Notes on the religious demography of Punjab and Bengal

India got partitioned in 1947 and the two provinces that got split were Punjab and Bengal. But the religious demography of the two provinces took different routes.

Today the Muslim population in Indian Punjab is 1.6%, while the muslim population in Indian Bengal is 27.1%. So while there was huge emigration of Muslims from Punjab, there was less so in Bengal.

There are still districts in Indian Bengal today where Muslims are in a majority

      • Murshidabad : 66%
      • Malda : 51%
      • Uttar Dinajpur : 50%

Nothing like it in Punjab where they are in single digits across the state. This is notwithstanding the fact that back in 1947, Muslims were upwards of 40% in the districts of Amritsar, Ferozepur, Gurdaspur and Jallandhar – all of which are in Indian Punjab today.

Punjab started with a clean slate. On either side of the border. Whereas Bengal remained complex demographically, with significant Hindu population in East Pakistan and significant Muslim population in West Bengal.

But the trends of the minority population on either side of the border in Bengal, have been radically different since 1947.  In Bangladesh / East Pakistan, Hindu population has reduced significantly since the 1950s. A decline from 22% in 1951 to 11% in 2015

Hindu population in East Pakistan / Bangladesh

1951 22.05%
1961 18.50%
1974 13.50%
1981 12.13%
1991 10.51%
2001 9.20%
2011 8.96%
2015 10.70%

In sharp contrast in Indian Bengal, the minority population has thrived, with a significant increase since 1951. Here are the numbers –

Muslim population in West Bengal

1951 19.85%
1961 20.00%
1971 20.46%
1981 21.51%
1991 23.61%
2011 25.25%
2011 27.01%

But it’s also interesting to see the demographic evolution of the provinces before Partition. We tend to think that the demographic mix changed decisively in 1947 (more so in Punjab than Bengal). But that doesn’t mean there weren’t major changes in late 19th / early 20th century.

Let’s take a look at the religious composition for Punjab as a whole (both Indian and Pakistan portions) before 1940s. We see a declining Hindu population in Punjab as we move from the 1880s to 1940s. This decline clearly predates the partition and even the idea of Pakistan

Religious demography of Punjab : 1881 to 1941


Notice the sharp decline in Hindu population from 43% in 1881 to 29% in 1941. But this drop is not accompanied by a rise in Muslim numbers. Rather we see a near-doubling in the Sikh % between 1881 and 1941. One is not sure if this was indeed driven by conversions, or whether Sikhs became more conscious of their distinct identity and separateness from Hindus in those six decades, leading to a surge in Sikh census figures.

When we look at East Bengal pre-independence, we see a similar story. The demographic change long predated the partition, though it was less drastic than in Punjab. Please note these numbers are for East Bengal and not united Bengal.

Hindu pop in East Bengal : 1901 to 1951

1901 33.00%
1911 31.50%
1921 30.60%
1931 29.40%
1941 28.00%
1951 22.05%

I found these numbers fascinating, The purpose is not to emote people, but to understand the long run trends and the underlying demographic reality of these provinces.

We tend to think of partition as a “discrete” event. But demographic change long predates Partition, and it continues long after the Partition particularly in Bengal (both West and East). Less so in Punjab, as Punjab was rendered homogeneous on either side by Partition violence.

References : 

The numbers on Punjab demography were sourced from this very fine paper –

(The author tweets @shrikanth_krish)

24 thoughts on “Notes on the religious demography of Punjab and Bengal”

      1. @Razib Khan
        I am not a population modeling expert; nevertheless here are the numbers.

        Assuming the current decadal growth rates in West Bengal hold, we have (1, 2):
        Decadal growth rate of Hindus, D(H) = 10%
        Decadal growth rate of Muslims, D(H) = 23%
        According to current population ratio, we have T(H) and T(M) equal to 70 Hindus and 30 Muslims respectively out of a population of 100.

        After 80 years, we have:
        T(H) = 70*1.1^8 = 150
        T(M) = 30*1.23^8 = 157
        T(M) > T(H) at the end of the century.

        This is an approximate model, but it is very likely that Muslims will be the largest group in West Bengal at the start of the next century. Frankly, all this does not matter as I believe they will remain patriotic citizens of India. I would appreciate if you can correct anything that I did wrong here.


        1. Assuming the current decadal growth rates in West Bengal hold

          This is an approximate model, but it is very likely that Muslims will be the largest group in West Bengal at the start of the next century.

          this is the problem. you assume a model holds, and say that this is likely. it’s likely IF THE MODEL HOLDS.

          it seems plausible but any inspection of linear models shows they are very weak in terms of track record. to give an example, Catholics in n Ireland generations ago presumed they’d have a massive demographic advantage in 2020. that didn’t happen. why? the catholic birthrate collapsed.

          anyone with familiarity with demographic modeling’s past history knows that you always need to bake in huge uncertainties into anything beyond 1 generation.

          google Israel and its unpredictable path (haredi fertility and persistence was unanticipated, so it has prevented the demographic timebomb people were worried about in the 1990s)

    1. the drop off in 1946- 1947 was a one time event – partition
      since then the muslim population has been increasing steadily

      1. yes obviously. that’s at the link. my point is i didn’t know there was so much migration of muslims to the east. I’ve never met any of these ppl. perhaps they are at the border

  1. Srikanth, can we speak offline?

    Is this (which Razib also cited) the source of your West Bengal population demographic breakdown by religion?

    Razib, a lot of nonmuslims moved to West Bengal. This had a denominator affect that lowered the percentage of muslims living in West Bengal.

    A lot of muslims from Bangadesh also moved to India. Could many of these be Indian muslims who moved to Bangladesh and then moved back to India?

    A lot of middle class people who “lived” in West Bengal had a 2nd home or ancestral village in east Bengal.

  2. Razib, a lot of nonmuslims moved to West Bengal. This had a denominator affect that lowered the percentage of muslims living in West Bengal.

    ah, that is what it is!


    A lot of middle class people who “lived” in West Bengal had a 2nd home or ancestral village in east Bengal.

    yeah. some of the people who lived in west bengal went back and forth for a while, but i think that’s ending as those properties are not sustainable due to laws

    1. Razib, I think it is three things. But need to check with Bangla sources to verify.
      1) A lot of Indian muslims moved to Bangladesh in 1947 and then returned to India later on. (something on the order of over 30 million “Bangladeshis” moved to India to escape Islamism and Pakistan.)
      2) A lot of nonmuslim Bangladeshis moved to West Bengal causing a denominator effect
      3) Some Indian muslims moved to Bangladesh and they or their children married and stayed. (suspect this effect is smaller than the top two)

      Note (1) above is I think is part of the reason why PM Modi quietly messaged Bangla people saying that illegal Bangladeshi immigrants would not be deported under NRC.

      Also note that PM Modi and the BJP has allied with many West Bengal muslims (including indirectly through third parites). My suspicion is that this is also heavily from note (1). West Bengal also has rising Islamism. But I would like to verify this by asking Bangla sources.

  3. Currently West Bengal and Bangladesh are socially and economically integrating with strong support from Sheikh Hasina, West Bengal CM Mamata Banerjee and PM Modi. {CM Mamata Banerjee and PM Modi are political opponents. I personally respect CM Mamata Banerjee.} I think security in Bangladesh has improved over the last 3 years (since the fiasco of Islamists murdering journalists and politicians). This is leading to more cross border business collaboration, Bengali Indians buying homes in Bangladesh and Bengali Indians visiting Bangladesh.

    I personally am a fan of Sheikh Hasina.

  4. Long-term it would be good for India and Bangladesh to have open borders like India and Nepal. Short to medium-term, and perhaps far more realistically, it would be good if India and Bangladesh had the kind of relationship that the US and Canada have.

    It’ll be vital for Northeast India.

    1. I feel lot of this advantages of inter connectivity are overblown. There is literally no advantages with closer Indo Bangladesh ties. Apart from feel good factor. Indo Nepal open border is itself a relic of religious and cultural reasons and not due to trade. An open border would have not materialized had it been negotiated today

      West Bengal has almost all the advantages which the north East needs to prosper. And let alone north East, West Bengal itself is stagnant and living of the advantages for the British raj.

      The reason of India”s East being backwater is less due to inter connectivity reasons and more due to its own inherent issues.

      1. “I feel lot of this advantages of inter connectivity are overblown.”

        Connectivity is the single most important factor for economic growth. Even within India, villages within 5 km of a city grew a lot richer, while those farther away actually became poorer. And connectivity to the sea is even more important.

        Bangladesh is growing rapidly, and we will need to develop an integrated power grid with them and Nepal soon.

      2. West Bengal is about average for most development metrics within India. So it’s not all that bad. It’s far more developed than UP, Bihar, MP and about on par with a state like Rajasthan.

        Also while yes the Indo-Nepal border is open because of historic reasons, there’s enormous untapped potential in that relationship, as well. Especially hydroelectricity. An integrated energy market would be huge.

        Really the whole “BBIN” region (Bangladesh, Bhutan, India, Nepal) needs to cooperate more.

      3. Inter connectivity b/w countries is the issue here. India and Nepal have an open border and as much inter connectivity as possible infrastructure wise. If the argument is better infrastructure than yeah its better. But considering S-Asian standard of infrastructure its there b/w India and Nepal. India and Nepal actually are at the end of their relationship, and i wont be surprised if we would have a visa system in the near future.

        On development of various India’s states, yeah Bengal can take solace from the fact that its above UP,Bihar etc. While its peers in 47 like Maharastra, Gujarat,Delhi have long surpassed it. And comparing a Bengal with states like Bihar is like comparing Hyderabad with Amravati.

        1. There is a world of difference between Nepal’s elite, which is bureaucratic and non-entrepreneurial, versus’s Bangladesh’s elite.

    1. The transit corridor is of course important, both in a practical sense and even in a psychological sense. But there’s even more to it than that.

      India and Bangladesh share major rivers, and if they can cooperate in building infrastructure for hydroelectricity production, mitigating floods, and shipping, both sides stand to gain substantially.

      And then there’s all sorts of benefits that we can’t really predict or spell out easily. Think of the example in Canada where for a very long time Canada tried to build a strong East-West national economy instead of just following organic trade routes and markets that existed even prior to colonization (mostly North-South given North American geography). Now with NAFTA and so on a transnational region like the Great Lakes, for example, functions economically as though it doesn’t really have borders, to substantial mutual benefit.

  5. Demography played a part, but I think the importance of the British militarizing the Punjabi Muslim cannot be understated.

    Without the Punjabi Muslims making up a third of a British Indian army, even a 75% Muslim Punjab would have made its peace with Hindu majority India.

Comments are closed.

Brown Pundits