² The terminology is Williams, Raymond'. The Sociology of Culture (New York: Schocken Books, 1981), p. 13.Google Scholar

³ Studies of ‘British realms of action’ have included not only the institutions and praxis of British administrators themselves, but also those of a westernized Indian elite who operated through the institutions of rule established by the British. Studies of ‘Indian realms of action’ have examined cultural and religious activities, as well as ethnographic constructions of Indian society.Google Scholar

⁴ Hay, Douglas, ‘Property, Authority and the Criminal Law’ in Hay, et al. ), Albion's Fatal Tree: Crime and Society in Eighteenth Century England (NY: Pantheon Books, 1975), p. 25. For discussion of the rather different treatment of property in the colonial criminal system, see footnote 16.Google Scholar

⁵ Our attention has been called to the subject by Michel Foucault, in works such as Discipline and Punish. More to the point here, however, are studies that take into account not just the texts penned by the ruling class, but the context in which these texts were produced. See the seminal study by Hay, et al. (eds), Albion's Fatal TreeGoogle Scholar, as well as the more recent works by writers such as Sharpe, J. A., Crime in Early Modern England 1550–1750 (London: Longman, 1984)Google Scholar and Beattie, J. M., Crime and the Courts in England (Princeton University Press, 1986).Google Scholar

⁶ Sharpe, Crime in Early Modern England, p. 143.Google Scholar

⁷ This process of domestication has been the subject of several studies. See, for instance, Metcalf, T. R., Land, Landlords and the British Raj: Northern India in the Nineteenth Century (Berkeley: University of California Press, 1979Google Scholar; Pamela, Price, ‘Resources and Rule in Zamindari South India, 1802–1903’, PhD dissertation for the University of Wisconsin-Madison, 1979Google Scholar; Nicholas, Dirks, The Hollow Crown (Cambridge University Press, 1988.).Google Scholar

⁸ See Arnold's discussion of ‘colonialist writers like Sir Percival Griffiths’ who ‘equated the police with the “rule of law” and the “firm establishment of law and order” under the British’. Arnold, David, Police Power and Colonial Rule: Madras 1859–1947 (New Delhi: Oxford University Press, 1986), p. 3.Google ScholarArnold, is quoting from Griffiths, , To Guard My People: The History of the Indian Police (London: 1971), p. 2—a volume based primarily on reminiscences of members of the colonial police (the primary materials for this volume have been preserved in the India Office Library).Google Scholar

⁹ Washbrook, D. A., ‘Law, State and Agrarian Society in Colonial India’, Modern Asian Studies 15:3 (1981): 649–721.CrossRefGoogle Scholar

¹⁰ Washbrook, ibid, pp. 652–3. We might also note that Washbrook's analysis goes some way toward explaining why only the civil law proved effective in protecting property in India. The major distinction one finds between the European studies of criminality and those of crime in a colonized area seems to be the lack of significant state commitment to protecting private property against criminal depredations. Laws were enacted, but relatively few resources or policy concerns were directed to the issue. But that is the subject for another essay.

¹¹ Rulers, Townsmen and Bazaars: North Indian Society in the Age of British Expansion, 1770–1870 (Cambridge: Cambridge University Press, 1983).Google Scholar

¹² See not only the extended discussion in Bayly of this theme as it relates to merchants, but also the examples of ‘little kingdom’ behavior of landed magnates provided by Pamela Price, Nicholas Dirks, Thomas Metcalf et al.Google Scholar

¹³ The implications of this division of function is explored in much greater detail in Freitag, , Collective Action and Community: Public Arenas and the Development of Communalism in North India (Berkeley: University of California Press, 1989).Google Scholar

¹⁴ Hay, et al. (eds), Albion's Fatal Tree, p. 53.Google Scholar

¹⁵ Bayly, , Rulers, Townsmen and Bazaars, pp. 265, 301, 295, 371, 318.Google Scholar

¹⁶ Gordon, Stewart N., ‘Scarf and Sword: Thugs, Marauders, and State-formation in 18th Century Malwa’, Indian Economic and Social History Review 4 (12 1969).Google Scholar

¹⁷ I have yet to analyze the actual relationship of this brief system to popular Hinduism; it has been suggested to me that key aspects seem to reflect widely-held beliefs and practices of lower caste villagers.Google Scholar

¹⁸ Although we must note that recruitment of individuals occurred frequently, and gangs often included a variety of caste and ethnic types. Nevertheless, some gang members could trace their thag connections back several generations.Google Scholar

¹⁹ See the files for the T & D in the National Archives, Home Dept; the Hervey reminiscences; and the various volumes by Sleeman on Thags, Budhuks, Megpunnaism (child theft), etc.Google Scholar

²⁰ SirWilliam, Sleeman Henry, Ramaseeana, or a Vocabulary of the Peculiar Language Used by Thugs… (Calcutta: G. H. Huttmann, Military Orphan Press, 1836), pp. 32–3.Google Scholar

²¹ Regarding social banditry, it seems more likely that the real peasants who were neighbors to thag gang members suffered by having to work thag lands as well as their own. See the proud testimony of thags in Sleeman, Ramaseeana.Google Scholar

²² See, for instance, Sleeman, W. H., Report of the Depradations Committed by the Thug Gangs (Calcutta 1840) for discussion of landlord–thag relations.Google Scholar

²³ See Price, ‘Resources and Rule …’ for the cultural value of litigiousness for ‘Little Kingdoms’, and Bernard, Cohn, ‘Representing Authority in Victorian India’ in Cohn, , An Anthropologist Among the Historians and Other Essays (Delhi: Oxford University Press, 1987) for discussion of the Durbar.Google Scholar

²⁴ General Charles, Hervey, Some Records of Crime…being the diary of a year [1862] (London: 1892), Vol. 1, p. 27.Google Scholar

²⁵ See, for instance, Taylor, C. Meadows, Confessions of a Thug… This is not to argue that more negative subconscious formulations did not underlie this treatment of the Indian ‘Other’. But here I am focusing particularly on the more marked contrast between positive images evoked for thags and those created later in the century for ‘criminal tribes’.Google Scholar

²⁶ There were earlier, more limited, experiments that set the pattern for the Thagi and Dacoity infrastructure–particularly in Bengal. An informer system bearing some similarity to this had also been developed in England. See Freitag, , ‘Collective Crime and Authority’ in Yang, (ed.), Crime and Criminality in British India (Tucson: University of Arizona Press, 1985).Google Scholar

²⁷ [No author] The Thugs or Phansigars of India, comprising a history of the rise and progress of that extraordinary fraternity of assassins…compiled from original and authentic documents published by Captain W. H. Sleeman (Philadelphia: Carey and Hart, 1839), vol. II, p. 23 emphasis in the original. (Apparently this is an American abridgment of Sleeman's Ramaseeana perhaps with extracts from other Sleeman volumes. My thanks to Tony Stewart for finding this volume for me.)Google Scholar

²⁸ Ibid, pp. 125, 124, 128. The ‘more or less’ refers to the fact that the sword, dagger and box were found, but not the bodies.

²⁹ Ibid, pp. 133–4. As Shahid Amin has noted, the full implication of the approver in the crime committed, and his ability to add telling ‘facts’ with corroborating ‘evidence’ were critical elements in winning approver status. ‘Approver's Testimony, Judicial Discourse: The Case of Chauri Chaura’, in Guha, Ranajit (ed.), Subaltern Studies V.Google Scholar

³⁰ Shahid Amin, ibid. It is important to note that Amin's analysis is about a much matured form of approver testimony, adapted for use against nationalists, and thus all of his arguments would not hold for this initial form of the mode. Nevertheless, his general analysis provides many sophisticated insights.

³¹ And became perpetuated. See discussion, below, on the late nineteenth century rediscovery of ‘ethnographic evidence’ regarding Thag belief systems.Google Scholar

³² Sleeman, , Ramaseeana, pp. 141–2, 187 (emphasis added).Google Scholar

³³ The non-crime related ramifications of this cultural division of labor are explored in more detail in Freitag, , Collective Action and Community: Public Arenas and the Development of Communalism in North India (Berkeley: University of California Press, 1989).Google Scholar

³⁴ See, for instance, From Sepoy to Subedar for both observations on the peace and prosperity brought by the British and, through implication, of the elaboration of alternative employment for those who might have taken up some form of banditry in the 18th century.Google Scholar

³⁵ See Edith, Branstedter, ‘Human Sacrifice and British–Kond Relations, 1759–1862’ and Stewart Gordon, ‘Bhils and the Idea of a Criminal Tribe in Nineteenth Century India’ in Yang, (ed.), Crime and Criminality in British India (Tucson: University of Arizona Press for AAS Monograph Series No. XLII, 1985).Google Scholar

³⁶ Among the most frequently cited is H. M. Eliot. We may take as representative his handwritten notes; most revealing are citations such as the following, which punctuate his encyclopedia on Indian groups: ‘Mohur Sing, a very intelligent man of – – says that there are 96 gotras of Togas and 1600 of Gaurs of Hariana’. Thus the process of selection by the British was twofold-in terms of both the subject matter and the informants deemed reliable-but, nevertheless, ultimately had to rely on the information provided by the elite stratum with which such administrators interacted.Google Scholar

³⁷ Amin, Shahid, ‘Introduction’, to William Crooke, A Glossary of North Indian Peasant Life (Delhi: Oxford University Press, 1989). See also Chaube's contributions to North Indian Notes and Queries at the end of the nineteenth century.Google Scholar

³⁸ Thornton, Edwarde, Illustrations of the History and Practices of the Thugs (London: William H. Allen & Co., 1837)Google Scholar; IOL & R S V 248, North Indian Notes and Queries, vol. I (1891–1892).Google Scholar

³⁹ See especially, Yang, Anand Y., ‘Dangerous Castes and Tribes: The Criminal Tribes Act and the Magahiya Doms of Northeast India’, in Yang, (ed.), Crime and Criminality, pp. 89–107Google Scholar; Arnold, David, ‘“Criminal Tribes” and “Martial Races”: Crime and Social Control in Colonial India’, paper presented to Seminar on comparative Commonwealth Social History, University of London, 1984; Sandria B. Freitag, ‘Collective Crime and Social Engineering: The Treatment of Criminal Tribes in Nineteenth Century British India’, paper presented to the annual Midwestern Conference on South Asia, Madison, Wisconsin, November 1982. The thagi precedent proved all-important in providing certain leaps of legal logic that could be called upon when designing the Criminal Tribes Act, which provided not only for a process whereby the crimes of a few could be cited to establish the guilt of many, but which also stipulated the need to specify the administrative remedies to be adopted by the provincial government (these ranged from registration and surveillance to incarceration in colonies). Despite the structural continuity with the innovations introduced by the Thagi and Dacoity Department, and the rhetorical links made by administrators between the thags and the criminal tribes, the law nevertheless conveyed the impression that these were intermittent strategies designed to deal with problems that arose sporadically. In part the very ‘P.R.’ generated by Sleeman required this approach, for he had conveyed the impression that the British Raj had ‘solved’ the problem of thags, and thus the criminal tribes had to be seen as a new problem. In part it also related to the uneasiness of those concerned to replicate the English ‘rule of law’ in India; these opponents, located in both the administration and the judiciary, had resisted attempts in the 1850s and 1860s to categorize and deal with entire groups of people as criminal. Such a system did triumph in the 1870s, as the NWP, Oudh, and Punjab governments pushed to define groups as criminal by reference to their caste identity-a legal characterization that rendered ‘crime’ a genetically transmitted trait that could be adduced for collectivities of people. But more than genetics was used in this definition; this essay suggests the importance of space and place in shaping the relationship between state and community in British India.Google Scholar

⁴⁰ Letter from Woodburn, J., Chief Secy to Govt, NWP & O, to Secy to the GOI, Home Dept, dated Allahabad, 17 December 1889, p. 2. IOR. L/P/6/291 (file 2250 of 1890).Google Scholar

⁴¹ Letter from Ollivant, Colonel A., Senior Deputy Inspector-General of Police, NWP & O, to Personal Asst. to the Inspector-General of Police, NWP & O, dated 27 February 1887, L/P/6/291, file 2250 of 1890, p. 1. The arbitrariness of the characterization may be noted from Ollivant's following point, when he asks: ‘The question now arises under what name we should in future reports speak of these people?’ After detailing difficulties, he suggests that because Aligarh is the headquarters of the group and in that area they are known as Sansiahs, and because ‘Mr. Warburton's inquiries have given some notoriety to the name… it is better to adhere to the title of Sansiahs.’ It is not possible from these records to gauge the accuracy of the administration's opinion that for all practical purposes, these subgroups did, indeed, constitute a single community.Google Scholar

⁴² Ibid., pp. 3, 6.

⁴³ See Freitag, S., ‘Collective Crime and Authority in North India’, in Yang, (ed.), Crime and Criminality, pp. 140–64.Google Scholar

⁴⁴ Kela, Bhagwan Das, Human Adish Jatiyan, 1950.Google Scholar

⁴⁵ Bhagwan Das Kela, ibid. Thanks to Sanjay Sharma for providing several proverbs: personal communication, February 7, 1987.

⁴⁶ Treatment of the Sansiahs seems to have been strikingly typical. See Stuart Blackburn's reconstruction of the similar history of the South Indian Kallars, ‘The Kallars: A Tamil “Criminal Tribe” Reconsidered’, South Asia 1978, pp. 38–49.Google Scholar

⁴⁷ Descriptions of Sansiahs in a letter from the Inspector General of Police to Chief Secretary NWP & O. IOR. NWP & O Police Proceedings for 1891, vol. 3839, April A proceedings no. 34–42.Google Scholar

⁴⁸ For the profound implications of the concept of ‘tribe’ as invented and invoked by the Punjab provincial government, see Gilmartin, David, Empire and Islam (Berkeley: University of California Press, 1988), chapter one.Google Scholar

⁴⁹ See the full discussion in Yang, A., ‘The Making of a Criminal Tribe’, in Yang, (ed.), Crime and Criminality.Google Scholar

⁵⁰ See B proceedings for December 1901, U.P. Police Proceedings for abolition of this system. The implications of anthropometry are discussed in detail in Steven Jay Gould, The Mismeasure of Man(NY:W. W. Norton & Co., 1981).Google Scholar

⁵¹ Risley, H. H., The People of India.Google Scholar

⁵² Gould, , The Mismeasure of Man, p. 132.Google Scholar

⁵³ Why the provincial governments of UP, Punjab and Bombay used this legislation so much more heavily than did the other areas of British India is a question well worth exploring in greater detail.Google Scholar

⁵⁴ Letter from Woodburn, J. (see note 40).Google Scholar

⁵⁵ Colonel Ollivant (see note 41), pp. 1, 6.Google Scholar

⁵⁶ It is important to note that the British were unwilling to exercise this control without putting in place the legal structure to make it appear that this ‘covert’ mechanism was, indeed, part of the rule of law. See, for instance, the documents in which the overzealous local officer suffered severe condemnation by the provincial government when he attempted to invent criminal tribe controls prior to creation of the Act. IOL. P/235/vol. 53A, NWP Police progs dated 12 August 1865, index no. 7, progs no. 24 & 25; index no 12, progs nos 2 & 3.Google Scholar

⁵⁷ Muzaffanagar: Gidhias, Kanjars; Meerut: Gidhias; Muthra: Radhua Kanjurs; no alias groups in Aligarh.Google Scholar

⁵⁸ See, for instance, Freitag, ‘Collective Crime and Social Engineering’ (see note 39).Google Scholar

⁵⁹ The phrase is used later for another such reformatory at Chunar. UP State Archives. File no. 283 of 1914. Printed as UP Government Police Proceedings nos3 & 4, March 1915.Google Scholar

⁶⁰ UP State Archives, File 796 for 1901, Box 5, Police Dept NWP & O. Progs for April 1897, nos 29–97.Google Scholar

⁶¹ See, for example, UP Police B Progs for April 1891 no. 13; January 1892, no 28; and July 1893, no. 22.Google Scholar

⁶² These families were taken from the prison settlement at Sultanpur, and were replaced in the settlement by ‘planted out’ families. This round-robin of ‘criminal’ bodies surely contradicted the original distinction between ‘more’ and ‘less’ dangerous Sansiahs.Google Scholar

⁶³ UP State Archives. Home Police. Box 58 I. File no. 715A. Notes & Orders. ‘Settlements of the Sansiahs’.Google Scholar

⁶⁴ Note by J. J. D. LaTouche, dated 9–4–96, ibid.

⁶⁵ Note by Secretary to Lt-Governor, dated 14–5–96, ibid.

⁶⁶ For full documentation of this event, see IOR. NWP & O. Provincial Police Proceedings for 1897, vol. 5136 for February, April, July, Augusts and November. The ‘Notes’ kept with these printed proceedings are located in the UP State Archives, File 796C for 1901, Box 5, Police Department NWP & O Progs for April 1897, nos 29–97.Google Scholar

⁶⁷ See note from F. C. d'M. dated 19–1–97, and Note to Secy from R. G., same date, ibid.

⁶⁸ Letter from IG Prisons dated 11–10–97, ibid. In the end, approval was given not on these embarrassing grounds, but because Kumbha's place as sweeper had been taken by a Dom (member of another ‘criminal tribe’), and the support the district was now paying Kumbha's extended family was reckoned a ‘dead loss to the state’.

⁶⁹ Letter from Under Secretary (Judicial), dated 6–4–96, ibid., p. 3.

⁷⁰ IOR. P/8903. UP Police Progs for November 1912, progs no 7, serial no. 28, p. 4.Google Scholar

⁷¹ UP State Archives, file no. 68 of 1911. Printed Police A Proceedings for July 1913, no. 118–62 (pp. 115–38); June 1915, pp. 49–57; November 1915, pp. 39–40.Google Scholar

⁷² In this case, by using his visit to the area as the rallying point for an attack on a police thana in which some 200 rioters killed the local officers. See especially Amin, Shahid, ‘Gandhi as Mahatma: Gorakhpur District, Eastern UP, 1921–2’ in Guha, (ed.), Subaltern Studies III.Google Scholar

⁷³ Quoted in Amin, ‘Approver's Testimony, Judicial discourse’, pp. 191, 199.Google Scholar

⁷⁴ Ibid., p. 186.