1. See Bernard Cohn, “The Role of the Gosains in the Economy of Eighteenth and Nineteenth-Century Upper India,” Indian Economic and Social History Review 1, no. 4 (1964): 175–82; Dirk Kolff, “Sanyasi Trader-Soldiers,” Indian Economic and Social History Review 8 (1971): 213–20; and Christopher Bayly, Rulers, Townsmen and Bazaars: North Indian Society in the Age of British Expansion, 1770–1870 (Cambridge: Cambridge University Press, 1983), esp. 125–44.

2. Bayly, Rulers, Townsmen and Bazaars, 242.

3. Sushobhan C. Sarkar, “A Note on Puran Giri Gosain,” Bengal Past & Present 43 (April-June 1932): 83–87; see also Jonathan Duncan, “An Account of Two Fakeers, with their Portraits,” Asiatic Researches 5 (1808): 45–46.

4. See Suranjan Chatterjee, “New Reflections on the Sannyasi, Fakir and Peasants’ War,” Economic and Political Weekly 19, no. 4 (28 January 1984); and Atis K. Dasgupta, The Fakir and Sannyasi Uprisings (Calcutta: K. P. Bagchi, 1992), 34–40.

5. See David Lorenzen, “Warrior Ascetics in Indian History,” Journal of the American Oriental Society 98, no. 1 (1978): 72–75, who examines briefly the underlying causes of the conflict. Jamini Mohan Ghosh, Sanyasi and Fakir Raiders in Bengal (Calcutta: Bengal Secretariat Book Depot, 1930), provides a narrative of the skirmishes primarily from the imperial military perspective.

6. Ms. letter from then Captain James Rennell, dated 30 August 1766, in possession of his grandson Major Rodd, cited in Col. Henry Yule and A. C. Burnell, Hobson-Jobson: A Glossary of Colloquial Anglo-Indian Words and Phrases, and of Kindred Terms, Etymological, Historical, Geographical and Discursive (London, 1903), s.v. “Sunyasee” (p. 872).

7. Of course, the late nineteenth-century novelist Bankim Chandra Chattopadhyay took a keen interest in the conflict as a Bengali nationalist, the result of which was the novel Anandamatha (1882; first translated into English as The Abbey of Bliss by Nares Chandra Sen-Gupta [Calcutta: P. M. Neogi, 1906]), a portrayal of sanyasi soldiers as prototypical Indian freedom fighters. On “rebellion” as a misnomer, see Lorenzen, “Warrior Ascetics in Indian History,” 72–75.

8. Hence the term naga (Sanskrit nagna and Hindi nanga, meaning “naked”) by which they are generally known. However, nakedness was and continues to be a bone of contention between Shaiva and Vaishnava nagas at the kumbh and elsewhere. See Surajit Sinha and Baidyanath Saraswati, Ascetics of Kashi: An Anthropological Exploration (Varanasi: N. K. Bose Memorial Foundation, 1978), 121–22. There are also indications that the term is related to nag, or snake, especially given the symbolic importance of the snake to both Shaiva and Vaishnava arms.

9. Government of India, Foreign Department, Secret Branch proceedings, nos. 5 and 6 of 21 January 1773, National Archives of India, New Delhi.

10. W. H. McLeod, The Evolution of the Sikh Community: Five Essays (Delhi: Oxford University Press, 1975), 1–19. McLeod argues that the arming of Sikhs was not a consequence of unilateral directives by the high-caste Khatri leadership (particularly the tenth guru, Govind Singh) in response to Mughal persecutions in the seventeenth and early eighteenth centuries, but occurred over a much longer period and was the result of social and not political change. By deemphasizing the political context, however, McLeod fails to explain why Jat peasants were armed in the first place; instead he assumes the a priori existence of a proud martial tradition and vigorous physical culture among early medieval Jats prior to their attraction to Sikhism. (Here McLeod relies on Irfan Habib, “Presidential Address,” Proceedings of the Punjab History Conference 1971 [Patiala: Punjabi University, 1972], 49–54.) McLeod thus asserts that “we may be sure that the Jats did not enter the Panth empty-handed. They would have been bearing arms many years before Guru Arjan died in Lahore [in Mughal captivity under mysterious circumstances, in 1606]” (12). This argument has received as well some criticism from within the Sikh community, since it runs counter to the accepted hagiography of the gurus, particularly that of the tenth guru, Govind Singh. On his critics, see McLeod, The Sikhs: History, Religion, and Society (New York: Columbia University Press, 1989), 40, 126.

11. J. N. Farquhar, “The Fighting Ascetics of India,” The Bulletin of the John Rylands Library 9 (1925): 444; and “The Organization of the Sannyasis of the Vedanta,” Journal of the Royal Asiatic Society, July 1925, 483 n. 1. Ghurye, Indian Sadhus, 92, 101, agrees that shudras provided the main breeding ground for warrior sanyasis. This continues to be the general perception among sanyasis, as indicated by Ramchandra Giri, mahant of Damami math near Sitamarhi in north Bihar, in conversation with the author, 25 November 1994. For challenges to the communal basis of the arming of Shaivas, see Ghurye, Indian Sadhus, 108–9; cf. also the version given in Sinha and Saraswati, Ascetics of Kashi, 94–95. The Dasnami order is said to have been founded by the ninth-century south Indian philosopher-theologian Shankaracharya.

12. For example, even though Swami Sadananda Giri, Society and Sannyasin [A History of the Dasnami Sannyasins] (Rishikesh: Swami Sadananda Giri, 1976), 26, disputes the assertion that shudras were recruited into the order as warriors, he admits nonetheless a distinct scorn on the part of orthodox Dasnamis for the extant akharas of soldier sadhus (37).

13. Ramanand’s status as a follower of Ramanujacharya became a contentious issue in the early twentieth century, at which time a radical faction succeeded in transforming the hagiography of the order by removing Ramanujacharya from the preceptor genealogy, or guru-parampara. See chapter 2 for an extended discussion of the debate surrounding this ideological change and its social and cultural dimensions.

14. For the Galta tradition, I rely on Richard Burghart, “The Founding of the Ramanandi Sect,” Ethnohistory 25, no. 2 (Spring 1978): 129–31.

15. Cf. Burghart, “Founding of the Ramanandi Sect,” 131. Two of the banned disciples of Ramanand, namely, Kabir and Ravidas (and particularly the former), are known for their pointed critiques of caste hierarchy.

16. Van der Veer, Gods on Earth, 139; B. P. Sinha, Ram Bhakti men Rasika Sampraday [The Rasika sampraday in Ram worship] (Balrampur: Avadh Sahitya Mandir, 1957), 119–21.

17. On the social dimensions of nonmonastic military recruitment during this period, see Kolff, Naukar, Rajput and Sepoy.

18. Van der Veer, Gods on Earth, 88.

19. See Monika Thiel-Horstmann, “Warrior Ascetics in Eighteenth-Century Rajasthan and the Religious Policy of Jai Singh II” (unpublished essay, no date). I am grateful to the author for providing me a copy of this essay.

20. In addition to Thiel-Horstmann, see Gopal Narayan Bahura and Chandramani Singh, Catalogue of Historical Documents in the Kapad Dwara [royal warehouse], Jaipur (Amber-Jaipur: Jaigarh Public Charitable Trust, 1988), v–vii, on Jai Singh II’s growing attraction—which, in part, accounts for his interest in Vaishnava affairs—to the Bengali Vaishnavism of Chaitanya and the Gauriya Vaishnava Goswamis of Vrindaban.

21. Documents nos. 1176 (undated, from the nine Ramanandi mahants), 1483 (also undated), 1277 (referring to bairagis, dated 29 April 1733), and 1275 (dated 28 March 1736), listed in Bahura and Singh, Catalogue of Historical Documents in the Kapad Dwara, Jaipur.

22. See nos. 1506, 1507, 1518, and 1520 (all undated), ibid. The last of these, in which several Ramanandi mahants make a series of specific pledges to Jai Singh II, is reproduced as well in A. K. Roy, History of the Jaipur City (New Delhi: Manohar, 1978), 26, and is discussed in Thiel-Horstmann, “Warrior Ascetics in Eighteenth-Century Rajasthan.”

23. In this argument, soldier monasticism and the entry of the lowly into the monastic orders would have been joined in the institutional memory of the monastic orders at some later date. Pursuing this line of reasoning would require great care in amassing, and dating, oral tradition, since such tradition would be held to reveal more about the contemporary constitution of the orders than about recruitment alleged to have occurred in the distant past. I should note that I am raising all these questions in current research on the history and historiography of armed monks in northern India.

24. Perhaps the strongest indication of their continued relevance in the religious life of northern India is the central ritual role played by the military akharas during the main bathing days of the kumbh mela, India’s premier pilgrimage festival, which lasts a month and takes place every three years, alternating between Hardwar, Prayag (Allahabad), Ujjain, and Nasik. During the last maha (great) kumbh at Prayag (in early 1989), the main day for immersion in the triveni (the confluence of the Ganga, Yamuna, and subterranean Saraswati) fell on 6 February: approximately five million people, including most importantly tens of thousands of naga sadhus whose akhara processions are the centerpiece of the event, immersed themselves in the sacred water on that day alone; the total festival population was over ten million. See “Pilgrims Pouring in for Holy Dip,” Hindustan Times, 6 February 1989, 10; and “Over a Crore for Kumbh,” Hindustan Times, 13 January 1989, 9, for details of important bathing dates and astrological calculations.

25. For a discussion of this typology, see Peter van der Veer, “Taming the Ascetic: Devotionalism in a Hindu Monastic Order,” Man, n.s., 22, no. 4 (December 1987): 680–95.

26. W. G. Orr, “Armed Religious Ascetics in Northern India,” Bulletin of the John Rylands Library 25 (1940): 96. This adage appears in slightly different form in a collection of proverbs by C. E. Luard, compiled around 1900. Luard, “Central Indian Proverbs,” Mss.Eur.E.139, Oriental and India Office Collection, British Library, London, ff. 381–82.

27. This is not to say that such documents do not exist. To the contrary, substantial records are housed in monastic institutions throughout the north; the issue is primarily one of access, particularly to records held by the military akharas. For example, van der Veer, Gods on Earth, 154, reports that he was refused access to the papers in the Hanuman Garhi, the main naga institution in Ayodhya.

28. Francis Buchanan, An Account of the District of Purnea in 1809–1810 (Patna: Bihar and Orissa Research Society, 1928; reprint, New Delhi: Usha Jain, 1986); An Account of the District of Bhagalpur in 1810–1811 (Patna: Bihar and Orissa Research Society, 1939); An Account of the Districts of Bihar and Patna in 1811–1812, 2 vols. (Patna: Bihar and Orissa Research Society, 1934; reprint, New Delhi: Usha Jain, 1986); An Account of the District of Shahabad in 1812–1813 (Patna: Bihar and Orissa Research Society, 1934; reprint, New Delhi: Usha Jain, 1986); and “An Account of the Northern Part of the District of Gorakhpur, 1812,” Buchanan-Hamilton Papers, Mss.Eur.D.91–92, Oriental and India Office Collection, British Library, London. Given the dates for the Shahabad account, the true dates for Gorakhpur are probably 1813–14. See Appendix 1 for a discussion of Buchanan’s life and work. The accounts of the Bengal (proper) and Gorakhpur districts have never been published in their entirety, though substantial extracts were published in Robert Montgomery Martin, The History, Antiquities, Topography, and Statistics of Eastern India. . . . , vol. 1: Behar and Shahabad; vol. 2: Bhagulpoor, Goruckpoor, and Dinajapoor; vol. 3: Puraniya, Rongopoor, and Assam (London: Wm. H. Allen, 1838). I refer to this work below as Eastern India.

29. See, for example, Buchanan, Bihar and Patna, 1811–1812, 1:57–262 (“Topography of the Division”), and 313–85 (“Of the Hindus”); and Shahabad, 1812–1813, 38–151 (“Topography of the Division”), and 182–226 (“Of the Hindus”). In each work, the sections entitled “Of the Hindus” include the discussions of both caste and sect. It should also be noted that Buchanan’s interest in the religious dimensions of north Indian life seemed to increase as he progressed from Bengal into Bihar and eastern Uttar Pradesh. As evidence one need only peruse his Purnea, 1809–1810, which offers comparatively less detail on sects, gurus, and belief; in the subsequent accounts westward, Buchanan devoted much greater attention to such matters and even provided detailed numerical data, which I discuss below. Buchanan’s seeming lack of interest in such matters in his accounts of Purnea and of districts in Bengal proper may have been due to the Bengali pandits who accompanied him, who would have seen no reason to comment on religious details familiar to them in Bengal; once they entered Bihar and eastern Uttar Pradesh they may have had occasion to remark on the dramatic differences in monastic tradition and forms of worship.

30. Buchanan, Bihar and Patna, 1811–1812, 1:375. Dasnamis also sought service outside British territory after 1800, as was clear by the brief visit of Kamptagiri’s force of fifteen hundred mounted soldiers to the encampment of the Maratha Mahadji Scindia in Malwa in 1809. See Thomas D. Broughton, Letters Written in a Maratha Camp in 1809 (London: J. Murray, 1813), 129. Kamptagiri was a disciple of Kanchangiri, himself a disciple of the celebrated gosain commander Anupgiri who served the nawabs of Awadh and others in the latter half of the eighteenth century.

31. H. H. Wilson, Sketch of the Religious Sects of the Hindus (Calcutta: Bishop’s College Press, 1846; reprint, New Delhi: Cosmo, 1977; first published 1828–31), 53. He added, however, that “there are, it is true, exceptions to this innocuous character, and robberies, and murders have been traced to these religious establishments” (53–54).

32. On rasiks as part of a threefold typology of Ramanandi asceticism (along with naga and tyagi), see van der Veer, Gods on Earth, 159–72; on rasiks as practitioners of Vaishnava devotion and performance, see Lutgendorf, The Life of a Text, 309–21. Both Lutgendorf and van der Veer rely in large part on the detailed study of Bhagwati Prasad Sinha, Ram Bhakti men Rasika Sampraday.

33. This theorization of the ascetic is from Khare, The Untouchable as Himself, 25.

34. Buchanan, Account of Shahabad, 1812–1813, 63–64.

35. Shankaracharya, according to his advaita (or nondualist) philosophical vision, posited plural reality to be a product of delusion and argued that cognitive unity could only be achieved through the exploration and perfection of knowledge. Ramanujacharya responded to this strict monism by positing that while final and unequivocal truth is to be found in the divine, the divine expresses itself in the multiplicity of the material world; therefore, to seek to discern the falseness of plural reality is a singularly misguided endeavor. Ramanuja favored bhakti as the best and highest way of reconciling the contradictions of existence, perceiving divinity, and achieving supreme bliss.

36. See McLeod, Evolution of the Sikh Community, 6–7, on the several religious strands in Guru Nanak’s teachings.

37. Buchanan, Bihar and Patna, 1811–1812, 1:67–68. In the published version of this passage, kholesah is spelled khalesah, an error that would have rendered Buchanan’s observation devoid of meaning were it not for his qualifier (“or original Sikhs”) and his use of the term kholesah elsewhere (see, for instance, the description on p. 368).

38. Many Dasnamis sanyasis today contend that Shri Chand was saddened (udas) at being passed over by his father for leadership of the religious community, and hence his followers have since borne the appellation “Udasin,” or full of sorrow (see Ghurye, Indian Sadhus, 141–43). Ghurye views this etymological explanation with skepticism, noting that “even the sectarian Udasins themselves are hard put to it to provide a rational explanation of the term.”

39. Ved Parkash, The Sikhs in Bihar (Patna: Janaki Prakashan, 1981), 152: “Amar excommunicated the Udasins, lest the new Sikh religion should meet the same fate as the other mendicant orders of the country.” Parkash does not elaborate on the nature of that fate.

40. H. A. Rose, “Udasis,” in James Hastings, ed., Encyclopedia of Religion and Ethics (New York: Charles Scribner and Sons, 1921), 504. Rose notes that one of these subsects, Bhagat Bhagwan, claims a large following among Udasins in Eastern India. This statement is corroborated by Ghurye, Indian Sadhus, 145, who, however, notes a comparatively small following in western Bihar. Though Buchanan refers to the prevailing Nanakpanthis in Bihar only as Kholesah Sikhs and Nanakshahis, it seems probable that they belonged to the Bhagat Bhagwan subsect of Udasins.

41. On the nineteenth and twentieth centuries, see Rajiv A. Kapur, Sikh Separatism: The Politics of Faith (London: Allen and Unwin, 1986); and Fox, Lions of the Punjab. Sachchidanand Sharma, Udasi Sampraday aur Kavi Sant Rena [The Udasi Sampraday and the Poet-Saint Rena] (Dehradun: Sahitya Sadan, 1967), 22–23, argues that many Udasins no longer wish to acknowledge their historical connection to Guru Nanak and the Sikhs because the Akali-led Gurudvara reclamation movement of 1921 displaced Udasin control of many Sikh shrines in the Punjab.

42. On this point, see H. H. Wilson’s observations with respect to brahman versus ascetic gurus in his Sketch of the Religious Sects of the Hindus, 30–31.

43. It is entirely possible, however, that tensions within the Ramanandi sampraday over the question of status and attitudes toward caste may have had a role in the distinctions being posited between brahman and nonbrahman Ramanandis. See the following chapter on these tensions, which erupted into a full-blown dispute in 1918 and remain unresolved today.

44. Buchanan, Shahabad, 1812–1813, 219–20. Elsewhere (Bihar and Patna, 1811–1812, 1:374), Buchanan noted that “The term Vaishnav is not considered as disgraceful for a Brahman, as is the case in Bengal and in the south.”

45. Buchanan, Bihar and Patna, 1811–1812, 1:369.

46. Buchanan, “Account of Gorakhpur, 1812,” 139–345.

47. Buchanan, Bihar and Patna, 1811–1812, 1:68. Followers of Vaishnava gurus in Patna city itself were divided almost equally, according to Buchanan, between “Ramawats” (Ramanandis) and Radhaballabhis. However, outside of Patna Radhaballabhis were notable for their relative scarcity. Later, Buchanan noted (1:374) that

48. Buchanan, Purnea, 1809–1810, 274. However, an inkling of the social inequities that would come to divide the Ramanandi sampraday a century later can be discerned here. Buchanan noted that “Part [of the Ramanandis] are descended of Brahmans, have images, and bestow instruction on the followers of Vishnu, who worship that god under the form of Ram. There are also some Ramayits who are Sudras, and serve the others in bringing water and other such occupations, but are not allowed to eat in company” (my emphasis). The phrase “descended of Brahmans” points to the fact that whether or not the individual guru was himself a brahman, his guru genealogy (or guru parampara) could often reveal the individual’s predilections regarding caste commensality. I take up this point in detail in the following chapter.

49. Buchanan, Bihar and Patna, 1811–1812, 1:357.

50. Ibid., 1:352. Buchanan’s observations on the stringent gastronomy of Vaishnavas are repeated in Shahabad, 1812–1813, 220.

51. Buchanan, Bihar and Patna, 1811–1812, 1:376.

52. Ibid., 1:358, 369.

53. Ibid., 1:369. See also Purnea, 1809–1810, 269. Further, a perusal of the caste enumerations in any of the Bihar accounts (e.g., Bihar and Patna, 1811–1812, 1:330–50; Shahabad, 1812–1813, 195–211) will reveal many communities classified as shudra and lower who looked to Dasnamis as spiritual guides in the early nineteenth century.

54. Buchanan, Bihar and Patna, 1811–1812, 1:387–89; see also 358. Parshuram Chaturvedi, an encyclopaedic source on medieval saints and bhakti literature, has noted more recently that Udasins have assumed many of the superficial traits of Hindu sadhus, and have assimilated many standard Hindu customs; see Uttari Bharat ki Sant-Parampara, 425.

55. Indeed, Buchanan mentioned one wealthy Dasnami near Patna who transferred his monastic allegiances to the Nanakpanthi order (Bihar and Patna, 1811–1812, 1:81). Buchanan noted as well that the individual in question was known for his eccentricity and occasional bouts of violent behavior, brought on in part by his disappointment at the loss of followers subsequent to his conversion. “Being a violent man, this disappointment, it is said, has made him outrageous and fearless, and it is alleged that he attacks all traders passing his house with loaded cattle, and partly by importunity, partly by force compels each to give him a trifle, and they do not think it worthwhile to complain.” It should be noted, however, that there were exceptions to the Nanakpanthi-Dasnami affinities. For example, Buchanan mentioned elsewhere (Purnea, 1809–1810, 272) that about seventy akharas of Bengali Vaishnavas (followers of Chaitanya) had been Udasins, or Nanakpanthis.

56. Ghurye, Indian Sadhus, 142–43. “Vaishnava ascetics” for Ghurye meant primarily those sadhus associated with the Ramanandi sampraday, while “Saiva sadhus” meant those attached to Shankaracharya’s Dasnami organization. “Vedanta” refers to the “the end of the Vedas,” signifying the philosophical discourses based on Vedic texts that led to the resurgence of Hindu thought with Shankaracharya and, hence, monist.

57. Sinha and Saraswati, Ascetics of Kashi, 138. Nevertheless, it would also seem that at least some Udasins joined ranks with Ramanandis, a point I return to in chapter 2.

58. Amardas Udasi, Udasi Mat Darpan (no bibliographic information provided), 491–92, cited in Sharma, Udasin Sampraday, 23.

59. Government of India (hereafter GOI), Census of India, 1891, vol. 5: The Lower Provinces of Bengal and Their Feudatories (Calcutta: Bengal Secretariat Press, 1893), table 16, 76; GOI, Census of India, 1901, vol. 6-A: The Lower Provinces of Bengal and Their Feudatories, part 2 (Calcutta: Bengal Secretariat Press, 1902), table 13-A, 194–256. As I note below, however, terminological imprecision with regard to religious identity rendered much of the census data on monasticism of little value.

60. See Richard Eaton, “Approaches to the Study of Conversion to Islam in India,” in R. C. Martin, ed., Approaches to Islam: Religious Studies (Tucson: University of Arizona Press, 1985), 107–23. A more elaborated consideration of the process, termed “inclusion, identification, and displacement,” can be found in Eaton’s more recent work, The Rise of Islam and the Bengal Frontier, 1204–1760 (Berkeley and Los Angeles: University of California Press, 1993), 268–90 and passim.

61. Sinha and Saraswati, Ascetics of Kashi, 49–52, 115.

62. Van der Veer, Gods on Earth, 37 (for the quote) and 143–46. On Shuja ud-daula’s patronage of the sampraday, see Buchanan in Martin, Eastern India, 2:485.

63. Wilson, Sketch of the Religious Sects of the Hindus, 67–68.

64. Herbert Hope Risley, The Tribes and Castes of Bengal (Calcutta: Bengal Secretariat Press, 1891; reprint, Calcutta: Firma Mukhopadhyay, 1981), 1:533–34; Buchanan, Bihar and Patna, 1811–1812, 1:334; Shahabad, 1812–1813, 198. Unfortunately, Risley’s survey was organized strictly according to caste, listed in alphabetical order by jati nomenclature (cf. the methodology employed by Buchanan, described above); Risley paid little direct attention to religious communities that appealed to supracaste loyalties, in part because the existence of such loyalties undermined the supposed impermeability of caste boundaries on which his racial theories were grounded. (On these theories, and the anthropometry and nasal indeces that sustained them, see Risley, The People of India [Calcutta: Thacker, Spink, 1915] and Christopher Pinney, “Colonial Anthropology in the ‘Laboratory of Mankind,’” in C. A. Bayly, ed., An Illustrated History of Modern India, 1600–1947 [London: National Portrait Gallery, 1990], 252–63.) As a result, even though the information under each of Risley’s jati headings contained potentially valuable descriptions of the culture, religion, and mythology of the eastern Gangetic Plain, it is more difficult to elicit religious trends for the population as a whole. In addition, Risley referred not to specific monastic communities, such as Ramanandis and Dasnamis, but to the broad sectarian components of modern Hinduism, namely Vaishnava, Shaiva, and Shakta belief. Nevertheless, if the peasants of Gangetic India found Vaishnava belief and institutions increasingly relevant to their lives, it is certain that this boded well for the Ramanandi sampraday.

65. Jogendra Nath Bhattacharya, Hindu Castes and Sects: An Exposition of the Origin of the Hindu Caste System and the Bearing of the Sects toward Each Other and toward Other Religious Systems (Calcutta: Thacker, Spink, 1896), 29. Bhattacharya’s observation is particularly revealing. First, it implies that many “low” people (no doubt a reference to shudra and untouchable peasants and artisans) had advanced economically in the preceding century, and that they were directing their newfound wealth toward the articulation of a new religious and social ideology. Second, it captures the sensitivity of monks to both prevailing public opinion and hard economic realities. In other words, notwithstanding the fact that Bhattacharya—an elite, educated Bengali pandit—took a dim view of the process (he decried the spread of Vaishnavism as little more than the constant, mindless repetition of Hari, or Vishnu), he saw it as entirely natural and appropriate that sadhus would desire to attract followers, for whatever motives, be they material or spiritual.

66. Buchanan, “Account of Gorakhpur, 1812,” 336. This observation occurs in Haripur town, Nichlaul thana.

67. Of course, the term possessed a strong Vaishnava connotation in Bengal and western Uttar Pradesh. Bengali Vaishnavas were followers of the ecstatic bhakti of Chaitanya; see Sushil Kumar De, Vaishnava Faith and Movement in Bengal (1942; 2d ed., Calcutta: K. L. Mukhopadhyaya, 1961). An account of the Vaishnava gosains who worship Krishna and Radha in and around Mathura and Vrindaban in the nineteenth century can be found in F. S. Growse, Mathura: A District Memoir (1880; 3d ed., revised and enlarged, Allahabad: Northwest Provinces and Oudh Government Press, 1883), 192–98.

68. Buchanan, Bihar and Patna, 1811–1812, 1:374; Shahabad, 1812–1813, 53 (Buchanan spells the name Arah; later British-Indian maps refer to Arrah); Buchanan in Martin, Eastern India, 2:483; Risley, The Tribes and Castes of Bengal, 1:300.

69. For instance, in Bhagwanpur, a large village three miles north of Nawada town in Gaya District, the assistant subdivisional officer observed that “there is . . . an old Thakurbari which is erected by Gosain-ji, which is taken care of by the residents.” Village Notes (hereafter VN), Gaya District, Thana Nawada, no. 400, Gaya Collectorate Record Room, 1909–14, Bihar. Other similar instances of gosains associated with Vaishnava temples are described in VN, Patna District, Thana Barh, no. 254, Patna Collectorate Record Room, 1909–14, Bihar; VN, Gaya District, Thana Mufassil Gaya, no. 35, Gaya Collectorate Record Room. The “Village Note” surveys were carried out to facilitate land settlement operations in Bihar in the early 1900s. Each village note was a four-page form upon which an assistant subdivisional officer recorded his observations and interviews in an ordered sequence of fifteen categories. The social and cultural complexion of a village was captured at the end of every village note in section 15, entitled “General Notes.” For a discussion of some research implications of the village notes, see James R. Hagen and Anand A. Yang, “Local Sources for the Study of Rural India: The ‘Village Notes’ of Bihar,” Indian Economic and Social History Review 13, no. 1 (January-March 1976): 75–84.

70. VN, Gaya District, Thana Aurangabad, no. 934, Gaya Collectorate Record Room. The subdivisional officer added, “There is a pucca Samadhi [burial site for a gosain] near it also built by him. The gosain Bisunpur is a very old and respected man. People go to him and give him money as offering. He has many chelas.” Additional instances of Shaiva gosains can be found in VN, Gaya District, Thana Nawada, no. 445, Gaya Collectorate Record Room; VN, Patna District, Thana Hilsa, no. 78, Patna Collectorate Record Room; VN, Shahabad District, Thana Piro, no. 189, and Thana Mohania, no. 663, Arrah Collectorate Record Room, 1909–14, Bihar.

71. It has been suggested as well that the monastic distinctions evident today among Ramanandis—namely, between nagas, tyagis (or austere ascetics), and rasikas (or aesthetes)—enabled them to absorb all manner of Shaiva ideas and practices and relate them to the worship of Ramchandra; while such structural looseness certainly would have contributed to the increasing strength of the Ramanandi sampraday, I am wary of relying on a modeling of the Ramanandi present to explain the Ramanandi past. It is possible, for example, that the absorption of Shaiva ideas and practices produced, or at least further accentuated, some of the present-day typological distinctions between Ramanandis. On the structure of the present-day Ramanandi sampraday, see van der Veer, Gods on Earth; the general point was raised in a reader’s report on the present work, received in January 1994.

72. Buchanan in Martin, Eastern India, 2:483; Wilson, Sketch of the Religious Sects of the Hindus, 204.

73. See Risley, The Tribes and Castes of Bengal, 1:26; and William Crooke, The Popular Religion and Folklore of Northern India, 2d ed. (London: A. Constable, 1896; reprint, New Delhi: Munshiram Manoharlal, 1978), 1:86.

74. Ghurye, Indian Sadhus, 79.

75. GOI, Census of India, 1891, vol. 5, table XVI, 76; GOI, Census of India, 1901, vol. 6-A, part 2, table XIII-A, 194–256; GOI, Census of India, 1911, vol. 5: Bihar and Orissa, part 3: “Imperial Tables” (Calcutta: Bengal Secretariat Book Depot, 1913), table 13-A, 97–119.

76. GOI, Report on the Census of Bengal, 1872 (Calcutta: Bengal Secretariat Press, 1872), 136.

77. GOI, Census of India, 1911, vol. 5: Bihar and Orissa, part 1: “Report” by E. A. Gait (Calcutta: Bengal Secretariat Book Depot, 1913), 239 (remarks by L. S. S. O’Malley).

78. The subjective nature of the census interview as a methodological failure of previous censuses is raised by E. A. H. Blunt, director of the 1911 census for Uttar Pradesh, in vol. 15: United Provinces of Agra and Oudh, part 1: “Report” (Allahabad: Government Press, 1912), 131.

79. Ibid., subsidiary table 8-A, 156–57. Even in the eastern Uttar Pradesh districts, Vaishnavas had the edge on Shaivas and far outnumbered Shaktas.

80. At the risk of jumping ahead to the last decade of this century, it can be suggested that the successful (if short-lived) recruitment of many—but by no means all—sadhus into the political campaigns of the essentially middle-class Hindu right (particularly as spearheaded by the Vishwa Hindu Parishad and organized around the reclamation of the birthplace of Ramchandra) is an attempt by some sadhus to reoccupy a position of political and social centrality in modern Indian life.