Seven—
State and Community:
Symbolic Popular Protest in Banaras's Public Arenas

Sandria B. Freitag

Despite its long-lived centrality of place in Indian perceptions, Banaras has not figured prominently in accounts of modern Indian history. This is primarily because it does not lend itself to the approach previously taken by most historians, which focused on the response of a Western-educated elite to the British Empire.

As the richness of subject matter discussed in this volume suggests, such a narrow focal point ignores much that provides texture and significance in the lives of Indians. It also leads to assumptions about popular values and collective motivations that are quite suspect: to the extent that collective protest figures at all in this style of history, it is seen either as popular sentiment manipulated by the elite or as a result of elite loss of control over mass action—the "insensate violence" of the "bazars and mohullas of Indian towns" (Robinson 1974:6). Yet, as the preceding chapters suggest, if the historian's focus shifts from the Western-educated elite (in any case a relatively unimportant identifier in Banaras, even for those who were) to collective activities that express group values and processes of community identity, a very different picture emerges. The historical dynamic captured by this focal point underscores the changing definitions of state and community and their interrelationship. In this view of modern Indian history, Banaras plays a key role.

This essay is meant to suggest the historical interpretations made possible through an analysis of collective activities. It is based on the premise that such actions are not "insensate" but quite rational; that they may be subjected to careful scrutiny to yield evidence about the values and motivating forces of the crowd; and that, moreover, crowd

behavior should be seen, not as peripheral to the dynamics of historical change, but at its very heart.^[1]

Given the paucity of primary sources that accurately reflect the values and perceptions of participants in collective action, particularly for the nineteenth and early twentieth centuries,^[2] those interested in the crowd must analyze its actions. Methodologically, this approach calls for a careful consideration of the full range of actions taken collectively: these possess the potential to speak to us through their shared symbolism, through the general ambiance of the public spaces they occupy, and through their methods of mobilization.^[3] As a technique the approach owes much to what Victor Turner has called "comparative symbology," in which the analyst must investigate "the relationships between symbols and the concepts, feelings, values, notions, etc., associated with them by users, interpreters or exegetes." Particularly for the historian, changes over time in the use of these symbols are central to the analysis, for the symbolic actions express "social and cultural dynamic systems, shedding and gathering meaning over time and altering in form" (Turner 1979:12–13).

Collective actions in the public spaces of an urban north Indian environment tended to be of three types: public space performances, collective ceremony, and popular protest. These forms have been dis-

cussed in the Introduction to Part 1. Although distinct in significant ways, they developed out of shared cultural assumptions and, taken together constitute a world that should be considered a coherent whole and analyzed on its own terms. This world may best be described as composed of "public arenas," or activities that represent public expressions of collective values and motivations (Freitag 1989). By analyzing the characteristics shared by the variety of types of collective actions we begin to see a consistent picture of symbolic behavioral expressions. In turn, this consistency of symbols highlights historical change when we compare a series of collective actions over time. In particular, the concept of "public arena" activity is intended to emphasize the "structure-in-process" or "process-ual" elements (to use Turner's terms) that are inherent in this world. That is, it is a world in which changes (in, for instance, the functions of particular groups and the relationships between them) affect, and become reflected in, structure that alters with each iteration. By studying the ritual symbols of this world in conjunction with other expressions of political power and cultural values, we discover what historical change is all about. Turner, too, urges that ritual symbols be studied

in relation to other "events," for symbols are essentially involved in social process . . . performances of ritual [are] distinct phases in the social processes whereby groups became adjusted to internal changes . . . and adapted to their external environment. . . . From this standpoint the ritual symbol becomes a factor in social action, a positive force in an activity field. (Turner 1979:13)^[4]

Several of the chapters in this volume treat the first of these types of public-arena activities. In this chapter I examine in more detail the latter two types, suggesting their significance in constructing a history of north India. At the heart of this processual analysis of symbolic collective behavior are events of 1809–1811. During these two years a style and a symbolic rhetoric of protest emerged in Banaras which differed in important ways from the style that evolved in other U.P. urban centers. After examining the protests of those critical two years, we look, in turn, at collective protests and ceremonials in the 1890s and 1930s in order to chart changes over time. In particular, these changes suggest

for us the significance of public arenas in accommodating important historical change, expressing through structure-in-process forms the significant alterations in relationships among urban groups, as well as between the state and its constituent communities.

Shared Culture of Public Arenas

Early-nineteenth-century sources refer, albeit briefly, to an array of collectively observed ceremonial activities, most of which were identified by the administration as "religious." Certainly their subject matter was religious, as the observances frequently reenacted stories from sacred texts. Particularly at this time, however, such ceremonials constituted statements of shared civic identity, of "Banarsipan," as much as they did a specific religious identity. The Muslim petition submitted in 1809, for instance, referring to a disputed site, said that "for some years, the lower classes of Hindus and Muhammedans have annually celebrated the marriage of the Laut and have divided the offerings between them" (Board no. 9093:168).^[5] Similarly, in referring to a site shared by a temple and a mosque, the Acting Magistrate noted that "the Muslims have frequently participated with the Hindus in the offerings presented to the idols" (Board no. 9093:262A). Even thirty years later, a British observer resident in the city commented that "on most occasions of festive and multitudinous assemblage, the distinctions of religion give way, and the scene bears more the character of a fair than of a religious meeting" (Prinsep 1831, quoted in Eck 1982:254). Indeed, the shared civic sense fostered by such public ceremonials is suggested by Acting Magistrate Bird when he noted in 1809 that "the religious ceremonies of the Muhammedans and Hindus are so inseparably blended" that any attempt to "disunite them" would constitute a "new arrangement" (see fig. 15).^[6]

This same source provided a brief description of an observance of Bharat Milap^[*] in 1809, which sketches the nature of these shared cultural activities. The story for the occasion is staged each year on a small field permanently set aside for the purpose in muhalla^[*] Nati Imli. After reenacting the reunion of Rama^[*] with his brother Bharat, the observance opens out to become a general celebration shared by thousands of people (see N. Kumar 1984:263). In 1809, although the seeds of conflict had been sown which would soon erupt in a virtually unprecedented riot between groups of Hindus and Muslims, the citywide ceremony was deemed sufficiently important that "it was amicably agreed upon to suppress and conceal their mutual differences during the cele-

[Full Size]

Fig. 15
Drawing of a tazia procession during the Muharram observance, 1831. James
Prinsep, Benares Illustrated  (Calcutta, 1833). 
Courtesy of Ames Library, University of Minnesota. (Compare with figure 13.)

bration . . . and refer them after the expiration of the holiday to the decision of the Court" (Board no. 9093:44).

Part of a preeminent occasion for collective play,^[7] this festival occupied an important place in the inventory of Banaras's public-arena activities. Perhaps the most appropriate documentation of this is found in a vernacular newspaper almost a century later, when the Bharat Jiwan commented on "the Bharat Milap of Nati Imli which is famous in this and all nations. . . . None would object perhaps to calling it the foremost of Kashi's melas and festivals, because on that day all Kashivasis—women, aged, children, Hindus, Muslims and the English—feel a rush of Rambhakti [devotion to Ram^[*] ] in their hearts . . . it would have to be an invalid or disabled person who does not go to see it" (Bharat Jiwan , 30 Oct. 1893, 8, quoted in N. Kumar 1984:264). Its eminence can be further attested to by the growth around it of a number of legends (see N. Kumar 1984).

While the occasion was doubtless particularly important to the artisans and other lower-class groups of the city (e.g., Ahirs carried in the swarup s, or actors, playing Ram, Laxman, and Sita^[*] ), the newspaper went on to note by name the important local leaders who participated by observing the event.^[8] Even today, one of the measures signifying the importance of the Bharat Milap^[*] observance is participation by the Maharaja of Banaras, who enters first as "the king" who then "becomes front rank of the audience, completing the sense of the event as that of a total people, led by their king, witnessing and worshipping the momentary arrival of their God in their midst. . . . The Maharaja, in the words of informants, is the people and the kingdom, and his exchange with Rama is symbolic of that of the necessarily anonymous masses [who] throng to the spot" (N. Kumar 1984:266).

By the middle of the nineteenth century this functional role of the Maharaja as symbol of "the people and the kingdom" had become virtually unique in U.P. As the Rohillas, the Nawabi of Awadh, and other successor states gave way to the British Empire, those who performed the integrative function described here were either dispossessed or had their status reduced and trivialized to that of mere large landlords.^[9] Nowhere else in U.P. did such a coalition (of merchants, Maharaja, and Gosains) succeed in protecting and extending the shared local culture. How did this unique situation evolve in Banaras? In the remainder of

this chapter I will argue that the answer to that question can be found by focusing on public arena activity; this focus points up a series of renegotiations of the relationship between state and community which helped preserve a particular form of local culture even while it accommodated changes in the roles and connections between leaders and followers in Banaras.

The Maharaja's own public ceremonial, as it was elaborated over two centuries, typifies the processual elements of collective observance. As was noted in the Introduction, the Hindu merchant corporation of the city, and the Bhumihar dynasty based in Ramnagar, cosponsored this unique version of Ramlila^[*] .^[10] By virtue of its thirty-one-day length and its patronage by the dynasty, this Ramlila remains, to the present, "the most extensive, best performed, and draws the largest audience of any" Ramlila in the subcontinent (Schechner and Hess 1977). Its key attributes emphasize expression of the "relationship between government, Maharaja, and ordinary people." Schechner and Hess note that the day-to-day events of the performance follow the outlines of Tulsidas's epic poem, the Ramcharitmanas^[*] , "but some events—notably those of great iconographic effect, the processions—are given higher focus: they are good theatre, and especially display the Ramlila's two leading performers, Rama and the Maharaja." This provides the widest possible appeal, incorporating all Banarsis even as it emphasizes the integrative role of the ruler.

Although the Maharaja does not figure in Tulsidas's narrative, or in the other, ubiquitous enactments of it staged throughout north India,^[11] "each day's [Ramnagar-sponsored] performance begins only after the Maharaja arrives, either on elephant or horse-drawn carriage or in his 1927 Cadillac limousine." Since the performance ceases and resumes again around the Maharaja's daily sandhya puja , it is the Maharaja who imposes, twice a day, a "processional rhythm" on the performance. Indeed, the authors note that "The Maharaja and Rama are mirror images of each other, the twin heroes of the Ramnagar Ramlila. The Maharaja is as much a mythic figure as Rama." Throughout the Ramlila, his actions and attire reflect his mythic kingly role as "upholder of reli-

gion, repository of tradition and authority," patron of the arts and learning. At the same time, he has a second mythic role, that of representative of Shiva, the lord of ancient Kashi.

The Ramnagar dynasty quite deliberately pursued certain political ends (see chapter 1) through Ramlila^[*] patronage. Important gains were made for Hindu culture in the process—especially noteworthy in the face of the political power of the (Muslim) Nawab of Awadh. But other, internally significant political ends were served as well by this elaboration of a popular observance—ends that reinforced the relations with the lower classes of Banaras, frequently Muslim, through the emphasis on the Maharaja as ruler of all. While Rama^[*] "fulfills his destiny . . . as the bearer of Hindu culture" (Schechner and Hess 1977:54), Tulsidas's equal emphasis on devotion over orthodoxy, of shared brotherhood over community and caste divisions, enabled the dynasty to use this event for integrative purposes, even with Muslim weavers. Thus the observance documents the uniquely Banarsi emphasis on the role of local ruler, with whom all Banarsis could identify. We may speculate, too, that the physical locale of the Ramnagar observance—with its permanent constructions representing various places in the subcontinent located in a space at some remove from the neighborhoods where popular forms of competition were focussed—could function in integrative ways not available to other observances of Ramlila.

Early-Nineteenth-Century State and Community

This description of collective ceremonials suggests the structural outlines of the public arena as it existed at the end of the eighteenth century in Banaras. Against this background established by integrative ceremonial, however, the early nineteenth century witnessed two unprecedented occasions of collective protest—a riot in 1809 between groups of Banarsis over sacred space, and a protest in 1810 against an innovative tax imposed by the British on Indian dwellings. By examining these two occasions, we gain a sense both of the process or dynamic that underlay structure and of the turn-of-the-century historical changes that had to be accommodated by that structure.

Sources have always identified the 1809 protest as a religious riot, an accurate characterization in terms of the targets chosen, the symbolic actions of the crowd, and the fact that groups of "Hindus" were generally arrayed against groups of "Muslims." The event left residents much shaken and was perceived at the time and thereafter as extraordinary—for such religious unrest, particularly in terms of the scale of destruction, was unprecedented in Banaras. As a petition from the qadi and mufti (the law officers of the Nizamat Adawlut, who resided in Ba-

naras) put it, this was "a disturbance of a more alarming nature than has ever been witnessed in this country . . . for three days this city was filled with rapine, fire and murder" (Board no. 9093:220).

The site of the original dispute was a neutral area between the pillar known as the Lat^[*] Bhairava and a neighboring Imambara^[*] .^[12] To underscore the neutrality of the area, weavers insisted in their 1809 petition that the idols set up on this ground were removed by the Brahmin in charge "when the Muslims met together for the purpose of prayer"; "if there happened to be any which could not be conveniently taken away, they were carefully concealed with grass" (Board no. 9093:169). While this version probably exaggerates the transitoriness of Hindu claims to the ground, it does explain the strong reaction of the weavers to the act of devotion that, in their perception, completely changed the nature of the area. For, into this shared space, came a worshipper of Hanuman who, in fulfillment of a vow, attempted to replace a sheltering mud structure around Hanuman's statue (perceived by Muslims as more "temporary" in nature) with one made of stone.

At first the weavers seemed content to appeal to the qadi, and agreed to continue participating with Hindus in the observance of Bharat Milap^[*] . At the Friday prayers that followed conclusion of Bharat Milap (October 20, 1809), however, they violently removed the stone structure and damaged nearby sacred objects. Word spread, bringing indignant Rajputs to the site. Kotwal and Acting Magistrate both rushed to the scene; they left again when it seemed that order had been restored. Instead, however, the Rajput crowd "committed some slight injury" (in the words of the Acting Magistrate) to the Imambara. The weavers' reaction drew on both their training in sword-and-stick performances (gained at physical fitness gymnasiums, or akhara s^[*] ) and on the symbolic actions they performed during Muharram: "collecting in considerable numbers armed with swords and clubs, [they] hoisted a standard and, exclaiming Imam Hossein and beating their breasts, marched towards the city" (Board no. 9093:45). Continuing their symbolic action, they marched to the temple of Vishveshvara (often transliterated Bisheshwar), which stood next to the mosque Aurangzeb had erected on the site of an older temple. Those left behind invoked religious symbols as well: after pulling down the pillar and breaking it into pieces, they then "slaughtered a [sacred] cow at the foot of the Laut and sprinkled [the broken column] with the blood of the animal" (Board no. 9093:47).

Reaction among Hindu groups in the city varied. The Raja disarmed his men voluntarily and removed them from the city during the violent period. The "brahmins and superior orders" gathered at the ghats and fasted. Rajputs and Gosains, slowly gathering force, invoked symbols appropriate to their own roles in seeking revenge. Late the next morning the Rajputs gathered together again and returned to the Imambara^[*] . After destroying it completely and murdering the four or five members of the caretaker's family, they countered the blood symbolism of the slaughtered cow by killing a hog (in Muslim eyes an unclean animal) near the principal tomb. From the Imambara they moved on to destroy the Dargah of Fatima and then turned to the weavers' quarters: they could target domestic quarters and Muslim bodies, for they did not need good relations with the Muslim lower classes of the city. Another crowd, this one composed of Gosains, concentrated instead on Muslim sacred spaces rather than their domestic quarters; it began pulling down the Gyan^[*] Vapi^[*] (often transliterated Jnana Vapi) mosque next to the Vishveshvara Temple. "The whole of Benaras," the Acting Magistrate observed, "was in the most dreadful uproar and confusion. The temples were shut, and multitudes of armed Hindus were assembled in every quarter, directing their rage chiefly against the lives and properties of the Joolahers [Julahas^[*] , i.e., weavers] and the Butchers"—indeed, "the whole quarter of the Joolahers was a scene of plunder and violence" (Board no. 9093:48–51). In contrast, the very lack of action by the Raja also proved significant from this vantage point. He remained aloof, withdrawn from the violence, an exemplar of the appropriate behavior by an integrative ruler. In the aftermath of the riot, he worked to restore peace, treating with the British administrators, on the one hand, and working with the leadership of various muhalla s^[*] , on the other.^[13]

Beyond the overt level of symbolism that classifies this conflict as "religious" lie further meanings. From this vantage point, the specific identification of the combatants, not as "Hindus" and "Muslims," but as Marathas, Rajputs, Gosains, and weavers, becomes significant. Thus, while not denying the religious characterization, we can find much additional meaning by placing these acts in the larger context of the political economy of Banaras at the turn of the century. To begin with, although the British East India Company had formally assumed control from the Raja in 1794, the frailties of the then-reigning Raja may have postponed a perception of finality regarding the transfer of power. By 1809, however, his successor had been agitating unsuccessfully for more than a decade to have the agreement put aside. He had also "lost

face" in negotiations over changes in structures of rural management.^[14] These indicators of loss of power by the Bhumihar dynasty could only have heartened both Marathas and Rajputs.^[15] The latter, as landholders who resided in the city but were supported by their holdings located throughout the surrounding district, still occupied "an important position in the district" in the nineteenth century, but they had had "their former predominance" destroyed "by the rise of the Benaras Rajas."

Yet new claims to power and leadership by Marathas and Rajputs necessarily imperiled the relationships already established in the city—both the triumvirate of influence and control exercised by the dynasty, Gosains and merchant-bankers, and the interdependence of the weavers and merchants that was expressed culturally through the figure of the Raja. I would argue that the underlying symbolism of the rioters' actions may be taken to express these configurations as much as they expressed conflict between "Hindus" and "Muslims."

During 1809, then, we can discern a first phase in the process by which symbolic structures in the city adjusted to accommodate momentous political change. For this phase, conflicts among those who held power in Banaras—which affected as well the relationship between artisans and their merchant patrons—found expression through a religionfocused symbolic vocabulary. This indicates that the participants, leaders and followers alike, reacted against other Indians perceived as competitors, in the field most open to them—that in which public arena activities were staged.

In this respect what happened in Banaras differed little from the style of urban conflict developing throughout the nineteenth century in U.P. While becoming reified in other urban centers, however, this style of symbolic activities did not come to dominate the public arenas of Banaras. To understand why, we must examine with some care the aftermath of the 1809 riot, as well as the collective protest staged just fourteen months later against imposition of a house tax. For, as the agitation died down in the city, the attendant actions of the various participants—including the British East India Company—reflected changes occurring in Banaras's structures of urban self-rule. The evi-

dence is clearest for the kotwal (police) and qadi (courts), but emerges as well for the neighborhoods and the relationship between community and state.

As was noted in the Introduction to Part 1, the functioning of the kotwal and qadi was essential to the well-being of an urban central place. The kotwal required the confidence of all communities resident in the city. In Banaras by the turn of the century, however, evidence strongly suggests that the preexisting relationship between kotwal (as representative of the state) and communities had been eroded. Perhaps the first indicators emerged in 1803 when the British administration instructed the kotwal to become involved in the administration of the phatakbandi tax. The measure, we are told by the Gazetteer , was "stoutly resisted" (Nevill 1909a:166). Significant further erosion of the kotwal's position is described in some detail in the documentation of the 1809 riots. The kotwal himself acknowledged this diminution in popular recognition of his authority by requesting that he be allowed to retire.

The shape of self-rule in Banaras was most profoundly altered, however, by administrative steps taken after the riot.^[16] First, the administration began altering both the composition and the organization of the urban police force. Second and even more significant, the administration decided that certain persons would be "brought to trial not for the outrages each had committed against the other, but one common offence against the peace . . . as an offence against the State." These persons had been selected as appropriate to stand as symbolic representatives for their communities because, "from acts and circumstances which have come to light, [these were the people who] appear to have instigated and encouraged, purely upon religious principle, the disgraceful excesses which were committed, or who were themselves personally concerned in the commission of some overt act of violence and outrage" (Board no. 9093:272–77).^[17]

These attempts by the British to alter the relationship of community and state are particularly significant. For evidence suggests that this kind of alteration was at the very heart of the collective protests of these years. The exasperated Acting Magistrate noted, for instance, that "the disturbance is found to have originated in the abuse of that privilege which the Natives have been permitted to enjoy, of assembling among themselves to deliberate on questions of common inter-

est." Further, in condemning the Rajput leader, he notes that "it is clear from the whole tenor of his conduct that he considered this a dispute in which the public authority had no business to interfere. That as the injury had been offered to the Hindus at large it was for them alone to determine the measure of their revenge and unite in the common resolution of inflicting it" (Board no. 9093:261, 295). As the conflict that erupted the following year demonstrated, the ability and inclination of Banarsi communities to gather and decide collectively on courses of action remained strong—and it directly threatened British perceptions of their state authority.

Yet, in the face of these clear messages, the British increased their interference. Thus the administration introduced the changes described earlier in the policing system; passed a proclamation that provided for closer control over collective gatherings;^[18] and in December 1810 attempted to introduce a new house-tax that would provide greater revenues for running the municipality, in the process bringing the revenues and administration of the neighborhood-based chaukidari^[*] system under closer administrative control.

These intrusions by the imperial state created a new relationship between state and community, one unappealing to the residents of Banaras. Moreover, the style of interaction differed from that demanded by Banarsis. Acting Magistrate Bird willingly worked unceasingly behind the scenes, where he repeatedly relied on personal, face-to-face relationships with the leaders of each community; but they wanted more. They wanted him to act publicly, to place his own person physically in public arena spaces for symbolic purposes. To underscore his perception of the audacity of one protest,^[19] for instance, Bird noted that

on the morning of the 24th the [leaders who became] prisoners assembled with the whole body of goshains and seating themselves upon the ghats remained there in spite of all remonstrance until the agitation occasioned by it, threatening a renewal of those horrors from which the city had recently been rescued, I was compelled to go in person to remove them; for this conduct they . . . collected not like the brahmins on the 23rd for religious principles, but for the purpose of obtained concessions. (Board no. 9093:304; emphasis added)

Renegotiating the Relationships

Reacting to the administrative innovations introduced in the wake of the 1809 riots, the House Tax Protest of 1810 provides a clear symbolic

statement of the second phase in the early-nineteenth-century process during which residents of the city reiterated the distinct Banarsi-styled relationship between state and community. In 1809, the first phase, Indians had acted out symbolic protests against other Indians. In 1810, however, these antagonists turned instead against the British, concentrating on issues that intruded into their preexisting system of urban self-rule. In the first phase, moreover, Banarsis had faced symbolic challenges to the internal relationship established between the ruling group (referred to here as the "triumvirate" of Ramnagar dynasty, merchant-bankers and Gosains) and artisans within the city; the style of the second phase suggests that the participants had resisted these challenges. In 1810, then, it is the underlying continuity of structure and protest that, together, stand out as the principal components of public arena activity.

Pre-British structures of community organization provided critical lines of mobilization to protest the house tax:^[20] while "all ranks and description of Banarsis joined in the protest" (Acting Magistrate Bird quoted in Dharampal 1971:6), the initial petitions were submitted by each neighborhood or muhalla^[*] (Boards Collections no. 7407:94; Heitler 1972:241). Occupation as well as neighborhood provided organizational networks, as the administration recognized when it tried to negotiate with occupational chaudhuri s^[*] (headmen).

Forms of protest followed time-honored merchant patterns, described as "sitting dharna." Once again taking oaths to shut down the bazaars and withdraw their services until "the outcry and distress" in the city induced the magistrate to accede, more than twenty thousand inhabitants "sat" in an open field in the city. Caste panchayats not only maintained order, they coerced those whose support was wavering for this "panch" form of protest.^[21] Symbolically, the orderly withdrawal to the field demonstrated as well to the British the effectiveness of mechanisms of urban self-rule:

Vast multitudes came forth in a state of perfect organization: each caste trade and profession occupied a distinct spot of ground, and was regulated in all its acts by the orders of its own punchayat, who invariably punished all instances of misconduct or disobedience on the part of any

of its members. This state of things continued for more than a month; and whilst the authority of the British Government was, in a manner, suspended, the influence of the punchayat was sufficient to maintain the greatest order and tranquility.^[22]

As was suggested in the Introduction to Part 2, such "combinations" became possible by working through urban linkages that overarched particular caste- and occupation-specific units. By the early nineteenth century such social integration proved capable of bringing to the field of protest representatives of most families in Banaras and its rural environs (see Dharampal 1971:13, 6). Money was raised through these same networks to support family members whose wage earners had withdrawn to the field: "The individuals of every class, contributed each in proportion to his means, to enable them to persevere, and considerable sums were thus raised."^[23]

This was, then, a popular protest, supported by virtually all levels of society and ordered by a variety of mechanisms reflecting local community organization. Even when the government decided to exempt religious buildings and lower-class dwellings, the protest remained general. Moreover, beyond the organizational level, the rhetoric and symbolic action of the protest provides important evidence of the symbolic structures brought to bear during this conflict.

To begin with, as the petitions make clear, Banarsis viewed this new taxation as an intrusion into their customary charitable practices. The city was filled, they noted, with widows, Brahmins, and other poor who were housed in structures that could not be maintained if taxes had to be paid on them as well. Threats to this practice undermined basic tenets on which Banaras had been established: "In this holy city the rajas of ancient times from a principle of virtue and in the hopes of everlasting fame built houses and fixed salaries and settled perpetual donations for the subsistence and residence of brahmins, paupers and mendicants, there are hundreds of such houses appropriated to brahmins" (Boards Collections no. 7407:106–13).

Furthermore, both rhetoric and action deliberately posed their system of self-rule against state perceptions of appropriate behavior. In their final petition to government, the protestors insisted they had not

created a "disturbance," as the magistrate charged, but had registered a complaint in a legitimate manner. They noted that

the manner and custom in this country from time immemorial is this: that, whenever any act affecting every one generally is committed by the Government, the poor, the aged, the infirm, the women, all forsake their families and their homes, expose themselves to the inclemency of the seasons and to other kinds of inconveniences, and make known their affliction and distress. (Quoted in Heitler 1972:250)

It is significant, too, that their objections to the tax centered, not on the rate at which it was pitched, but on the very legitimacy of the form of demand. Their petition noted that

in the Shera and Shaster, together with the customs of Hindostan . . . houses are reckoned one of the principal necessaries of life, and are not accounted disposable property . . . in this country, in the times of Mohamedan and Hindoo princes, houses were never rendered liable to contributions for the service of the state. (Quoted in Heitler 1972:253)

They objected, as well, to the notion that representatives of government should go into their houses to determine the value to be assessed. Finally, they pointed out that already they were paying other taxes, such as stamp, transit, and town duties; in particular they referred to the phatakbundi .

The contrast between the new house-tax and the phatakbundi was an important one. The phatakbundi operation had been organized locally within each muhalla^[*] ; the substantial sums, collected by the inhabitants themselves, then were used to pay chaukidar s—also chosen and supervised by the inhabitants. The new taxing arrangements would have represented a very substantial infringement on local self-rule. Thus the antagonists clearly perceived the conflict to be between British authority and Banarsi community control. In early January the government of the East India Company responded to Bird with the observation that Bird should announce the compromise of the phatakbundi in the way "best calculated to allay the disposition to riot and resistance to public authority, which appears so generally prevalent among the lower orders in the city of Banaras." Yet three weeks later, Bird despairingly noted that "the people still continue collected as they were," and only fatigue and disappointment were likely to break up the tent city. He was fully aware of the implications of the month-long protest: "I cannot but feel very forcibly, that such a state of things being permitted to continue in defiance of public authority, has already weakened, and weakens daily still more and more, those sentiments of respect, which it is so essential that the community should entertain for the government of the country" (Boards Collections no. 7407:121–24, 185–88).

The tactics with which Banarsi communities fought the state had been carefully designed to protect their legitimacy. The crowd's careful adherence to nonviolence tied Bird's hands, for he felt that only violence would justify use of the military. The tactics, as he noted, thus forced city life to a standstill, leaving the administration little choice but to rescind the tax. Beyond the petition cited above, we have other evidence that the crowd felt itself fully justified in its action. Another petition, with a "style and contents" characterized as "disrespectful" by the judges of the Provincial Court to which it was addressed, baldly declared the "deadly evil" that the tax represented. Indeed, the petitioners continued, "if our bad fortune be such, that you are induced to wish our leaving this for other countries, we trust that you will be pleased to order what we have expended in buildings to be paid to us out of your treasury" (Boards Collections no. 7407:203).

Demands for an appropriate state-community relationship moved beyond the actual issue of taxation, to the increasing reluctance of the new government to symbolically participate in public arenas. The protestors underscored this last grievance by their demand on January 23 that "they were willing to disperse, providing [the Acting Magistrate] came to them [in the field where they were residing] in person to request it" (Bird, 28.1.1811, quoted in Dharampal 1971:30; emphasis added). After his indignation in 1809, the magistrate flatly refused this time. Instead, he turned to the Raja of Banaras to represent the state in public. The Raja

proceeded with all the distinctions of his rank to the place where the people were collected, the mob soon listened to his exhortations, and returned to their homes, and the Rajah, selecting from among them, fifty of the persons principally concerned in the disturbance, brought them, to [the magistrate] to acknowledge their offences; while the Rajah himself interceded in their behalf and solicited [the magistrate] to endeavour to procure both for themselves and for the subject of their complaint, the indulgence of the Government. (Bird 28.1.1811, quoted in Dharampal 1971:32–33)

The role of the Raja in this process thus proved critical. While his power was doubtless bounded by the British government, his behavior matched public expectations for a traditional ruler. On the one hand, he could represent symbolically the culture shared by upper and lower classes, by Hindus and Muslims alike. On the other hand, he provided the only symbol of authority on which both the Banarsi and the British could agree—the only figure who could operate both in the reformulated system of self-rule emerging in Banaras and as an intermediary figure judged legitimate by British administrators. We might note that his symbolic usefulness continued even after he convinced the residents

to abandon their tent city. Once the crowds had dispersed and business and services resumed, the Raja returned to the magistrate. On this visit he again acted on behalf of Banarsi residents, bearing their petition addressed to the Governor General in Council.^[24]

In this second phase in Banaras, then, we find that the rituals of protest had resulted in creating a structure in which urban residents— rallied by merchants, Gosains, and Brahmins, and brought together around the figure of the Raja—moved against the imperial government. That they joined together to focus their protest, not against each other but against government measures imposed from above, reflected a society more culturally integrated than that of any other urban center in U.P. Where, in other urban centers public arena activities became increasingly focused on conflicts among communities, in Banaras they focused instead on the relationship between state and community.

This is not to say that Banaras was wracked by collective protest in the nineteenth or early twentieth centuries. On the contrary: for a variety of reasons—relating to its relative economic stability (compared, for instance, with Kanpur) and its Bhojpuri cultural continuity with the surrounding countryside (see Pandey 1983)—very few collective confrontations occurred in Banaras. Those few, however, until the 1930s followed the pattern we saw established early in the nineteenth century. They centered on issues affecting the relationship between state and community. We turn now to an examination of examples of collective protest in Banaras in 1891 and 1931 to discern further shifts in the processual character of public arena activities.

The Changing Style of Banarsi Protest:
1891 and 1913

In the 1890s, while the rest of the province rioted over the issue of Cow Protection, the collective violence in Banaras was directed at statemandated technology. The administration began introducing a new water-supply scheme into many of the larger cities of U.P.; by 1890 the attendant tax increases to fund the scheme had been levied in Banaras. This municipal taxation, we should not be surprised to learn, was "generally viewed with disfavor by the poorer classes of traders and shopkeepers. The butchers have already [by March 17, 1890] struck

work, and other classes are expected to follow suit" (Raj-ul-Akhbar , 17 March 1890; Bharat Jiwan , 2 March 1890; both in SVN). This intrusion of tax-financed technology exacerbated the perceptions of hardship which had originated with a grain shortage. Particularly hard hit by the great rise in grain prices, Muslim weavers experienced "grave" reductions "in circumstances from the falling off in the demand for those rich fabrics for which Banaras was famous." They had also approached the Collector for financial assistance; as the Officiating Commissioner noted, they suffered more in such circumstances than did others in the city, for "they have no such resources as poor Hindus who are fed in thousands daily at the various chhatras maintained by opulent Hindus from charitable motives."^[25]

While not directly related to the waterworks issue, an accompanying act of protest resulted in the city's collective self-denial of pleasure in the face of new taxation and grain scarcity. No observances were held that year of the "Water Carnival," or Burhwa Mangal, a festival in which decorated boats paraded along the river. A crowd predominantly composed of weavers ensured that this ostentatious occasion of play did not occur, by visiting the boatmen (mallah s) and warning them against participation.^[26]

Actual construction shifted public focus to the waterworks themselves. As a symbol of encroachment, the technology being applied to water proved strikingly apposite.^[27] Moreover, Banarsis watched closely the fate of a small temple contiguous to the construction site of the pumping station at Bhadauni, three miles outside of town. Early complaints noted that the engineer repeatedly entered the temple with his shoes on. Soon, however, it became apparent that construction endangered the temple itself. The Municipal Board rather cynically decided to simply permit construction to continue—thus leaving a hole all around the temple some ten feet in depth—so that the temple would "fall down of itself in the monsoon" (White Demi-official). As agitation mounted "the principal pandits and bankers of Banaras" resolved at a public meeting to send a protest memorial to the government and to seek the assistance of the Sujan Samaj of Banaras (Bharat Jiwan , 3 November and 1 December 1890, SVN), an organization of Western-

educated pleaders and clerks opposed to any form of direct taxation (White Demi-official).

Collective activities thereafter combined traditional-styled protest actions and Western-styled argumentation. At February public meetings called by the Samaj, participants protested "against the alleged improper assessment of the house tax and [urged the administration] to consider the subject of electing a better class of men as members of the Municipal Board" (a denial of the legitimacy of those who participated in British-sponsored institutions). Crowds continued to gather each day at the construction site and temple. A similar-sized body "of about 4,000 men" marched on the Collector's house to complain "of hardships caused to them by the scarcity of grain and increase of municipal taxation" (White Demi-official; Raj-ul-Akhbar 13 April, 1891, SVN).

Violence finally erupted when the Municipal Board refused to take action to protect the temple. As bankers and shopkeepers closed their storefronts, from five to six thousand people assembled at the temple site. Increasingly agitated, the crowd raised the cry "Destroy the machinery!" and pitched boiler and pump into the river. It is significant that the protest particularly targeted symbols of imperially imported technology. Within two years of major protests that, in the rest of U.P., pitted Hindus against Muslims (i.e., the Cow Protection riots of 1893), Banarsi crowds, instead, tore up water-supply pipes, then rushed through the city to destroy streetlamps, the telegraph office, and the railway station.

Some of these collective activities continued styles used in earlier protests: businesses closed, crowds marched to petition administrators, others gathered at the scene where their values seemed most directly threatened. While the crowd acted in ways somewhat different from those of a century earlier, this symbolic behavior still conveyed concern over shared values represented in public spaces. Crowds concerned with the threatened temple, for instance, included Muslim weavers, lower-class Hindus, and the Brahmins and pandas attached to the temple (White demi-official). That is, within these changing but still culturally continuous modes and alliances of protest, the rioters continued to reflect the two characteristics that distinguished Banarsi protests from those in other U.P. urban centers: (1) the state, rather than a perceived community of "Others," remained the target; and (2) public arena activities still operated in broadly inclusive ways, expressing protest still shared to a significant extent by the mercantile elite of the city and the artisan masses. Comparing this Banarsi protest of 1891 with an ostensibly similar one in Kanpur in 1913, for instance, we can see the importance of these distinguishing characteristics (for full description of Kanpur 1913 riot, see Freitag 1989, chapter 7). Where, in Banaras, a

protest that had begun over a temple had been rerouted against the technology of the state, in Kanpur the protest (over the road routed through part of a mosque) evolved into a dramatic and influential statement on the "martyrdom" of Indian Islam. Where in Banaras the symbolic enactments of protest centered on destroying streetlamps and telegraph equipment, in Kanpur they included processions of mourning and the rebuilding of a section of the mosque.

While the Banarsi style of protest remained distinctive over the nineteenth century, however, it was changing as well. The "structure-in-process" elements of symbolic public actions trace these changes. Most important, in contrasts between the agitations of 1809–1811 and that of 1891, we can begin to discern, between the lines of eyewitness descriptions, an increasing gap between the actions of the crowd and those of the Banarsi elite. Unlike 1809, for instance, descriptions of the crowd in 1891 suggest that no upper-caste Hindus joined the lower-class crowds who gathered at the temple or marched on the Collector. While there may still have been hinge individuals or groups who connected protest meetings and these crowd actions, they are less visible by the end of the century (perhaps the most prominent was Raja Shiv Prasad).

The gap between lower-class and upper-class residents of the city widened even more in the early twentieth century. Other chapters in this volume suggest that the period directly preceding the 1930s may have proved critical in this process (see Introduction to Part 1). As Kumar has documented, perhaps the most dramatic example is the Nakkatayya festival (designed around the moment in the Ramlila^[*] when Ram^[*] cuts off the nose of a demoness). From a festival of reversal in which "antisocial" behavior was expected and condoned for those who marched in the army of the demoness, it first became sanitized, then suffered the withdrawal of elite patronage. Other festivals and events were affected as well, including Burhwa Mangal and even Muharram, which in Banaras had been observed not only by Sunnis and Shias but also by Hindus (N. Kumar 1984).

This process of separation did not begin within a religious frame of reference as it generally did in the rest of U.P. Instead, connected primarily to the growth of nationalism, it was played out in public arenas as a division between upper and lower classes. Yet a large proportion of the lower-class urban population in Banaras was Muslim: most of them weavers, they lived "a hand to mouth existence . . . without exception dependent on the good will of their Hindu employers."^[28] The dramatic

shift in public arena relations, the unprecedented separation of activities patronized by the elite from those participated in by the lower classes, thus had implications for relations between religious groups as well:

For a period extending beyond the memory of living man Benaras appears to have been free of instances of severe communal disorders although periodically small disturbances have occurred owing to the close proximity of Hindu and Mahomedan sacred places. In 1930, however, signs of increasing [tension] became apparent. In that year two movements came to the fore. On the one hand there was a very pronounced Hindu movement represented by the Congress volunteers and on the other hand there was a Mahomedan movement represented by the Tanzeem organization. A violent clash between the two communities was narrowly averted at the Ram Lila at the end of September.^[29]

The ironic and sudden shift in Banaras to an apparently communal style of popular consciousness in 1931 could not be expressed in more symbolically significant terms than the 1931 riot there. The narrative is quickly told. The autumn of 1930 witnessed intense boycotting by Congress of educational establishments, which attracted a strong contingent of student supporters. These, in turn, concentrated on picketing schools, as well as cloth and liquor shops in Banaras. Agitation declined, however, after Congress was declared an illegal association in October, although intermittent pressure continued to be applied to cloth and liquor shopkeepers. In January activity resumed with enthusiasm when the Government of India released the All India Congress Committee. During the picketing in February, Congress volunteers beat a customer in the shop of Agha Muhammad Jan; he called the police in, and the volunteers were arrested. That night Agha Muhammad was severely injured on his way home. Before he died he named as his attacker a student from Banaras Hindu University, head of the Congress student volunteers. Two violent outbursts accompanied Agha Muhammad's funeral procession; one by a crowd awaiting the body near Victoria Park, another by those accompanying the body in procession. The latter had taken offense that Hindu merchants' shops remained open as they passed; their destruction of goods and shops symbolically mirrored the violence exerted by their opposites, the Congress volunteers who had picketed liquor shops. Three days later groups of Hindus retaliated during Friday prayer, choosing as their site of attack the Gyan^[*] Vapi^[*] mosque (which had attracted the attention of the Gosains in 1809).

Certainly on the face of it, this conflict looks much the same as ones

occurring in these years in Kanpur or Bareilly—involving a Hinduized Congress, alienated Muslims, and sacred spaces (see Freitag 1989, Part 3). What had happened in Banaras in the 1920s to change its own style of collective protest so that it matched, so much more closely, that of other urban centers? I would argue that two simultaneous and complementary processes had altered the relationship of state and community expressed in Banaras's public arenas. First, the question of who possessed the moral authority of the state had become clouded. While in other parts of U.P. the nationalists' claims to state power had led them to attempt to claim authority from the British colonial state, this proved more problematic in Banaras. There, the British had never succeeded in wresting moral authority from the Maharaja and his mercantile supporters. Thus the new nationalist claims challenged the previous integrative understanding achieved between the triumvirate and the lower-class residents of the city. Second, as essays in this volume and Kumar (1984) have documented, a separation of elite and popular culture had developed, which became reflected in public arena activities. The public culture of the lower classes that remained, after elite patronage had been withdrawn, revolved around identities that alternated between class and religion, depending on the particular historical context at the moment.

Not least important in this process was the impact of the depression on the interrelationship between Hindu merchants and lower-class artisans. While we have no detailed studies of this development, we do know, for instance, "that all the ills of the handloom industry . . . came to a head after the world economic crisis of 1929–30" (Fact-Finding Committee 1942:20). "The severe trade depression which has prevailed since 1929 has seriously reduced the average earning capacity of the handloom weaver . . . accentuated by the relative increase in the cost of yarn on account of the protective duty" (Tariff Board Report of 1932:70, quoted in Fact-Finding Committee 1942:20). This materially affected the relationship between weavers and the merchants who functioned as their supply sources and selling agents. To the extent that the relationship had previously included certain protections and benefits for the weavers, we may assume that this economic interdependency was imperiled by the inability of middlemen to protect weavers from economic distress; we certainly have indicators that there were shifts in this period in the shares of the market enjoyed by artisans of particular products (e.g., N. Kumar 1984:45). This social dislocation fortuitously coincided with the reintroduction of Congress nationalist agitations. As Nehru later observed, "The Civil Disobedience movement of 1930 happened to fit in[,] unbeknown to its own leaders at first, with the great world slump in industry and agriculture" (quoted in G. Pandey 1978:155).

Significantly, much of this nationalist agitation in Banaras was expressed in a very particular vocabulary, that of reformism. This development should not surprise us; from the eighteenth century the city had been dominated by a "code of piety and restraint," to use the description of Hindu-Jain merchant culture employed by Bayly (e.g., 1983:385). Banaras Hindu University served as a base for those interested in agitation in the city, hence the prominence accorded to student activists, and a recognition of the collective authority assigned to Hindu scholars in the city. Reflecting reformist values, then, nationalism in Banaras came to be expressed in efforts to curb consumption of alcohol, to remove excess from public festivals: that is, to Sanskritize expressions of Banarsi personal and collective culture. The extent to which elite patronage encouraged this development can be traced, for example, in essays in this volume on drama, the Hindi movement, and religious oratory. Arguments for sanitizing the festivals were sketched out in the pro-Congress Aj^[*] :

The aim of Lilas is to educate the public through the teaching of . . . Rama and other pure characters. . . . It is essential for such important Lilas to be pure. Two Lilas are specially splendid here: Bharat Milap and Nakkatayya. . . . Such shameful scenes should be completely stopped. The organizers of Ramlila can stop them in one day. If educated young men would make groups and plead with the makers of floats, these corrupt scenes could be done away with.^[30]

Although these values were supported by the elite of the city, they did not appeal to lower-class Muslim artisans, who, as we have noted earlier, made up almost one-fourth of the population. Their search for alternatives led them to public arena activities that focused either on lower-class interests and values (see N. Kumar 1984; chaps. 5 and 6), or on Islam. Fittingly, the form of Islam they supported also reflected reformist influence. Baba Khalil Das "for several months previous to these riots was organizing the Muslims of Banaras, mainly of the lower classes, to adhere more to Islam and to religious practices." In Benin Park and other Muslim quarters of the city, as well as in Victoria Park,^[31] the Baba organized his listeners into "processions which paraded the town til late[,] night after night singing and preaching Islam." These followers carried symbols to testify that they belonged to the Tanzim movement, including green badges, distinctive uniforms, and flags. When challenged by the magistrate—who cited Hindu complaints that Tanzim was aimed against them—the Tanzim organizers

published a list of "objectives" for the movement that emphasized social and religious reform. The government felt that it could not object.^[32]

The economically and socially unsettled context, then, had dislocated the interrelationship of Banarsi communities. These changes were played out in political terms in the competition between Congress agitations and Tanzim, which in Banaras were as much expressions of conflicting class values as they were of conflicting religious identities. Collective violence extended the competition to riots occurring around the murder of Agha Muhammad Jan; these, too, came to be seen as communally motivated. Yet such communally oriented unrest is perceived even now as an aberration in the normal pattern—as is indicated in the oral history cited by residents, who date participation by Hindus in Muharram as being commonplace "'before 1930s,' 'before the Hindu-Muslim riots' " (N. Kumar 1984:316).

Thus the processual shifts in the relationships in Banaras could reflect class or religious differences, depending on the context. But the legacy of the eighteenth century remained. In an effort to nullify such shifts, symbolic acts were organized. Staged in public arenas, these attempted to heal the rift in shared culture. In two town meetings led the following week by Pandit Malaviya, Raja Moti Chand, and others, "resolutions were passed deploring the communal riots, expressing sympathy with the sufferers and deciding to raise funds for repair of temples, mosques, shops and houses that had been damaged and for compensation of the injured." On the following night, "a large number of Mahomedans congregated in the Town Hall grounds where pan, illaichi and garlands were distributed to them by Raja Sir Moti Chand (whose idea this was), Dr. Bhagwan Das and other Hindu gentlemen. This may be regarded as the formal end of an outbreak."^[33]

Conclusion

In Banaras, then, state and public structures had been integrated symbolically through the person of the Maharaja and, as modern political forms emerged, the merchant elite allied with him. The Hinduized cultural style of the city had been shaped by the values and cultural patronage of merchants and other intermediary groups, who had maintained close ties through the hierarchically structured Banarsi society. Developments in the 1920s and 1930s strained these ties, however: among the most important of these developments were economic dislocation from the world depression, accompanied by an effort to ensure the moral authority of public arena activities through a Sanskritization

of popular culture. Alienated lower-class artisans instead developed their own collective activities, which ranged, depending on the context, from festivals staged by and for themselves to Islamic reformist activities such as Tanzim. The change should not be overstated: a shared culture of public arenas still exists in attenuated form today, in which activities featuring the Maharaja of Banaras continue to be seen as "significant" events shared by both elite and lower classes. Nevertheless, many public arena activities are created, organized, and sponsored from within the lower-class culture of the city—a measure of the changes in the processual nature of collective activities which has developed over the last century.

In this context, it is not insignificant that scholars have documented a significant expansion in recent years of a wide variety of collective activities staged publicly in Banarsi neighborhoods. As the welter of festivals around Ramlila^[*] attests, the moral authority of the public arena is shared now among competing actors: the Maharaja, the state, and those interested in expressing lower-class community identity.