Last week, I finally got around to digitize the old map of Lucknow's Shia and Sunni population, which an enthusiastic Census officer produced in 1961, and which I managed to acquire in full copy three days before I left India last December.1 Geocoding the scan and counting all the little dots (which represent 200 Shia or Sunni households) resulted in the following map of Shia population:

First of all, I am again astonished how much smaller Lucknow was only five decades back! While growth pattern of the last two decades were interesting enough (mapped here), it is easy to forget that Lucknow in the 1950s barely extended across the Gomti at all. But more than that, the digital map now allows for a more systematic comparison of sectarian population distribution over time - confirming, first of all, my earlier hunch that there was apparently a Shia concentration in the eastern Chowk area (Gol Darwaza, KGMU) as well as in some pockets in Aminabad - but not (yet) the "ghetto" of Kashmiri Mohalla...

Before I proceed to quantify that insight, though, let me emphasize three caveats. First of all, we don't know on which empirical basis and using which methods that Census officer drew his map in the 1950s: his accompanying account seems credible and insightful, but he makes no comment on the mapping process per se. How reliable his map is? We can't know. Likewise, my own map, based on the electoral rolls of 2011, comes with many qualifications. Both maps thus have a margin of error that remains hard to quantify. To complicate matters, both maps also rely on different sources of data - most likely survey data in 1961, but election data in 2011. Any or all of the following differences might thus stem from error margins or methodological deviation more than from true historical change. Sadly, there is little I can do about that possibility, apart from concentrating on strong effects only - and following up with ethnographic fieldwork later this year. But here you go - differences in sectarian population distribution 2011 and 50 years earlier (red areas signify a relative influx of Sunnis, green areas a relative influx of Shias):

On first sight, this map seems to be fairly green with only one large red area in eastern Chowk. Which would suggest that Sunnis concentrated while Shias diversified - in contradiction to both earlier research and my own ethnographic impressions.2 A more convincing explanation could thus be that Sunnis were, perhaps, more upwardly mobile than Shia, and thus moved to a greater extent into Lucknow's new suburbs (or that labor migrants living on the outskirts are more likely Sunni than Shia), leaving a Shia concentration in the old town while creating a Sunni concentration in new Lucknow. These areas are not covered by the above map, explaining the sea of green. Yet in an earlier map and a related analysis of residential segregation, I have shown that even today only few Muslims live across the Gomti. And a brief look in my dataset confirms that their sectarian makeup resembles that of the Muslim population in the old city: 30% Shia vs 70% Sunni.

But there might be yet another explanation: perhaps the Shia population grew more rapidly over the last fifty years than the Sunni one? This, indeed, seems to be the case. The 1961 map suggests that only 20% of Lucknow's Muslims were Shia - which was only two thirds of their contemporary population share vis-a-vis Sunni Muslims. Which explains why that first map was so green: there are few places without relative Shia influx, because the Shia population share grew everywhere. To assess sectarian change not just in toto across Lucknow, but in spatial breakdown, it thus seems reasonable to factor in that overall more rapid growth of the Shia population. From which the following map emerges (now, red areas indicate either a Sunni influx or at least a less than average Shia influx, while green areas indicate many more Shias than in the Census map):

Now this is very interesting. Even if I factor in the methodological caveats mentioned above and concentrate on areas with sharp differences only, it seems that there was a Shia influx into Muftiganj, Kashmiri Mohalla (consistent with recent findings), Kaiserbagh, and Wazirganj (my own area - interesting since it has always had an old Shia presence, which I already suspected to be more in imagination than in numbers). In contrast, eastern Chowk (Gol Darwaza, Medical College) in particular became much more Sunni, as did a pocket in Aminabad (no idea why - yet), and - though less clearly - some areas west of Kashmiri Mohalla. It will be very interesting to follow up on this pattern during my second round of fieldwork this summer.

Any ideas from you, my readers, are of course very welcome in the comments below...

I would like to acknowledge the use of the Oxford Supercomputing Centre (OSC) in carrying out this work.

  • 1. Census of India. (1961). Moharram in two cities (Lucknow and Delhi). In: Census monograph series, part VII-B: Fairs and festivals.
  • 2. Verniers, G. (2012). A minority within a minority: The Shias of Kashmiri Mohalla, Lucknow. In L. Gayer, C. Jaffrelot (Eds.), Muslims in Indian cities: Trajectories of marginalisation. London: Hurst.
Census of India - Moharram in Lucknow (1961).pdf43.98 MB