The Minority in Banaras: Intercommunity Power Friction and Heterogeneity

The contents of the paper given below have already been published in the interdisciplinary journal Voices. It’s being put on the blog for the kasiphiles who access this blog for information on Banaras.

 Muslims Chausatti Ghat

Banaras, Kasi or Varanasi has been seen as a microcosm for ages. It is said that the cosmos is present with all its constituents in their micro form in the city. The matters non-material can neither be proven nor disproven using materialistic formulae. Banaras is India in its micro form too, and an attempt can be made to prove that at both material and non-material levels. People belonging to various religions, languages, states, castes, communities, and, even countries have lived in the city for ages. As a system, Banaras had always had a dynamic equilibrium between its basic building blocks: the people who live there.

As far back as the light of history can show, Hindus of various castes and sects have always been a significant majority in the city. James Prinsep’s record of the population of Varanasi in the first half of nineteenth century shows that roughly 67 percent of the population was Hindu and 17 percent was Muslim (14). He had also mentioned the number of temples and mosques in various localities of the city. There were 1000 temples and 333 mosques in the city back then (13), i.e. the ratio of temples to mosques is 3:1. He had subdivided the urban zone into various areas viz. Dasaswamedh, Bhelupura, Adhampura etc. Later on, Sherring had provided amended data. Even then, the ratio of temples to mosques remained the same. He had reported the numbers of temples to mosques in the ratio of 1:3 at Jaitpura, and 8:9 at Adhampura, both with a high concentration of Muslims. At Dashaswamedh the ratio was 20:1, with the overall count as 1454 temples and 272 mosques (42). The present paper studies the area between Sonarpura and Shivala. The area covered here falls under what Sherring had kept under Bhelupura.

There are 83.7% Hindus and 15.9% Muslims in Varanasi according to the Census 2001. Muslims have always been a minority in the city, but a very prominent and significant one. They have contributed to the formation of the material and cultural essence of the city. They have been subject to a long tradition of essentialization and amorphization that has stripped them of any kind of agency and existential responsibility of becoming. They have become a prototypical “them’ for the majority of the members of the majority religion in the city.

Banaras is the oldest continually inhabited city of the world (Eck 5). Through centuries of conscious and unconscious magnetization, the city has formed religion (and caste) based domains in at least its older quarters. It’s the religious geography of one of the most densely populated zones of the city that informs this paper. The area chosen for the purpose of demonstration of what lies at macro-sociopsychological level comprises of the domains arrayed round the axis of Harishchandra Ghat Road. The axis divides the quadrilateral shaped area with the north and south lines that are parallel to it having Kooch Behar Kali Bari and Panchkot Kali Bari on them.


Kooch Behar Kali Bari

Like Prinsep’s study, no attempt will be made to include the hundreds of temples in the niches, under trees etc. that are very common in the Hindu quarters. Neither will all the temples in private houses be included in the total. The concept of a private mosque for a household does not exist, as far as I know. Only larger public temples (larger in comparison to the niche-tree type) and mosques proper shall be considered for the study.

As the number of Muslim places of worship is comparatively less, let’s begin with them. There’s a mosque at Shivala, right behind Panchkot compound. There’s another at Chhipi Tola near Hanuman Ghat and another by Sonarpura-Shivala main road, near Prasad X Ray. There are two shrines: one at Shivala Crossing and another near Sonarpura Crossing, near Kooch Behar Kali Bari.


Shri Kedareshwar Temple

Sri Kedareshwar Temple at Kedar Ghat is the nucleus of Kashi Kedar Khand. The clanging of huge bell of this temple can be heard, especially early in the morning, in a very large area around the temple. Some people find it convenient to designate as kedar khand the area within the boundaries of which the sound of the bell is audible. Therefore, the area covered in this paper roughly corresponds to the same. This fact alone will suffice to explain the large number of big, small and very small temples in the locality. The core of this zone is the temple that lends its name to the zone: Sri Kedareshwar Temple. Hanuman Temple near Hanuman Ghat, Chintamani Ganesh Temple at Harar Bagh, Kali Bari at Pandey Haweli and the temple made in the south Indian style near Harishchandra Ghat are the well known larger temples of the zone.

A newly constructed temple of Sai Baba at Harar Bagh draws huge crowd of the devotees, Hindus, of the locality. The rise of the popularity of Sai Baba amongst the Hindus of Banaras is a recent phenomenon, and the small temple (that was originally just a collection of various other Hindu idols) in an unknown lane of the city lends its name to the locality now. As mentioned already, there’s a really large number of smaller temples and idols on various ghats and lanes of the area that, if included to the grand total, will pull it beyond just one hundred. So, the ratio of the Hindu and Muslim religious places is roughly 10:1, that’s mush more in comparison to the same a century and half ago. There’s a definite mechanism of the construction of the houses of religious worship. They have a lot to do with the socioeconomic strength of the community to which they belong. The changing ratio tells its own story, but it needs to be interpreted for a better understanding.

A jump from religion to demography may seem abrupt at the surface level, but there’s an internal justice for the jump that becomes clearer gradually. The locations that have mosques or Muslim shrines also happen to be either densely populated Muslim domains or they are contiguous to such domains. The collective un/conscious of Hindus and Muslims has programmed their habitation patterns and religious microgeography too.

In any other circumstances it’d be called ghettoization, but looks could definitely be deceptive. The muhallas can never be called ghettoes and the populations of the muhallas of both the communities have houses constructed of similar material where live people of similar sociocultural and economic backgrounds. Hence, a benefit of doubt kind of thing starts its play in the mind. Yet, when surface is scratched, one sees a subtle and successful attempt at ghettoization here. Markers like fencing, barricading, restrictions on movement pattern and coercive control may be absent from these ghettoes, but there’s something about the polarized population’s mind that makes their ghettoes within. There are lanes of Hindus and those of Muslims that pride themselves on being 100 percent pure, i.e. there’s not a single house there that belongs to a member of the “other” community. Not only the ownership pattern of these lanes is homogeneous, but also that of those staying on rent. Of course, there are some areas where the boundaries of habitations fuse – where there’s a marked heterogeneity in the composition of population.

Homogeneity translates into absolute power, and power asserts itself in free display and construction of the markers of identity. Hence, all the mosques are in dense Muslim domains and the two shrines are by the main streets of Sonarpura and Shivala, nearly marking the limits of Muslim and Hindu populations on their sides. It may be just coincidental, or it may not be, that members of both the communities have (at least theoretically) free access to the shrines. The boundaries of Muslim domains (ghettoes?) and those of their Hindu counterparts touch these shrines. Jhanda Tare and Shivala are the Muslim domains whose boundaries touch the shrines either nearly or actually. The mosques, as mentioned already, are embedded in the domains, except in one case. The temples are deep inside Hindu domains.

Laat Bhairav, where it had all begun in 1809

Religion has always been one of the main sites of the contest of power in Varanasi. The history of riots in the city, that dates back to 1809, affirms the centrality of religion in power contests and the fusion of religion with socioeconomic factors creating the juggernaut that’s finally satisfied only when it had taken the sacrifice of many lives – both Hindu and Muslim. The location of the places of worship too becomes significant in a clime like that – more so, in heterogeneous than in homogeneous localities. The clearly and homogeneously Hindu or Muslim domains are generally free of any kind of property purchase pressure from the “other” community. There’s a taboo against selling property to the “other” and those with social roots in the locality and city know it well. So, aberrations happen very rarely.

The heterogeneous domains provide interesting opportunity for the study of domain re-formation, expansion and contraction. The area around the two shrines mentioned already and one mosque in heterogeneous localities prove the point well. Due to their importance for one community property around them is in high demand amongst the members of Muslim community. They feel the need to buy in order to homogenize the domain and tilt the power balance in their favour. It’s happening slowly but surely – the homogenization in favour of the community to whom the mosque and the shrines belong. Religious cartography and domainization are aspects and expression of power – the same power that was coveted and contested for back in 1990. The matter of the ownership of their place of worship has an uncanny hold upon the collective un/conscious of the people. It has been proven time and again and proven very conclusively through the rise of Bhartiya Janta Party in the 90’s.

No prominent temple of kedar khand is embedded in a Muslim domain. The same can’t be said about the Muslim places of worship. This indicates towards a larger phenomenon. The inter-community power distribution is proportionate to its comparative population density in a given locality. Thus, there’s no contest or confusion about power in homogeneous domains, it’s only in the heterogeneous ones where heat is generated due to intercommunity power friction. The quadrilateral covered in this paper has two nearly homogeneous Muslim domains at Shivala and Chhipi Tola, and one very little homogeneous strip near the shrine at Sonarpura. The last population is not being termed a homogeneous domain because it’s a weaver muhalla of around twenty-five Muslim houses, the last one of which is where the compound of Kooch Behar Kali Bari starts and the first one behind the shrine.

Varanasi has a long history of the parallel lines of communal harmony and disharmony between the two dominant communities of Hindus and Muslims. Harmony isn’t as conspicuous as disharmony because of the shock value of the later although the line of disharmony is neither thicker than the other one, nor is it unbroken. Moreover, its being an aberration sets each instance of disharmony apart from the continuous line of harmony. The flashed points on the line of disharmony are many, and 1809 is where the availability of detailed records began. The riots of 1809 were about intercommunity power friction in a heterogeneous domain, so were many riots in the later years including the riots of 1990’s. All those riots had something to do with a temple, an idol or a religious procession, and the point from where it began had almost always been in a heterogeneous domain.

The heterogeneity of domains has a close connection with the risk of disharmony’s overpowering harmony and culminating into a set of violent and hateful actions that are clubbed together and called a riot. There’s a close but not one-to-one relationship between them, as heterogeneity is a constant factor in many localities but disharmony, in its mild to extreme forms, appears only sporadically in the stream of time. What is it that creates differential response to the same stimulus? Why is the entropy of the system maintained in some cases and challenged in some other? Why and how do many heterogeneous Muslim domains remain harmonious most of the time?

Studies of riots point towards planned and concerted action, manipulation and the involvement of external agents in breaking down the equilibrium of the system. The mechanism of creating conditions favourable for riot-mongers has been studied in detail in many cases. Engineer and Pandey, both have shown how socioeconomic factors play a very crucial role in mobilization of masses after rallying them around something linked to their survival at basic level, e.g. safety of life or property. Social programming at intra and inter-domain levels plays a very central role in the formation and solidification of differential identities and in the vilification of the “other”. Those looking for opportunities to create communal friction get ample support from an already biased and paranoid mind programmed against the essentialized “other” since its infancy. Rumours have always played an unacknowledged yet significant role in the initiation and spread of unrest, especially when the situation goes out of control due to system failure. The intercommunity power friction has a full range of intensity, from its milder and harmless forms to the extreme one of full scale communal riots. The level of heterogeneity of a domain is one of the factors that determine whether friction will translate into full scale communal riot and violence or will remain smoldering for a long time, apparently harmless, waiting for the winds of circumstances to blow from over it and convert it into its more vicious ramifications. It becomes very important, therefore, that the heterogeneous domains are watched over closely and studied intently for devising prophylactic measures.



“Basic Data Sheet (Varanasi) – Census 2001”. n.d. Web. 17 December 2013. Pdf.

Eck, Diana L. Banaras City of Light. New Delhi: Penguin, 1993. Print.

Khan, Rafiq and Satyaprakash Mittal. “The Hindu-Muslim Riot in Varanasi and the Role of the Police”. Communal Riots in Post-independence India. Ed. Asghar Ali Engineer. Hyderabad: Sangam, 1997. Print.

Pandey, Gyanendra. The Construction of Communalism in Colonial North India. New Delhi: OUP, 1990. Print.

Prinsep, James. Benares Illustrated in a Series of Drawings. Varanasi: Vishwavidyalaya Prakashan, 1996. Print.

Sherring, M. A. Benares: The Sacred City of the Hindus. New Delhi: Low Price Publication, 1996. Print.



Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s