The Election Commission of India’s (ECI) special intensive revision (SIR) in West Bengal has not produced one uniform pattern of deletion, but two. One runs through the Muslim-dominated border belt, where scrutiny through the ‘under adjudication’ process appears to be driving a large share of exclusions. The other runs through the urban and industrial belt, where deletion is being driven far more heavily through the draft-roll route. Together, they reveal a political and administrative operation that does not appear to be random.Channels of exclusionThe total burden of deletion is spread across three channels: draft-roll deletions, deletions through the final roll under Form 7, and those marked ‘found not eligible’ under adjudication. Statewide, draft-roll deletion has been the biggest channel of exclusion. But the second largest is channel three – ‘found not eligible’ after adjudication, and that is where the communal and geographic pattern begins to solidify.This phase is also qualitatively different from the first stage of deletion. Those who were in the list of those ‘found not eligible’ by judicial officers were not simply absent names, but living people who had entered the process, appeared before judicial officers, and were then struck off after failing to satisfy documentary linkage demands tied to the 2002 electoral roll. In this sense, theirs is not just a deletion but an adjudicated disenfranchisement.To follow what is being said below, click on the two dashboards here and here, to check the data for yourself.Muslim population and the under-adjudication pipelineThe most striking trend is the link between the Muslim population and ‘under adjudication’ scrutiny. Constituencies with a higher Muslim population show a strong positive relationship with the share of voters pulled into the ‘under adjudication’ process. The same seats also show a strong rise in the number ultimately marked ‘found not eligible’ under adjudication. In plain terms, the more the population of Muslims in a constituency, the more likely it is to be dragged into the ‘under adjudication’ pipeline, and the more likely it is to see deletion through that route.That pattern is one of the strongest relationships in the data.It also sharpens the political meaning of what is happening in districts with high Muslim populations like Murshidabad and Malda, where the raw volume of ‘under adjudication’ rejections points to a disproportionate purge of Muslim voters, many of whom were called for hearing despite providing stable generational land, birth or legacy records. Linguistic and spelling discrepancies in Arabic- and Urdu-origin names across generations may also have pushed Muslim voters more deeply into hearings they were then poorly placed to survive.The Murshidabad-Malda BeltThe effect is clearest in the districts of Murshidabad and Malda, and extends into parts of Uttar Dinajpur, North 24 Parganas and South 24 Parganas. The names at the top of the list are Samserganj, Lalgola, Bhagabangola, Raghunathganj, Farakka, Suti, Mothabari and Jangipur, each of which stand out with very high volumes of voters marked ‘found not eligible’ under adjudication. All these constituencies have a high Muslim population.Metiaburuz stands out as the major urban outlier, with a very large number of voters struck off through the same route. Sujapur in Malda is another major site of under adjudication exclusion. These examples strengthen the argument that the Muslim belt is not just seeing higher scrutiny in theory but it is producing large ground-level exclusion counts.This matters because it changes the language of what is happening. In these seats, deletion is not merely an administrative clean-up of the rolls but is tied to adjudication and is thus being routed through a scrutiny process that lands hardest in minority belts.The border, refugee and erosion beltThe borders reinforce the same conclusion. Border constituencies show much higher ‘under adjudication’ exposure than non-border seats. They also show a significantly higher share of deletion coming through the ‘under adjudication’ route. But they do not show the highest draft deletion percentages. Once again, the border belt appears to be an adjudication-heavy zone rather than a metro-style deletion zone.Two additional layers make this border story harsher. The first is the crisis among Scheduled Caste refugee communities, especially Matua and Namashudra populations in frontier districts, for whom maintaining continuous documentary linkage back to 2002 is far harder than it appears on paper. Seats like Krishnaganj and Bagda show how high rejection rates during the adjudication process can fall on border-zone Scheduled Caste populations living with the long afterlife of displacement and insecure documentation.The second is river erosion. In districts like Malda and Murshidabad, where villages have repeatedly been displaced by the Ganga and Padma, the demand for localised documentary continuity becomes especially punishing. Bhagabangola and Farakka reflect this vulnerability. When voters uprooted by river erosion are asked to produce stable linkage documents from an older geography that no longer exists in the same way, natural disaster begins to function as a bureaucratic ground for exclusion.The urban draft-deletion shockAnd yet the story does not end there.Although Muslim-heavy constituencies dominate the ‘under adjudication’ side, they do not dominate the highest rates of draft-roll deletion. That distinction is crucial. A comparative analysis of the draft deletion or absent, shifted or dead/duplicate (ASDD) list and the list of those deleted after having been placed ‘under adjudication’ shows that Muslim concentration is strongly linked to ‘under adjudication’ exposure and adjudication-based deletion, but not to the highest raw draft deletion percentages. In fact, the worst draft-roll shocks are concentrated elsewhere, the urban belt.Kolkata North, Kolkata South, Howrah and Paschim Bardhaman emerge as the centres of a very different deletion regime. Here, the defining pattern is not unusually high adjudication, but unusually high draft-roll deletion.Constituencies like Jorasanko, Chowrangee, Howrah Uttar, Kolkata Port, Ballygunge, Shyampukur, Kashipur-Belgachhia and Beleghata have some of the highest deletion percentages in the state. But the share of deletion coming through the under adjudication route in these seats is comparatively low.That means the urban deletion shock is not primarily being driven by voters being found ‘not eligible’ under adjudication. It is being driven through the ASDD list.Urban India, rural India, two different risksThe urban-rural split tells the reverse story. Wholly urban constituencies show the highest rates of draft deletion and the highest overall contraction. Yet these same seats have the lowest ‘under adjudication’ share in total deletion. Urban seats are not where the ‘under adjudication’ machine is most dominant. They are where the ASDD voters were marked and removed from the electoral rolls.Migration patterns sharpen that reading further. In-migrant constituencies show much higher deletion and contraction than out-migrant constituencies. Out-migrant seats, by contrast, tend to be less deletion-heavy on the draft side but somewhat more ‘under adjudication’-driven. The implication is that voter mobility is being treated differently depending on geography. In the city and industrial belt, mobility seems to translate into vulnerability to draft deletion. In the border-rural belt, vulnerability appears to be channelled more through scrutiny and adjudication.The caste factorThe caste story is different, and politically revealing.Granular analysis of SIR data does not show any strong positive relationship between Scheduled Caste (SC) concentration and the volume of those ‘found not eligible’ after adjudication as a whole. Nor does it show SC-heavy seats as the epicentre of deletion. In fact, the relationship runs the other way. Constituencies with a higher SC population tend to show lower deletion percentages and lower overall contraction. They also show lower Form 7 deletion. This does not mean SC-heavy seats are not affected. Many of them still show a substantial mix of draft deletion and ‘under adjudication’-based deletion, but they are not the sharpest red zones in the state.That distinction matters in a state where political narratives around exclusion are often flattened into one undifferentiated complaint. The data does not support a lazy story but a harsher one. Different communities and different geographies are being hit through different mechanisms.At the same time, the SC crisis becomes far sharper in border geographies and refugee-origin populations where documentary continuity is weakest. This makes the SC story more specific, but not less serious.Judicial variance and the politics of adjudicationThe data also raises another question: judicial arbitrariness. There is sharp variance in rejection rates across constituencies, with some mixed or low-Muslim seats showing very low ‘under adjudication’ rejection and others showing extremely high rejections. The implication is that the threshold for proving eligibility may not have been applied evenly. If that reading holds, then the issue is not just harsh rules. It is an unequal application of harsh rules.That matters because once the system enters the language of adjudication, it acquires the appearance of neutrality. But if the outcomes vary wildly by demography and geography, then adjudication itself becomes part of the political question.The inequality becomes even starker at the appellate stage. The tribunal process has exposed a disturbing asymmetry in who can realistically fight back and who cannot. Those with lineage, legal networks, political standing or public visibility appear to have a vastly better chance of clearing their names than ordinary workers and the rural poor. The promise of equal citizenship begins to collapse when the right to vote can only be reclaimed by those with social capital, legal assistance and access to public attention.For the working poor, this ordeal is also economic attrition. Every trip to a tribunal costs money, time and wages they do not have. For someone living on daily earnings, repeated travel, paperwork and lost workdays turn the fight for voting rights into a punishing financial burden. The right to vote survives in theory, but in practice the state begins to price the poor out of their own citizenship.Two regimes of voter exclusionMurshidabad is the clearest example of a ‘under adjudication’-dominated deletion belt. There, a majority of the total deletion burden comes through the ‘found not eligible under adjudication’ route. Malda shows a similar, though slightly less extreme, pattern. Uttar Dinajpur sits somewhere in between, with draft deletion and under adjudication deletion both playing major roles. By contrast, Kolkata North and Kolkata South are overwhelmingly draft-deletion districts. The same, though to a lesser extent, is true of Howrah and Paschim Bardhaman.Thus in Muslim-heavy constituencies, especially along the border belt, deletion is being disproportionately routed through a scrutiny framework that carries its own burden of suspicion and contestation. In the metro belt, deletion comes in a broader sweep through the draft roll. The result is a map of exclusion that looks administratively diverse but politically coherent.A pattern without an explanationThe hard truth is that deletion in SIR in West Bengal is not distributed evenly but follows a pattern.One cluster of constituencies is made to pass through the gate of ‘adjudication’. Another is cut down through regular process. One can call these technical categories, but the effect is social, geographic and political. The people most affected are not scattered at random across the state but are concentrated in identifiable belts.This turns the exercise into more than an administrative process and into a political map of exclusion.It shows a Muslim-heavy border belt being pulled deeply into the ‘under adjudication’ process even after most voters showed remarkable progeny linkage in the initial revision phase. It shows an urban-industrial belt being hammered by draft deletion through un-mapping. It shows border SC communities facing a distinct documentary crisis. It shows erosion and displacement being folded into eligibility judgments. And it shows that the story of voter exclusion in West Bengal cannot be reduced to a single number or one method. It was a layered exercise unlike other states and this is precisely what makes it more serious.