Music and Communal Violence in Colonial South Asia

Winner of the JaFran Jones prize for best student paper at the 2012 conference of the Midwestern Chapter of the Society of Ethnomusicology (MIDSEM).

In the final decades of the nineteenth century, music came to occupy a central role within patterns of recurrent communal, or sectarian, violence occurring in British India. Newspaper readership across the subcontinent during the period became all-too familiar with the term “music-before-mosque riots,” and the phenomenon it described. Specifically, “music-before-mosque riots” arose following the deliberate display of a musical procession, usually accompanying a Hindu festival, in front of a Muslim place of worship, causing offense and, very often, violence. I think of this as a “colonial” phenomenon, in the sense that it developed under a set of changing social conditions in which colonial state officials and various bodies of indigenous subjects interacted with each other. This, however, is not to say that the occurrence of music-before-mosque riots has been limited to the colonial era alone, which formally ended in 1947. This kind of communal violence still occurs in India, albeit far less frequently.

In order to convey the severity of the kinds of riots I'll be discussing, I'd like to quote a passage written by an anonymous correspondent from the Indian newspaper The Pioneer in 1882. The correspondent visited the city of Salem in South India, following a bloody series of riots that had just occurred over music passing before a mosque, in order to document the physical devastation. After arriving in the city, the correspondent reports that he

saw some terrible sights. I had read, of course, at the time, of the so-called Bulgarian atrocities, but I can hardly believe that there was any instance of cruelty practised by the Turks which would equal some of these. One Mussulman infant, lying dead, had had his arms torn from the body, and the intestines were protruding from a hatchet cut in the abdomen. A little further on one came across headless corpses of men and women – Mussulmans in all cases. In fact, I saw no Hindu dead. The place seemed as if it had been stormed. There were corpses on all sides, and the houses were burning. The [mosque] . . . was almost razed to the ground, and its rich furniture and chandeliers completely destroyed. Peeping into the wells in the premises of the mosques, I saw dead pigs thrown in with the corpses of Muhammadan children. (The Pioneer 1882)

This paper represents a preliminary study of music-before-mosques riots based on my MA thesis in Ethnomusicology at the University of Wisconsin-Madison. In this paper, I intend to situate these riots within a larger narrative of community and communal imagining as well as examine how popular conceptualizations of musical sound underwent a radical transformation through this phenomenon, imbuing music with a particular kind of power in terms of how colonial subjects forged consequential political relationships with each other.

Before presenting my argument, I want to first address that the music-before-mosques phenomenon has by no means evaded scholarly interest entirely. However, I feel that the ways in which scholars have approached the issue in the past have been inadequate in their treatment of  musical significance, and therefore my interest lies partially in responding to these studies. Early analyses of music-before-mosque riots were dominated by what I like to call the “fundamental incompatibility” hypothesis. The succinct version of this line of reasoning holds that the “Hindu religion,” understood as a monolithic belief system, demands the performance of musical sound in order to appease the gods. This compulsion towards the musical, according to the hypothesis, is in direct conflict with “Islam,” again understood as a monolithic belief system, which asserts the moral impropriety of music and thereby proscribes its performance. Thus, the result of two populations with such irreconcilable world views living amidst each other will inevitably be violence. The problem with the “fundamental incompatibility” hypothesis is that it fails to recognize the centuries of Hindu-Muslim cohabitation on the subcontinent during which time countless festival processions occurred seemingly without incident. Furthermore, it fails to recognize the heterogeneity in both theology and practice of the two groups unproblematically labeled “Hindu” and “Muslim.” One might cite instances of Islamic ritual in which music plays a central role, or note the well-documented “syncretic”1 tendencies of worship in South Asia prior to the late nineteenth century, wherein nominally “Muslim” subjects often participated in nominally “Hindu” festival celebrations, and vice-versa.

I'd like to address one other interpretation of the music-before-mosque phenomenon espoused in much twentieth century writing, one to which I'm much more sympathetic but which still seems inadequate for the purposes of music scholarship. Identifying this inadequacy is important for my paper, since it will help to frame my argument. In writing about riots involving music, many scholars tend to dismiss the significance of the musical dimension, as if it were a relatively interchangeable factor obscuring greater social tensions. I wish to reconfigure the analysis of these riots, offering greater emphasis to the musical. Rather than representing the expression of pre-existing conflict between pre-existing communities, I feel that these riots, driven by music, had a generative effect on the way in which these communities were being formed or reformed. Religious affiliations, unprivileged as markers of social identification before the period I'm interested in, became the basis for the production of boundaries defining and separating political communities. Music, or new and differing attitudes towards it, came to represent a point of differentiation between “Hindu” and “Muslim” communities, as well as a point upon which these identifications expanded to become national communities of shared political ideas, and not just local communities of shared theological beliefs. Robert Frykenberg, among others, has claimed that “Hinduism,” as a cohesive religion or community, is best understood as a colonially constructed category superimposed upon a “mosaic of discontinuities” (1989:41). What I'm arguing is that these riots laid the foundation for the imagining of a national, politicized community of Hindus. In turn, through violent opposition, the music-before-mosque riots phenomenon similarly shaped a national, and at times even transnational, community of Muslims. Through this phenomenon, boundaries were generated and reinforced.

In The Elementary Forms of Religious Life, Emile Durkheim argues that group integrity is produced and maintained through ritualized moments of effervescence (2001 [1912]). He writes that “because a collective feeling cannot be expressed collectively unless a certain order is observed that permits the group's harmonious movements, these gestures and cries are inclined to be rhythmic and regulated, and become chants and dances” (ibid.:163). Employing the language of music, Durkheim contends that through repeated patterns of collective performance comes group affirmation. Communal riots involving music represent a kind of effervescence, generating and shaping new forms of social identifications. Musical practice and conflict, which in these cases are clearly related, become an almost ritualized (in their patterning and association with annual festivals) mode of group consolidation for both sides of disputes, through radical opposition.

From a relatively early date, one can observe a broad geographical patterning of music-before-mosque riots. Within the first few decades of the twentieth century, large riots had occurred in major cities from Madras to Calcutta to Bombay. Something that is particularly fascinating to me is that, in juxtaposing this geographical spread of music-before-mosque riots and the development of the Indian nationalist movement, in some ways the music-before-mosque issue appears to be a national phenomenon that preceded the nation itself. What I mean to say is that, for many living in British India at the time, feelings of relation at a national level to other subjects along communal lines (as Muslims, Hindus, etc.) predated, or at times eclipsed, feelings of relation to a secular nation of Indians. Unlike the nationalist movement, the imagination of multiple nationalized communities within communal categories was by no means limited only to elites. Music-before-mosque violence was therefore a popular, “All-India” phenomenon at a time when few cultural forms pervaded the entirety of India's vast geography.

I'd next like to examine how shifting concerns of Hindu and Muslim leaders in India contributed to the development of this phenomenon. Bal Gangadhar Tilak was a central figure in the late nineteenth and early twentieth century nationalist movement. By most contemporary accounts, he was also a Hindu fundamentalist with strong anti-Muslim tendencies. One author characterizes Tilak's goal as being to “bring the Hindu community together . . . and provide a sense of Hindu solidarity . . . vis-à-vis the Muslims” (Barnouw 1954:81). Tilak more or less single-handedly reinvented the Ganapati festival in Maharashtra (Left, photo from Columbia University). This was previously a devotional festival celebrated in honor of the god Ganesh, but came to be a public spectacle with strong political motivations. Part of this reinvention was pragmatic: British authorities suppressed political meetings in public, while typically remaining more permissive with religious celebrations. Therefore, while openly organizing a political demonstration would have been unsuccessful for Tilak, he was able to promote his agenda under the guise of Ganesh worship and evade police suppression. The festival centralized musical performance, and through these performances we can detect anti-Muslim sentiment, especially in the texts of songs sung during the celebration. One such song, in translation, asked, “What boon has Allah conferred upon you, that you have become Mussalmans today? Do not be friendly to a religion which is alien, do not give up your religion and be fallen” (Muralidharan 1994:13). Another song states, “Disturbances have taken place in several places, and Hindus have been beaten. Let all of us with one accord exert ourselves to demand justice” (ibid).

The timing of the first Ganapati celebration in 1893 was also taken as a direct offense to Muslims, as it directly coincided with the observance of the Muslim festival of Muharram. In the past, many Hindus participated in Muharram celebrations, but Tilak called upon Hindus to boycott the festival that year, offering Ganapati as a new, and divisive, alternative. The festival, therefore, was also an attack on syncretic religious practices that previously had flourished in India.

Tilak also initiated a festival in honor of the seventeenth century Maratha historical figure Shivaji, whose celebrated status derived from his ability to repel the Mughals from the Maratha domain. Because of this, a public celebration of Shivaji held a dual purpose and effect. On one hand, it highlighted a case of indigenous triumph over foreign invasion, thereby paralleling the hopes of opponents to British colonialism. On the other hand, Tilak and many festival celebrants presented Muslims, exemplified by the Mughals, as foreign invaders themselves. In doing so, they increased hostility towards Muslims in Maharashtra. Unsurprisingly, the Ganapati and Shivaji festivals lead to several music-before-mosque incidents, including violence in 1926 in Nagpur and in 1927 in Surat, to which police responded by firing their guns “in an indiscriminate manner . . . at innocent people in the neighbouring shops” (The Times of India 1927).

Basically, Tilak was reconceptualizing what it meant to be a Hindu. He aided in shifting this category from designating a group of individuals with shared religious belief, enacted through rituals like festival celebration and musical performance, to a political body engaged in ongoing hostilities with Muslims.

The nineteenth century witnessed a proliferation of Islamic reformist movements, in South Asia and elsewhere. Many of these movements sought to directly synthesize theological interpretation and political engagement in the context of European colonialism. Beginning in the 1860s, the Deobandi tradition in India, following Wahhabism and other reform movements, began to politicize religious practice to such a degree that at times it seemed to give priority to the political over the spiritual (Allen 2005). The Deobandi movement had a significant impact on the political consciousness of many Muslims in South Asia and in other geographical regions, positioning itself, through its theological teachings, against the encroachment of European colonialism. It explicitly denounced music and dancing as one of the basic tenets of its doctrine. I do not think that the proscription of music and the Deobandi political position are unrelated. I think that it is possible that musical practice (or its denouncement) became a cultural node upon which religio-political identifications might be established and expressed. This might be even more clear in another example, that of the Tablighi Jama'at movement, which had strong roots in the Deobandi school.

Scholars have argued that Tablighi Jama'at, founded in 1926, emerged, in no small part, as a reaction to the growing presence of Hindu fundamentalism in India, and framed its theological teaching within those terms (Mayaram 2004). The practice of music is unequivocally linked to religious “Otherness” within the rhetoric of the movement, which often includes clear and pejorative references to Hindu gods. Condemnations of “false gods” and their followers are paired with the denouncement of music, thereby portraying musical practice as the point of differentiation between followers of Tablighi Jama'at and Hindus, Buddhists, or Jains.

As Tilak called upon Hindus to stop participating in Muslim festivals, many Muslims, according to Suranjan Das, began to reevaluate their own syncretic traditions and participation in Hindu festivals (1990:30). Das writes that issues involving the permissibility of music near mosques or in general “which hardly worried Muslims in the past, now emerged as an important nucleus around which Muslim communal solidarities developed” (ibid).

I'm going to present an example of a music-before-mosque incident that illustrates the way in which music had attained a contentious kind of power. In October 1924, several disturbances were reported during celebrations of the Durga Puja festival in the greater Calcutta area. A group of celebrants gathered outside one mosque. The Times of India reported that “stone-throwing was indulged by the Mahomedans within the mosque . . . [and the] Hindus retaliated by playing band and music” (Times of India 1924; emphasis mine). If the language chosen by the journalist represents an accurate telling of the events, this account is significant as it portrays music as something more than merely a precipitating or incidental factor in the clash. After engaging in a physical encounter, the Hindus involved “retaliated” with their music. If considered in terms of retaliation, the performance of music in front of mosques no longer seems a matter of triviality amidst deeper social issues between religious communities. Rather, the performance of music, in this context, is itself a form of violence. Understanding the use of music as a weapon of retaliation, one might begin to reconfigure interpretations of violence in music-before-mosque riots.

Neil Whitehead suggests that violence can be “part of the expression and creation of identity and group organization” (2004:59). He writes that “almost all theoretical and research approaches to violence begin with the assumption that, at its core, violence represents the breakdown of meaning, the advent of the irrational, and the commission of physical harm” (2007:1). To the contrary, Whitehead demonstrates that violent acts certainly can and do have cultural meaning.

Proceeding from this point, one can attribute productive potential to violent expression in its ability to structure meaningful social relationships. However, in scholarship addressing the music-before-mosque issue, music (and subsequent conflict) is portrayed as having an effect that is more divisive than integrative, and more destructive than productive. Without a doubt, this trend, in part, represents a consequence of limited theoretical understandings of violence. Whitehead wrote of a tendency for scholarship “to focus on the political and economic conditions under which [violence] is generated,” at the expense of more extensive examination (2007:6). This illustrates precisely the problem with previous studies of music's involvement in Indian communal riots. Scholars are typically preoccupied with analyzing those factors that contribute to what appears to be complete failure in “normal” social processes, rather than what is created from the riots. Evidence suggests that communal cohesion, prior to the series of riots with which this project is concerned, was much looser and more localized (Das 1991; Bhattacharya 2007). But, with each incident being reported in national newspapers as a “communal riot,” these occurrences reproduced themselves across the Subcontinent and became a widespread phenomenon, and furthermore helped to produce the very communal groups between which the violence was occurring.

Music and violence typically exist on opposite ends of the interpretive spectrum within popular discourse, which holds music to be largely innocuous and violence to be universally destructive. However, a careful consideration of these concepts from a theoretical point of view, as well as their particular articulation in the case of music-before-mosque incidents, demonstrates their complexity and many points of intersection. Issues of public ritual and processional performance in symbolic terms represent one such point of intersection. These issues are crucial here in understanding festival music and violence as expressions of creative agency.

From the perspective of imperial power, marching music expresses domination and territoriality. The idea of a highly stylized and ostentatious movement through space with great sonic volume implies control over that space in some ways. Given the direct involvement of British administrators and legislators in music-before-mosque issues (a point upon which I will not be able to expand today), one might argue that the recontextualization of Hindu festivals in the late nineteenth and early twentieth centuries might carry similar implications with regard to space and power. The deliberate movement of music processions, established already as a form of violence themselves, before places of quiet worship might be viewed as an assertion of territorial control; an interpretation that seems to correspond with the trajectory of increasingly Hindu-leaning nationalist movements gaining momentum around the same time. Hermanowicz and Morgan identify two potential effects of public rituals. One is an “integrative function,” which I understand as solidarity-production along the lines of Durkheimian effervescence. The other is the ability to express “power and conflict rather than unity . . . [thereby allowing] parties in a dispute to claim positions of dominance” (1999:199). In a discussion of parades, Michael Ashkenazi corroborates these elements of public ritual. He writes that parades “serve a local-integrative function . . . [or] may also indicate or emphasize fragmentation” (1987:35). Hermanowicz and Morgan continue to state that the dual functions of the public ritual “need not be mutually exclusive . . . [as they] can contain processes of both social cohesion and conflict . . . [and] both mark distinctions and promote unity, sometimes within a single ritual event” (1999:199). The violence apparent in music-before-mosque riots during festivals, in all of its many forms, might actually be perceived as part of the festival celebration itself. Through the construction of meaning from historical and religious interpretation, as well as the relationship between riots both structurally and as a media phenomenon in newspapers, these cultural forms contributed to the emergence of “Hindu” and “Muslim” as national categories of communal identification. Victor Turner wrote that public celebrations “[attempt] to manifest, in symbolic form, what [a social group] conceives to be its essential life, at once the distillation and typification of its corporate experience” (1982:16). While some might find it disturbing, rioting and murder over music in this case seemingly qualifies within Turner's conceptualization of the public ritual.



Allen, Charles. 2005. “The Hidden Roots of Wahhabism in British India.” World Policy Journal 22(2):87–93.

Ashkenazi, Michael. 1987. “Cultural Tensions as Factors in the Structure of a Festival Parade.” Asian Folklore Studies 46(1):35–54.

Barnouw, Victor. 1954. “The Changing Character of a Hindu Festival.” American Anthropologist, New Series 56(1):74–86.

Bhattacharya, Tithi. 2007. “Tracking the Goddess: Religion, Community, and Identity, in the Durga Puja Ceremonies of Nineteenth-century Calcutta.” The Journal of Asian Studies 66(4):919–962.

Das, Suranjan. 1990. “Communal Violence in Twentieth Century Colonial Bengal: An Analytical Framework.” Social Scientist 18(6/7):21–37.

––––––. 1991. Communal Riots in Bengal 1905–1947. Delhi: Oxford University Press.

Durkheim, Emile. 2001 [1912]. The Elementary Forms of Religious Life. New York: Oxford University Press.

Frykenberg, Robert. 1989. “The Emergence of Modern ‘Hinduism’ as a Concept and as an Institution: A Reappraisal with Special Reference to South India.” In Hinduism Reconsidered, edited by G.D. Sontheimer and H. Kulke, 1–29. Delhi: Manohar.

Hermanowicz, Joseph and Harriet Morgan. 1999. “Ritualizing the Routine: Collective Identity Affirmation.” Sociology Forum 14(2):197–214.

Mayaram, Shail. 2004. “Hindu and Islamic Transnational Religious Movements.” Economic and Political Weekly 39(1):80–88.

Muralidharan, Sukumar. 1994. “Patriotism with People: Milestones in the Evolution of the Hindu Nationalist Ideology.” Social Scientist 2(5/6):3–38.

Anon. 1882. “The Riots at Salem [from our Bangalore correspondent].” The Pioneer, 25 August:p3.

––––––. 1924. “Hindu-Moslem riots: More outbreaks.” 1924. The Times of India, 10 October:p10.

––––––. 1927. “Outbreak reported from Dacca: Moslems' alleged attack on Hindus.” The Times of India, 7 May:p12.

Turner, Victor. 1982. Introduction to Celebration: Studies in Festivity and Ritual. Washington, DC: Smithsonian Institution Press.

Whitehead, Neil. 2004. “On the Poetics of Violence.” In Violence, 55-78. Santa Fe: School of American Research Press.

––––––. 2007. “Violence & the Cultural Order.” Daedalus 136(1):1–11.



1 This term has been challenged in contemporary scholarship with regard to its usefulness. I include it because of its historical invocation within this discussion.


Julian Lynch's picture
Julian Lynch

Julian Anthony Lynch is a PhD student in Ethnomusicology and Anthropology at the University of Wisconsin-Madison. His work focuses on issues of conflict in musical practices in colonial and contemporary India. He completed an MA thesis in Ethnomusicology on that subject in 2012.