Page 2 :
BANDAGI AND NAUKARĪ 61, Timurid invasion (1398/99) and the fragmentation of the Tughluqid, dominion.' The inordinate persistence of this representation is surpris-, ing, not least of all because it is in the face of some recent scholarship, on the fifteenth and sixteenth centuries, not least of all by Dirk Kolff, and Simon Digby.2 Dirk Kolff's observations on the culture of service,, naukari, the search for military employment compensated by a salary, and/or other social and political rewards, was placed as a distinguishing, feature of the fifteenth and sixteenth centuries, setting this period apart, from the one prior to Timur's invasion. In Kolffs analysis, neither the, decline of the Sultanate nor Timur were primary contributors to the, making of this new political culture. Instead naukari was closely linked, to the military labour market in the subcontinent where the opportu-, nities for, and availability of, armed personnel had grown exponentially, through the fifteenth into the sixteenth centuries. It was this world,, rather than that of the Delhi Sultans, that had an abiding impact on, the Mughals., 3., BANDAGĪAND NAUKARĪ, STUDYING TRANSITIONS IN POLITICAL, CULTURE AND SERVICE UNDER THE, NORTH INDIAN SULTANATES,, THIRTEENTH-SIXTEENTH CENTURIES*, Sunil Kumar, ' Note Peter Jackson, The Delhi Sultanate: A Political and Military History,, Cambridge: Cambridge University Press, 1999, whose penultimate chapter,, "Stasis and Decline: Firuz Shah and His Successors', is followed by 'Epilogue:, c. 1400-1526'., 2 Dirk Kolff, Naukar, Rajput and Sepoy: The Ethnohistory of the Military Labour, Market in Hindustan, 1450–1850, Cambridge: Cambridge University Press,, 1990; Simon Digby, War Horse and Elephant in the Delhi Sultanate: A Study of, Military Supplies, Karachi: Orient Monographs, 1971; Digby, 'Anecdotes of a, Provincial Sufi of the Delhi Sultanate: Khwäja Gurg of Kara', Iran, 32, 1994:, 99-109; Digby, 'Before Timur Came: Provincialization of the Delhi Sultanate, through the Fourteenth Century', Journal of the Economic and Social History of, the Orient, 47(3), 2004: 298-356; Digby, 'Two Captains of the Jawnpur Sultan-, ate', in Circumambulations in South Asian History: Essays in Honour of Dirk H.A., Kolff, Jos Gommans and Om Prakash (eds), Brill: Leiden, 2003, pp. 159-78;, Digby, "The Indo-Persian Historiography of the Lodi Sultans', in Les Sources et le, temps, F. Grimal (ed.), Pondichéry: lÉcole française d'Extrême-Orient, 2001, pp., 243–61; Digby, "Abd al-Quddus Gangohi (1456-1537 A.D.): The Personality, and Attitudes of a Medieval Indian Sufi', Medieval India: A Miscellany, Bombay:, Asia Publishing House, 1975, vol. 3, pp. 1-66; Digby, 'Dreams and Reminis-, cences of Dattu Sarvani, a Sixteenth Century Indo-Afghan Soldier', Indian Eco-, nomic and Social Economic and Social History Review, 2(1), 1965: 52-80, and, 2(2): 178-94. Digby's contributions are considered slightly later in the chapter., HISTORIES OF THE DELHI SULTANATE are usually organized into rather, simple binaries: years of centralized governance of the Sultans of Delhi, are measured against years of decentralized rule under the 'regional, Sultanates. Historians frequently arrange their narratives to describe the, origin-apogee-decline career graph of the Sultanate where 'stasis and, decline' is usually the chapter before the epilogue on the years after the, * This chapter has profited from the critical interventions of Samira Sheikh,, Francesca Orsini, Munis Faruqi, Tanika Sarkar, William Pinch, Pankaj Jha, and, Ali Anooshahr. I am grateful to Francoise 'Nalini' Delvoye for the opportunity, to first think about the subject in my lectures at the EPHE in 2006. Anjali, Kumar, as always, was a part of its writing. This would have been a lesser work, without Sikandar Kumar's perspicacious engagements with its arguments. I am, grateful to Ali Anooshahr and Pankaj Jha for making the texts of Mushtaqi and, Vidyapati available to me.

Page 3 :
79, AFTER TIMUR LEFT, BANDAGĪ AND NAUKARI 63, Despite Kolff's innovative rereading of the post-Tughluq and pre-, Mughal centuries (1414-1556), his work also begged the question: what, was the prehistory of this culture? From what elements was it consti-, tuted? What range of personnel and ties did political service encompass, prior to the fifteenth century? How did this alter and in what arenas?, In other words, even as Kolff elaborated on the quality of naukari, the, history of this culture remained murky. Notably, in his analysis it was, never clear why the Khaljis or the Tughluqs did not tap into the mili-, tary labour market in the same way as the later Afghans or the Jaunpur,, Malwa, and Gujarat regimes. Were the differences merely conjunctural, contingent, or were they a consequence of larger structural changes, Although Kolff's understanding of naukari was further inflected by the, notion of ethnogenesis, where groups developed their identities through, shared service, domicile, language, literature, and memories, these pro-, cesses were, in themselves, not unique to the cultural world of the fifteenth, century warband.' Many of the similarities and differences between the, period before and after Timur's invasion have not been systematically, explored and the vectors of analysis are still based on impressionistic evi-, dence concerning scale of military and political formations. Perhaps most, critically, there has been no comparison of the ways in which regimes, and their participants were discussed in the literary materials of the time,, or of the ways in which they constructed ideas of service, loyalty, and, leadership and their material dimensions-what aspects continued, what, changed, and how are they manifest through the thirteenth into the, fifteenth-sixteenth centuries?, pune, in representation and interpersonal relationships through the fourteenth, century, or a mixture of all these factors?, In as much as conventional historiography considers the fifteenth and, sixteenth centuries as substantive departures from a preceding period,, this chapter extends the provocative question posed by the editors, in the introduction to this volume and asks, 'How well do we know, the history of the thirteenth and fourteenth centuries that we should, consider the fifteenth and sixteenth centuries so different?' There are, considerable similarities in both periods: the armed peasantry were also, recruited in huge numbers by the Delhi Sultans (especially Ala' al-Din, Khalaji and Muhammad Shah Tughluq), and immigrant person-, nel, occasionally of very humble origins (including the Afghans) were, integral to Sultanate armies in the thirteenth and fourteenth centuries,, some going on to become governors of provinces and rulers of Delhi., This is actually quite astonishing because most historians have com-, mented on shifts in the nature of the literary materials available to the, historians from the fifteenth century. As the editors of the volume have, noted, the epoch of the great histories of Delhi, the tawarikh of Juzjani,, Barani, 'Isami had tapered to an end, there were few contemporaneous, Persian histories on the fifteenth and sixteenth centuries and far more, that were retrospective accounts produced in the Mughal chanceries or, in the courts of latter-day Rajput princes. There were also histories pro-, duced in a variety of vernacular languages following literary conventions, borrowed from Sanskrit and an assortment of Bhakti and Sufi-inflected, styles. But the sheer difference in the literary materials, the cognitive, worlds that they mapped, and their retrospective projections of social, and material conditions into an earlier period left historians trained in, reading the grand narratives of the Persian tawarikh of the Delhi Sultans, 3 For details on the huge standing army of the two monarchs see Peter, Jackson, The Problems of a Vast Military Encampment', in Delhi Through, the Ages, R.E. Frykenberg (ed.), Delhi: Oxford University Press, 1986, pp., 18-33, and more tangentially Irfan Habib, 'The Price Regulations of 'Ala'uddin, Khalji-A Defence of Zia' Barani', Indian Economic and Social History Review,, 21(4), 1984: 393-414., 4 All of the founding dynasts of the Khalaji, Tughluq, Sayyid, and Lodi, regimes were immigrants to the subcontinent and originally frontier commanders., For a more detailed response of the Persian literati to these groups of people,, see Sunil Kumar, 'The Ignored Elites: Turks, Mongols and a Persian Secretarial, Class in the early Delhi Sultanate', Modern Asian Studies, 43(1), 2009: 45-77,, and Kumar, 'Courts, Capitals and Kingship: Delhi and its Sultans in the Thirteenth, and Fourteenth Centuries CE', in Court Cultures in the Muslim World: Seventh, to Nineteenth Centuries, Albrecht Fuess and Jan Peter Hartung (eds), London:, SOAS/Routledge Studies on the Middle East, 2011, pp. 123-48., 5 Sec, for example, the early work of B.D. Chattopadhyaya, 'Origin of, the Rajputs: the Political, Economic and Social Processes in Early Medieval, Rajasthan', Indian Historical Review, 3(1), 1976: 59-82, oddly missing in Kolff's, bibliography. And from the adjoining region of Transoxiana and Afghanistan,, but considering historical processes that have a bearing on South Asia, see, Jean Aubin, 'L'ethnogénèse des Qaraunas', Turcica, 1, 1969: 65-94.

Page 4 :
BANDAGI AND NAUKARĪ 65, 64 AFTER TIMUR LEFT, or the Mughal period with no ready-made templates that they could, readily follow., Since these literary materials were very different from each other, in their readings of history, their literary styles, and authorial intent,, their interpretation needed, as Carlo Ginzburg reminded historians in, a different context, a great deal of caution. As Ginzburg noted, The, fashionable injunction to st..y reality as a text should be supplemented, by the awareness that no text can be understood without a reference, to extra-textual realities'.7 In other words, from the perspective of my, bandagi is derived from the noun banda, the Persian term for military, slave (plural: bandagān), and describes political relationships that were, strongly touched by traditions of servitude present in the practice of, military slavery. I use the term to discuss political relationships amongst, personnel who were not always, strictly speaking, 'unfree'. While these, personnel could have been free, their relationships were also strongly, touched with the paradigms of service associated with bandagi. On the, other hand, the adjective naukari as a means to understand political, culture may appear as more of a neologism. In its Persian usage the, noun naukar carried some of the meanings present in its antecedent, Mongol form, nökör (singular)/nököd (plural)-personal retainer, loyal, friend, comrade in arms, bodyguard-and within the limited context, of a dyadic relationship with a master, its meaning was very close to, banda-i khāşş. In its original Mongol sense, the nököd were free and, honourable servants, who had voluntarily accepted service with a great, lord; it had none of the pejorative meanings associated with slavery., In the pre-Timurid years of the Sultanate I have found no evidence in, argument, since the literary materials before and after Timur's invasions, are so different, and the evidence they proffer so incommensurate, any, comparative exercise of the cultural and political worlds of the two, periods has to have some degree of structural correspondence. I keep, that in mind as I try to disaggregate the elements that constituted politi-, cal culture and service at differing moments of time and investigate three, relatively precise subjects: the recruitment of changing kinds of person-, nel and what this might tell us regarding expectations and ambitions, of masters and servants; the ways in which political entitlement related, to the social and cultural profiles of individuals and groups and might, serve to alter it; and the manner in which Persian and other writers, framed and commented on departures from idealized norms of political, conduct. Because of the varied modes of literary presentation, I have, tried to be as precise as possible in presenting the contexts in which dif-, ferent personnel constructed a range of service relationships through the, long period of the thirteenth-sixteenth centuries, paying close attention, to the ways in which the literary materials commented on this subject, 8 While the term bandagi has not been used by historians (as yet) to unravel, aspects of pre-modern political culture, the term itself is not a neologism. See,, for example, Ziya al-Din Barani, Fatăwä-yi jahandäri, A. Salim Khan (ed.),, Lahore: Idarah-i Tahqiqat-i Pakistan wa Intishgah-i Punjab, 1972, p. 333. The, term seems to have had considerable currency as a title in the fifteenth and, sixteenth centuries. For a notable example, consider the title of Khizr Khan, (r. 1414-21), the founder of the Sayyid dynasty. See Yahya Sirhindi, Tarikh-i, Mubārak Shähi, M. Hidayat Hosain (ed.), Calcutta: Bibliotheca Indica, 1931,, p. 181 and the title of the monarch 'Bandagi rāyāt-i a'la' Khizr Khân. This, would translate literally as 'In the Service of the Exalted Banners'. Sirhindi's, anecdotes on the monarch's Sayyid status would suggest a reference to 'service to, the banners of the Prophet'. But there is enough ambiguity to allow readers to, consider a possible genuflection towards Timur: see Sirhindi (1931), p. 182, and, the title of the monarch stated as bandagi'i bandagân räyät-i a'la' and the conclu-, sion to this chapter for a further discussion of post-Sultanate usages of bandagi., 9 For a useful discussion of the Mongol nökör, see V. Vladimirtsov, Le Regime, social des Mongols, M. Carsow (trans.), Paris: A. Maisonneuve, 1948, pp. 110-30;, Igor de Rachewiltz, The Secret History of the Mongols: A Mongolian Epic Chron-, icle of the Thirteenth Century, Leiden: E.J. Brill, 2004, vol. 1, pp. 256-7; and, Gerhard Doerfer, Türkische und Mongolische Elemente im Neupersischen,, Weisbaden: Franz Steiner Verlag, 1963, vol. 1, pp. 521-6., over time., In my analysis of political culture and service, I use the two terms, bandagi and naukari hermeneutically to analyse the relationships of the, different kinds of personnel recruited by the Delhi Sultans and other, patrons between the thirteenth and sixteenth centuries. The adjective, * For an important overview on the sources of this period and the historio-, graphical issues they raise for the historian, see Francesca Orsini, 'How to Do, Multilingual Literary History? Lessons from Fifteenth and Sixteenth Century, North India', Indian Economic and Social History Review, 49(2), 2012: 226-46., Carlo Ginzburg, 'Checking the Evidence: The Judge and the Historian',, Critical Inquiry, 18(1), 1991: 84,

Page 5 :
66 AFTER TIMUR LEFT, BANDAGI AND NAUKARĪ 67, the Persian chronicles of the usage of the term naukar; it was, however,, in circulation in the Afghan records of the sixteenth century, but not, as, far as I know, in its adjectival form, naukari. Even if the term naukari, was absent in the pre-Timurid centuries, I am interested in researching, the elements that constituted relationships with naukars (or naukari, as I, refer to it) and here, happily, there is considerable information., I place naukari in a dialectical relationship with bandagi, with whom, it can be usefully juxtaposed for purposes of elucidating the different, kinds of political relationships and the tensions apparent in their rep-, resentations. In this context I found the diachronic mode particularly, useful in avoiding what might appear as a bandagi/naukari binary., While the elite personnel encompassed in these terms were always in, flux and their relationships and identities shaped by a variety of material, factors, they were not the authors of their histories. Moving through, time has the advantage of careful contextualization to enable an inflected, analysis of how Persian and other chroniclers responded to Sultanate, personnel with shared or contrasting backgrounds and conditions of, service. It allows me the opportunity to note continuities and transitions, by paying attention to the ways in which particular characteristics of these, personnel and their contexts were elided or foregrounded. Cohering, political relationships around terms such as bandagi and naukari, there-, fore, serves as points of entry through which contingent relationships, and their representations can be further interrogated to finesse particu-, larities, especially over time. Towards that end, the following section studies, the meaning of bandagi from the thirteenth to the sixteenth century in, somewhat greater detail., Malik Kafur Hazar Dinari, the Khalji general who led 'Ala' al-Din's, armies into the Deccan and managed court politics in Delhi at the end, of his master's reign;10 Khusraw Khan Barwari (r. 1320), the Khalji, slave commander who deposed his master, became Sultan of Delhi, and, had cordial relations with the Sufi Shaykh Nizam al-Din Awliya;" the, Tughluq slave 'Imad al-Mulk Bashir Sultani who was Firuz Tughluq's, commander of armies;12 and Malik Sarwar Khwaja Jahan, a eunuch and, Firuz Tughluq's commander, who went on to become Sultan of Jaunpur, (1394-9) and founded the Sharqi dynasty (1394-1457).13 We also, need to take into account the cadre of military personnel present dur-, ing the period of Lodi (1451-1526) and Sur rule (1540-55), contingents, that were described as khassa khayl. These contingents were retained by, the Sultans and his nobles, amidst which were a significant number of, slaves., It is not the persistence of slave retinues through this period that, is actually of such great significance, but its attendant implication:, through these centuries there was also considerable structural stability, in the modes of selecting and socializing personnel into a culture of, political service which I refer to as bandagi. This was evident in the, ways in which common siaves were transformed into bandagan-i khass., 10 On Malik Kafur Hazar Dinari, see the synopsis of Peter Jackson (1999),, pp. 174-7, 206-7., 11 Jackson (1999), pp. 158, 177., 12 Shams-i Siraj Afif, Ta'rikh-i Firüz Shähi, Maulavi Viłayat Husain (ed.),, Calcutta: Bibliotheca Indica, 1888-91, pp. 436-45., 13 For his appointment during the reign of the Tughluq monarch Sultan, Mahmud Shah II (1393-5), see Sirhindi (1931), pp. 156-7., 14 For an account of the khassa khayl, see I.H. Siddiqui, Some Aspects of, Afghan Despotism in India, Aligarh: Three Men Publication, 1969, pp. 111-17, and Siddiqui, The Army of the Afghan Kings in India', Islamic Culture, 39,, 1965: 223-7. The khassa khayl contingents seem to have been very similar to, the thirteenth- and the fourteenth-century qalb retinue-the central contingent, of the army that surrounded the ruler at moments of conflict and stayed with, her/him in the capital. The qalb contained a significant number of the ruler's, bandagan as well; see Sunil Kumar, Emergence of the Delhi Sultanate, Delhi:, Permanent Black, 2007, pp. 78, 132, 258-9, 321. For the recruitment of slaves, and their sons by Sher Shah, see also Kolff (1990), pp. 66-7., THE VALUE OF DÉRACINÉS: BANDAGI IN, THE PRE-MUGHAL PERIOD, For the longest time historians suggested that the usage of military slaves, was a condition unique to the thirteenth century Turkish' Sultanate, regime. The evidence contradicts this conclusion; the deployment of, military slaves continued well into the Lodi regime in the sixteenth cen-, tury and beyond. Although historians find it easy to recall Qutb al-Din, Ai-Beg, Iltutmish, and Balban as examples of important, pue, powerful, slaves in the thirteenth century, we need to add to this list of significant, examples of military slaves in the service of Delhi Sultans the eunuch