cent.

For two weeks, the main debate has been whether the SIR determined the outcome.

The straightforward answer is that it did not.

The Wire published two detailed analyses that are now the main sources on this issue, which I am using for my writing. Their counterfactual model, which adds deleted voters back to the rolls and assigns them to parties based on each constituency’s 2021 vote share, shows that eleven seats would flip back to the TMC. The BJP would drop from 207 to 198 seats, and the TMC would rise from 80 to 91. The overall result stays the same.

This result is important because it challenges the strongest claim that the SIR decided the outcome. Yogendra Yadav wrote in the Indian Express that 27 lakh deletions equal 4.3 per cent of votes cast, compared to the BJP’s 4.6 per cent lead. While the math is correct, it assumes every deleted voter would have chosen the runner-up, which isn’t realistic. The Wire’s counterfactual offers a more careful approach to the question.

The Wire also ran a worst-case scenario. If every deleted voter had supported the runner-up and every new voter backed the winner, then 87 out of the 137 seats that changed hands between 2021 and 2026 could be linked to roll revisions. This is the maximum possible effect, not what actually happened, but what the numbers allow.

I think that the reality is somewhere in between. Eleven seats were definitely affected, and up to eighty-seven could have been. The overall result stayed the same, but the impact was significant.

There’s another aspect to consider. Standard deletions and those made during Adjudication had different political effects. Standard deletions (such as removing names marked as absent, shifted, dead, or duplicate) did not give the BJP a significant per-seat advantage. In fact, these deletions were more common in TMC-held seats. Adjudication deletions were different. The Wire’s detailed study in Mothabari found that 76.8 per cent of these deletions happened in booths with high minority populations. At one polling station with 91.88 per cent minority voters, 385 names were put under Adjudication. The SABAR Institute in Kolkata has done booth-level research that international media have cited. In Nandigram, where Muslims make up 25 per cent of the population, 95 per cent of supplementary-list deletions were Muslim names. In Bhabanipur, where Muslims are 20 per cent, 40.1 per cent of the deletions were Muslim.

This shows what the SIR did at the booth level. It was not the same everywhere. In some districts, the roll revision functioned within procedural norms. In others, the same procedural framework was used selectively, resulting in disproportionate effects.

The targeting did not affect all communities equally. SABAR’s earlier research on unmapped voters (people whose details didn’t match the 2002 rolls) found that Matuas were over-represented among the unmapped, nearly twice the state average in their areas. BJP-supporting groups were also impacted. The effect was stronger in some places than others.

The simplest conclusion is the one The Wire reached: the SIR did not change the overall result. It made a difference in a group of close races. Standard deletions helped the BJP more in seats with tight margins, while adjudication deletions had a bigger effect in areas with many minority voters. The outcome would have been the same without it, but the impact was real.

This is what it means to consider both findings at once.

Why the TMC was vulnerable in the first place

The bigger issue isn’t just whether the SIR amplified the swing, but why the TMC became open to such a swing in the first place.

The best analysis I’ve seen was in the Hindustan Times on counting day. Two political scientists visited a village in Jalpaiguri, between the Jaldhaka river and the Sonakhalo forest. Before they could say anything, a woman asked if they were from a political consultancy representing the TMC. Their interpretation of her question is the sharpest insight into Bengal’s current politics I’ve read.

Their argument is about structure. The TMC’s local organisation hasn’t vanished. Cadres are still present at booths and in neighbourhoods. The machine still exists in form, but its function has changed.

When welfare became universal, and Lakshmir Bhandar reached two crore women as a rule-based entitlement rather than being distributed by party workers, local party members lost their main source of influence. In the past, local party workers decided who got state benefits. Universal cash transfers removed that power. Cadres could no longer selectively distribute benefits.

What the cadre kept was different. They still had the power to control who could participate and who was included or excluded from the rolls. Their role shifted from distributing benefits to acting as gatekeepers.

A woman living in a slum in Titagarh, near north Kolkata, told the HT authors that during the SIR, help with paperwork was taken away from households seen as non-TMC and given to others, regardless of religion. The discrimination was based on political loyalty, not community. The machine was still working, but only for some.

The leadership didn’t focus on internal reform. Instead, they built parallel structures.

Didi Ke Bolo sent citizen complaints straight to a central office, skipping the local cadre who were often the problem. Duare Sarkar brought government services directly to people’s homes, without needing local party workers. Political consultancies ran both programs. The leadership started depending on these new systems, separate from their own party workers, to understand what was happening and to talk to voters.

In 2021, many people said Didi was good, but her team was not; this view was clear to anyone who did fieldwork in Hooghly, Nadia, or North Bengal. By 2026, this feeling had turned into strong anti-incumbency. The parallel structures couldn’t fix it. They weren’t meant to repair the bond between party workers and citizens, only to work around it. But working around a problem isn’t the same as fixing it.

This is what made the TMC vulnerable when the SIR amplified the swing. The party machine had become more about taking than providing. The leadership built new systems instead of fixing old problems. Women’s support faded because Lakshmir Bhandar became a standard benefit in many states, not just a TMC achievement. Crimes against women, from R G Kar Hospital to Sandeshkhali, were handled in ways many voters saw as evasive. The Maa-Maati-Manush slogan, which once reassured investors, lost its meaning as central government schemes slowed and people saw relatives leaving for jobs in other states and abroad.

The party machine depended on one person’s grip on power. When that grip weakened, there was nothing left to support it.

What the BJP substituted

The BJP’s win in Bengal wasn’t like its victories in Madhya Pradesh or Uttar Pradesh, where it relies on a strong grassroots network.

The Sangh has never had deep roots in Bengal. Its shakha network grew from about 580 in 2012 to over 4,000 by 2024, but that’s still far less than the cadre strength built in the Hindi belt over generations. Instead of building this depth, the BJP used its control of the central government to substitute federal institutions for local organisations.

This substitution was implemented using five main tools.

Central armed police deployments neutralised the booth-level intimidation that had favoured incumbents in earlier cycles. The Election Commission’s deployment of 30 roll observers and 8,000 micro-observers in Bengal against four observers and zero micro-observers in Uttar Pradesh, the asymmetric institutional intensity that Yogendra Yadav documented in detail. The SIR itself as the procedural product of that asymmetric intensity. The central government’s freezing of approximately a trillion rupees in central funds to Bengal, including MGNREGS and PM Awas Yojna, which choked the state government’s delivery capacity at exactly the moment when its political vulnerability was deepest. And the gubernatorial office, which was repurposed from a ceremonial function to an instrument of confrontation.

The last of these is worth pausing on. RN Ravi was moved to Bengal as Governor shortly before the election. He came from a five-year tenure in Tamil Nadu, where he had performed exactly this role of confrontation with the state government, holding up bills, delaying budget assent, and refusing to read the standard Governor’s address. As Caravan reported in March, the Governor’s office in opposition-ruled states has often “found itself the locus of opposition to the state government, acting as a BJP mouthpiece and obstructing parts of the Mamata government’s agenda.” Ravi is now in Kolkata at the moment of transition. The new BJP government will take oath on his watch, on Tagore’s birthday, in an assembly he has just dissolved.

These five tools are not the same as the cadre networks the Sangh built in the Hindi belt, or those the CPI(M) has in Kerala, or those the DMK has in Tamil Nadu. They are different; they are federal structures used to compensate for the lack of local depth.

In Bengal, the BJP won power without building the kind of social network that usually leads to lasting governments. The next five years will show if these institutional substitutes are strong enough to fill that gap, or if the party will need to build a real cadre over time.

This is why Pratap Bhanu Mehta’s view in the Indian Express rings true. He called the Bengal win “a remarkable tribute to its unmatched combination of ambition, perseverance, and political ruthlessness.” These are not partisan words; they simply describe what happened. The BJP achieved in Bengal what it had aimed for over fifteen years, using every legitimate and some borderline methods, with a razor-sharp focus that no other Indian party has right now.

What today looks like

But the transition tells us something more, and it does so in real time. Five days after the results, Bengal is already changing, and these changes reveal what the new government has really inherited.

Auto fares that had gone up were rolled back within days. A Tollygunge auto operator in Behala told a journalist, “Yes, they were demanding Rs 5 extra. Today they are polite, charging the regular commuter the actual fare.” Public transport operators, film groups, and traders quickly adjusted to the TMC’s defeat.

The film industry is changing its image almost overnight. The Eastern India Motion Pictures Association, which two TMC-aligned brothers had used to control Bengal cinema for over a decade, cut ties with the TMC and declared itself a “non-political entity” within two days. Actor Parambrata Chattopadhyay, who once had to apologise for working with outsiders, said the federation president had become “a law unto himself, eclipsing both the former CM and the new CM.” The new president, Papia Adhikari, has announced, “the federation’s era of dictatorship is over.” She has promised to end the culture of bans and revive the industry.

Near New Market in Kolkata, hundreds of hawker stalls closed temporarily as traders tried to understand the new power dynamics. Some looked for political protection from the BJP. The Kolkata Police commissioner even had to ban the use of bulldozers in victory celebrations.

Metro projects that had been stuck for years are now moving forward. Just days after the results, the railways ministry approved projects worth Rs 671.7 crore for Bengal. Work on the 366-metre viaduct at Chingrighata, stalled for over a year, will start on May 15. The long-awaited Purple Line is expected to speed up. Land at Esplanade, previously occupied by a market, was cleared overnight.

This is worth thinking about. It’s not just how quickly a new government gets central project approvals or how fast auto operators and film producers change sides. These are familiar patterns in Bengal, as the Times of India noted this morning, going back to May 2011 when Mamata’s win caused a similar overnight shift. The term “Lal-Trinamool” described Left strongmen who switched to the new ruling party. The ideology changed, but the muscle stayed. Now, the same pattern is repeating with a new party in power.

But we should also notice what’s happening on the sidelines, where it’s less visible. This week, Suvendu Adhikari’s personal assistant was shot dead by attackers on bikes in Madhyamgram. A Trinamool worker was killed in Birbhum, and a BJP activist was lynched in Howrah. Trinamool offices have been set on fire or vandalised in Kolkata, Howrah, Asansol, Siliguri, and Baruipur. The usual cycle of post-election violence has started again.

By Bengal’s own difficult standards, this week’s violence has been less severe than in past transitions. Central security forces are still present and have helped prevent trouble in some sensitive areas. But this isn’t normal life. It’s the usual start to a regime change in Bengal, a pattern the state has followed for fifty years under three different parties. Congress started it, the Left Front made it routine, and the TMC continued it. Now the question is whether the BJP can finally break this cycle.

This is the real test Bengal voters have set for the new government. It’s not about the central schemes or faster metro projects; those will happen automatically as part of the federal system. The real challenge is whether the BJP can do what no party has done in fifty years: end the cycle of political violence. If they succeed, the double-engine sarkar will have achieved what earlier party networks could not. If not, only the party in power will have changed, and the same old questions will come back.

We’ll see early signs soon. If the new government stops or excuses revenge attacks by BJP supporters, that will tell us a lot. If syndicate operators who controlled contracts in TMC areas now work under BJP banners, as the auto operators did, that will tell us even more. These patterns should be clear by the end of May.

What this leaves us with

I’ll end with three final observations. These are not predictions.

The regional voluntary machine, which I’ve described as the only real counter to the BJP’s national machine, has now collapsed in two of its three strongholds. Bengal has fallen. Tamil Nadu has shifted to a new party within the Dravidian tradition. Only Kerala remains, where the Left Front and Congress-led front alternate, keeping the voluntary system alive. For the first time since 1969, the Indian Left has no state government. The Trinamool Congress lost both its state and its leader’s seat. The DMK was replaced by a five-month-old party. These are big structural changes. I’ll discuss them more in Part 3 of the series once things settle down.

The parallel-machine idea I introduced in Part 1, as a third type alongside voluntary and state-funded machines, has now become clearer. There are three subtypes: the institutional type, which uses independent bodies that act in line with the ruling party; the consultancy type, which hires private firms rather than relying on party workers; and the federal-substitution type, which uses central agencies, governors, and scheduling powers to compensate for weak state-level networks. The TMC used the consultancy model. The BJP used the federal-substitution model. Both replaced party workers with parallel systems. The democratic impact is different, but the basic structure is the same.

The HT article ends by asking if parties can keep avoiding their internal problems by building parallel structures. For the TMC, the answer is no; the problems came back as electoral losses. The same question now faces the BJP in Bengal over the next five years. Federal-substitution helped them win, but whether it can keep them in power, especially in a state with strong political traditions and Bengali pride, is another challenge.

We’ll get early clues from how the new government balances Bengali cultural pride with the Hindi-belt instincts of its national leaders. The oath ceremony on Tagore’s birthday is the first sign, and it’s a thoughtful one. What happens next will show if this careful approach is just for show or a lasting strategy and whether the new government can finally end the cycle of political violence that three previous governments could not.

A closing note

This article does not address whether the SIR crossed the line of electoral integrity needed for India’s democracy to consider the results legitimate. That question is important and is being debated in the courts and in public discussions.

Speaking as someone with decades of experience in Indian political campaigns, I see the factual question (did the SIR decide the result) and the legitimacy question (was the SIR an acceptable process) as separate. Both matter. The first has a clear answer, supported by the best research. Courts and other institutions are deciding the second, and I’m willing to wait for their judgment.

What I can say for sure is that the different political machines I’ve described in this series are now clearly at work. The BJP’s national voluntary machine, built over a century, hasn’t yet taken root in Bengal. The state-funded machine disappeared when the TMC lost. The parallel machine, especially its institutional form, stepped in to fill the gap. The new government, sworn in on Tagore’s birthday, will start without the party network that earlier governments relied on. Over the next five years, we’ll see if this is a lasting strength or a growing weakness. Bengal will show us.

Part 3 of the Invisible Machines series will be back on May 17, taking a broader look at regional limits, including Tamil Nadu and Kerala. Until then, these are the lessons from Bengal.