by Philip Dwyer
The Marava Form listed 11 questions, whose answers were filled in by local inspectors. The information it gathered was on the kaval system in a village—whether it existed, who its beneficiaries were, and who its victims. But in gathering this information, the form assumed the exercise of ‘Marava oppression’ through the practice of kaval in the colonial countryside. The very framing of the questions thus assumed the primacy of caste identity within the village as well as an inextricable link between caste and criminality. For instance, the form asked inspectors to list the number of houses in the village by caste, the attitude of powerful castes towards Maravars, whether Marava oppression arose from kaval, whether other villagers consented to kaval or resented it, would the oppressed villagers depose against Maravars, and so on. Responses to these questions varied across the villages, followed no clear pattern, and are of little help in estimating the actual prevalence of kaval. However, they do show the categories in which police functionaries repeatedly wrote about caste and crime. 24 On several forms, the answer to whether kaval existed was a simple ‘nil.’ Yet, these same forms also contained answers that presumably were informed by the questions. For instance, the information for Mavadi village claimed twice that there was ‘no kaval system’ and yet asserted that ‘Maravars will commit any type of crime if kaval is refused.’ 25 Likewise, the form for Keezhapillayarkulam village declared that ‘there is no oppression by Maravas’ but that the village munsif Rangasubbaraya Iyer ‘is afraid of the Marava’ and that ‘somebody will come out boldly to depose against the Maravas.’ 26 For Kokkulam village, the inspector stated that there was ‘no Marava oppression and no kaval,’ even while claiming that the villagers were ‘ready to repose in police cases against Maravars if offered police backing.’ 27
The questionnaire brought together identities of caste and class, and wove them into criminal categories that permitted policing. (For example, ‘What are the names of the leading oppressive Maravars and have they any means of livelihood other than oppression?’) Concomitantly the responses often tied together these elements, enabling native appropriation of the enmeshed categories of caste, class, and criminality. For instance, Pallikkottai village had no kaval system, according to the form, but:if kaval is refused the kavalgars give unnecessary trouble by impounding cattle, maiming them or administering poison to them. They also destroy the standing crops of the villagers…Most of the Maravas in this village have no means of livelihood except by the oppression committing property crimes, threatening extortion and setting fire to the houses and will commit any type of crime. 28
Despite their standardized format, the Marava forms from Tirunelveli depicted a complex situation on the ground. Documents from senior police officers however translated this messy data into accessible statistics and neater narratives of oppressive Marava kaval that needed to be policed and uprooted. Discussing kaval in neighbouring Ramanathapuram district, the Inspector-General of the Presidency F.A. Hamilton noted in 1929 that:in the 20 villages for which statistics have been gathered, there are 256 kavalgars receiving payment annually of Rs.14310. This gives an average payment of Rs.4-8-0 a kavalgar each month. These figures will convey some idea of how formidable is the system that has grown up and has now to be displaced by the provision of a number of police sufficient to protect the villagers of Chettinad from the…intimidation and extortion of the criminals in their midst. 29
These statistics were used in a police effort to eliminate kaval and influenced the placement of stations and men across Ramanathapuram. For instance, a new station and outpost were opened in Kadaladi to monitor 132 bad characters through beats that were to be undertaken to each of the attached villages as often as thrice a week. 30
The overlapping categories of criminality and caste were also reflected in the typology of persons who needed to be policed, which included types such as the suspected criminal, kavalgar, Notified Member, Habitual Offender, Bad Character, wandering gang, and Known Depredator (referred to as the KD). 31 In some records, these classifications do not mean much—the terms are used interchangeably or KD is used as an umbrella category. But others suggest that the typology of criminals translated to police practice: some suspects warranted a closer and more frequent watch, while others could be monitored less frequently. Some were monitored through daily beats, others through nightly beats, one-day beats, twice-weekly beats, miscellaneous beats, and so on. 32 For instance, a proposal drawn up to reallocate police resources within Ramanathapuram district in 1929 included a map of Mudukulathur—one of the regions within the district. 33 The map contained the boundaries of each village in the region, and the number of people to be policed in each, broken down into numerous, seemingly fine, categories, such as Marava population, Marava conviction, Non-Marava Conviction, those Suspected in Cases, those needing a Close Watch, and those who only warranted a Non-Close Watch, with legends for each of these categories. Papangulam village, for example, had 170 MPs, 8 MCs, 1 NMC, 8 Ss, and 12 CWs. Each station in the map was staffed, and some police outposts established, upon aggregating these numbers and calculating the distance to be covered on the beats.
Calculations for police staffing were often astonishingly localized and specific to the category of criminal under surveillance . In 1930, officials budgeted for two constables for Tirupattur station, who would be sent to patrol two villages within its limits, Kandavirayanpatti and Nachiapuram, where nine and seven kavalgars, respectively, resided. 34 Likewise, they suggested increasing the strength of the Neikuppai station, whose two beat constables were fully occupied in monitoring the 15 bad characters registered in the adjoining three villages. Sikkal station had 38 bad characters and four criminal tribe members residing in five villages, and needed four constables to check them, while Sayalkudi station had 29 bad characters residing in four villages, and therefore needed four beat constables. 35 Police officers considered it vital to patrol such areas which had registered criminals and criminal suspects, however few in number. This was usually at the cost of other swathes of land—forested, mountainous, or, simply, ‘quiet’—which were perceived as not requiring a regular police beat.
Coercion on the Beat
While the policing of criminal tribe members is conspicuous in the government archive, police surveillance was not limited to notified members but, in fact, extended to the larger population too. Every police station in Madras Presidency was required to maintain a narrative record on each village within its precincts. Known as the Part IV records, these contained ‘notes on important factions and disputes, especially between castes and communities, and regarding the commissions of serious breaches of the peace…In short any information which may be useful to a new station-house officer, having no previous experience of the station, should be entered in this register.’ 36 Some police stations contained typewritten forms that comprised the opening entry of the Part IV record for a village—the form asked for the geographical position of the village, its population, (listed by community), factions, ‘miscellaneous bad characters,’ festival and market days, kaval details, and so on. Regardless of whether a form was provided for the local inspector or not, the Part IV record sought to map each village within a station’s jurisdiction in terms of its geography, communities, spaces, and times that needed extra surveillance . For example, the description for Manur village mentions the location of the village relative to neighbouring large towns, its inhabitants whose activities needed to be monitored, and festivals that needed to be policed.This is a small village situated 9 miles north of Tinnevelly on the Tinnevelly-Sankarankoil road. Pallars form the bulk of the population. 37 There is a strong ill-feeling between the VM who is an acting man from Tinnevelly and the karnam the permanent resident of Manur, in which one is trying to entangle the other in some criminal case or other. 38 Both the village officers do not cooperate with the local police. There is an ayurvedic dispensary maintained by the Tinnevelly district board. Treatment in this dispensary is offered free. There is a temple which gets an annual income of
Rs.4000 which is under the management of the Tinnevelly temple committee. One Shunmugasundram Pillai is the manager of the temple. Every year in the Tamil month of Avani a festival called Moolam [unclear] is celebrated and a lot of crowd from the neighbouring villages of other taluks visit. 2 constables are usually deputed during the festival for bundobust. 39 Pickpockets are likely to visit. The police station is the only government building in this village. There is no other thing worthy of mention. [Sd. IP, Tinnevelly]. 40
Apart from detailing spaces and people to be watched, the description also made it clear that the rest of the village (men who beat their wives? landholders who exploited labourers?) did not merit police surveillance : ‘there is no other thing worthy of mention.’ 41 Descriptions such as the one above were often not dated, conveying the impression that the record captured the image of an unchanging village. 42 For instance, the record for Pallamadai describes the village succinctly, as follows:This is a small village, consisting of a few houses of Pallas, Shepherds, Nadars and Muhammadans. Pallikottai Maravas were doing kaval for this village and now it has been completely stopped and so this village is often troubled by Pallikottai Maravars. This depends upon a fairly big tank lying near for agricultural purposes. Police informants: 1. Thalayari Sankarasubbu Thevan, 2. Velliah Kone. 43
The note is signed by the inspector but not dated. 44 The details that the inspector chose to include in this four-line note are indicative of where the police saw need for their intervention. Kaval, as mentioned earlier, was criminalized by the colonial government and policed vigorously. Tanks were an important source of irrigation, and, presumably, of conflict too, in Tirunelveli. However, even as the inspector’s note captures some of the larger concerns of the colonial state, it displays the messiness of immediate, local knowledge in its classification of the villagers: two caste categories, one occupational category, and one religious.
As opposed to other police records, which could be periodically destroyed, the Part IV records were meant to serve as a continuous record of the jurisdiction of each police station. Opening remarks like the ones cited above were followed by periodic updates (a few times a year) filled in by successive police inspectors. These were brief reports on crimes, on suspected criminals who needed surveillance , or on tensions between communities that needed to be smoothened. Together, they formed a continuous, documented history of crime, or the lack of it, in a village. For the historian, the journal offers a glimpse of moments when colonial subjects challenged authority in ways that were violent and visible to the state. For policemen, who wrote and read these records, the journal shaped their knowledge of crime and, consequently, their policing practices—specifically, the direction and frequency of their beat.
While some entries in the Part IV record indicated that there was nothing requiring immediate police attention in that village, others warned of crimes that were brewing and had to be prevented, while the rest described crimes that had had occurred and were being investigated. The cadence of these updates suggests the role of routine, coercive police authority in maintaining ‘order ’ by preventing hostilities from escalating into conflict. Such hostilities often concerned ritual or social privilege, and took the shape of competing rights of various communities over village spaces: temples, streets, and burial grounds. 45 Police intervention ensured that simmering conflicts did not boil over, sometimes over a period of several years. For instance, in December 1932 the Manur station inspector noted the ‘frequent ill-feeling’ between the Pallars (considered low, ritually ‘impure’ castes) and Maravars of Mavadi village. 46 In one instance, the Pallars put up a bund near their burial ground to prevent their corpses from being washed away during floods. The Maravars objected to this and complained to the revenue authorities, who took action against the Pallars for encroachment. The inspector noted that ‘the feeling deserves to be watched.’ Nine months later, he updated his remarks to a terse ‘No trouble in the village Mavadi.’ Presumably, the tension abated but did no die, for in 1937 he again noted that there was no sign of ill-feeling between the two groups ‘after the warning given by (him).’
Therkululam village presents another instance where the police actively intervened to prevent challenges from lower caste groups to the established spatial order . 47 In the 1940s, some Hindus of the village, very probably belonging to a lower caste, converted to Islam: they were called the Navamuslims (‘New Muslims’). The late-1940s and early-1950s witness several instances when the Navamuslims tried to assert their rights over the village spaces, only to have their attempts foiled by the police. Successively, their attempts to bury a child in a plot disputed by the Hindus, to construct a mosque close to the village church, and later to build it near the Hindu temple, were all put down by the local magistracy and police. In 1946, the police sent a daily beat to the village; a few years later, Sub-Inspector Natarajan still had to send two constables ‘to watch the events,’ and to warn the parties ‘to not take the law into their own hands.’ Presumably, the police surveillance worked this time, for the next few entries in the station records graduate from reporting that there was no trouble in the village to, four years later, ‘there is nothing important’ in this village, thus dimming it from the police radar. 48
There is no mention of overt coercion in the police journals, but there are ‘warnings,’ as in the examples above: ‘there is no sign of ill-feeling at present, after the warning given by me.’ 49 Understandably, the notes rarely mention what exactly the ‘warning’ was. It may have been a euphemism for threats, or it may have been a warning that proceedings under the security sections of the Code of Criminal Procedure, 1898 would be taken against the disputing parties, or it may have entailed a judicious combination of the two tactics. While recording a caste conflict in Thazhiyoothu village in 1937, the Inspector wrote that ‘both parties [Hindu and Christian Nadars] have been personally warned that they would be run in under 107 CPC and are quiet now,’ 50 and, eight months later, that ‘there has been no trouble in this village between the R.C. converts and the Hindus after the warning given by the C.I.’ 51 Sometimes the language describing police intervention is more conciliatory. Following a conflict in Melapillayarkulam village in 1938, the inspector noted that ‘both the parties were advised to sink their differences and live amicably. They have promised to do so.’ 52 Similarly, after intervening successfully to ‘restore goodwill’ between two groups in Chittanpacheri, the police noted that ‘the S.I. must visit this village often and be in touch with the feelings and satisfy himself that the compromise is genuine and that the parties are keeping quiet.’ But, he continued, ‘at the slightest manifestation of trouble, there should be no hesitation in taking security action…’. 53
To the historian, the vocabulary of social harmony, of the absence of ‘trouble’, seen in these police writings suggests instead the effective functioning of strong arm of the state: the maintenance, in fact, of Order . 54 Even if coercion was not overt, the very presence of policemen, the extra beats deputed to areas of ‘trouble’ may have been menacing enough to subdue protest. Station-house records indicate that the police gaze and stave were frequently, and effectively, redirected to maintain order.
Increased police surveillance may not have been simply procedural, a matter of sending a beat constable to a disturbed location. Rather, the language used in the journals emphasizes the act of watching, suggesting that surveillance was quite deliberate. For example, ‘the village should be frequently visited and the feelings watched’; 55 ‘this is a crime country…(that) requires police attention during dark nights’; 56 ‘S.I. will watch further development…’; 57 ‘S.I. will watch the situation’; 58 ‘the situation is, however, needs (sic) frequent watch’; 59 ‘S.I. shall watch the feelings between the parties’; 60 ‘Rama Koravan and his brothers… were registered as C.W. suspects…They are under close watch. This village requires very close attention during dark nights…’; 61 and so on. The use of the acronym ‘C.W.’ (Close Watch) suggests that the typology of
surveillance mentioned earlier was not restricted to policy, but also translated to practice. More broadly, the detailed, but relatively distant, plans made by senior police officers to ‘manage populations’ towards ensuring the smooth functioning of the colonial economy transform in these journals into the more immediate, bodily practices of the local Sub-Inspector. 62