Introducing the Sultanate Frontier Military Commanders

Social and political formations in the Indus and Gangetic plains were not unduly troubled by political developments in the mountainous Hindu Kush or Karakorum regions in the north-west. These lands were too poor and fragmented to support large state systems and the pastoral inhabitants of the area indulged in relatively localized plundering expeditions into the plains. Although trade routes into Iran and Central Asia were more easily disturbed by the turbulent politics of the region, Afghanistan seized the attention of political regimes in north India only when the area became a part of larger geopolitical developments.

In the tenth through the twelfth centuries this happened when the Ghaznavid and Ghurid regimes attempted to sustain their control over eastern Iran by the revenues extracted from north India. The challenges posed by these developments were completely dwarfed by the Chinggisid invasions of the thirteenth centuries. The Mongols seized much of western Punjab and periodically threatened the Gangetic plains, destroying agriculture, displacing pastoralists and pillaging cities. Beyond the very real threat of Mongol depredations was the ‘great fear’ that gripped the land in the 1220s and after, when it seemed as if a holocaust of proportions already witnessed in eastern Iran, Transoxiana, and Afghanistan was awaiting north India.Footnote ¹

The need to secure the Sultanate regime from Mongol marauders led to the delineation of a ‘frontier’ that needed to be defended. At least during the early-thirteenth century this was carried out through garrisoned cantonments in the Punjab. These cantonments were placed under trusted slave-commanders, the bandagan-i khass, of the monarch. They were of Turkish origin, but the patronage of their master, together with systematic efforts to bond and incorporate them in the household of the monarch, oriented their allegiance away from their ethnic roots and towards the realm of Delhi. Judging by their military records Turkish slaves did not hesitate in opposing the invading Mongol hordes, which carried in their train a large number of Turks. In other words, shared ethnicities notwithstanding, Sultanate commanders on the frontier were acculturated to serve a regime that oriented them in ways quite distinct from their original steppe habitats.Footnote ² The frontier between the Mongols and the Sultanate was therefore marked by different cultural and political orientations, even if the social groups who inhabited the region were not always dissimilar.

This paper is concerned with developments of a slightly later period—the decades after the 1250s—when political fragmentation within the Delhi Sultanate and the Mongol confederacies complicated relationships across the Punjab frontier. In the years after Shams al-Din Iltutmish's death (1236) increasing competition amongst Sultanate slave commanders drove discontented amirs into alliances with the Mongols.Footnote ³ Concurrently, the old concordance amongst the Chinggisid descendants was ending and the lands of eastern Iran, Afghanistan and Transoxiana were populated by rival political dispensations. Internal conflict and the search for alternative opportunities pushed many Mongol commanders and their subordinates into Hindustan and the service of the Delhi Sultans.Footnote ⁴ Although Mongol raids into north India continued through the second-half of the thirteenth century, there was considerable migration of Mongol and Turkic groups searching for Sultanate patronage and instances of disaffected Sultanate amirs looking for allies in Mongol camps.

In focusing upon the second-half of the thirteenth and the early-fourteenth century, the first part of my paper draws attention to the recruitment of frontiersmen by successive Sultanate regimes to guard the Punjab marches from Mongol depredations. The old traditions of policing the frontier by slave commanders slowly shifted to include new bodies of immigrants who had intruded into the region. These developments were first noticeable in the reign of Balban (1266–87) and Kaiqubad (1287–90) and then more apparent during the succeeding Khalaji regime (1290–1320). Political fortunes in the marches of the Punjab, areas that lay in the interstices between the Mongol and Sultanate dominions, fluctuated constantly and military successes (and failures) were often transient. Service here, remote, as it was, from the cohering and disciplinary structures of Delhi, also allowed for great opportunities. Ambitious Sultanate commanders adroitly used frontier manpower resources to accumulate large war bands and construct local reputations as warriors and patrons even as they remained marginal, distant groups in the courtly intrigues of the capital. And yet, when the opportunity presented itself, these frontier commanders possessed sufficient assets and initiative to march into Delhi, seize power and establish their own dynasties. Nor were these exceptional moments in the history of the Sultanate. Although in this paper I study the Khalajis peripherally and give greater attention to the early Tughluq regime, it is important to note that every Sultanate dynasty from 1290 through 1526—the Khalaji, Tughluq, Sayyid and the Lodis—had frontier origins.

The following sections of my paper unravel the social and cultural backgrounds of the frontier commanders and study the ways in which these might have complicated their relationship with Delhi. During the Khalaji regime many of the frontier commanders and their contingents were of Turkish or Mongol background and had a record of past service with Chinggisid subordinates active in the Afghanistan region. Hence the curious paradox in the deployment of these commanders: many of the frontiersmen patronized by the Delhi Sultans shared a history of past service and cultural affinities with the very people who periodically threatened the Sultanate.Footnote ⁵ Although their loyalties and investments in the cause of the Delhi Sultans must have been adequately ascertained to justify their deployment, these frontier commanders had not undergone the processes of training and acculturation characteristic of the bandagan-i khass. They were not a deracinated group but had arrived in the Sultanate with intact lineage networks and were linked to significant parts of their retinues by shared natal, ethnic and/or past service associations. Although they had accepted service with the Delhi Sultans and went on to become monarchs themselves, we are indifferently informed about the extent to which these military commanders and their retinues had made the transition from their old steppe-descended, frontier milieux to the urbane world of Delhi. Certainly, prior to their arrival as Sultans, Delhi's literati had looked askance at people of similar social and cultural profiles. What was their reaction when groups of frontiersmen arrived in the capital as lords and masters?

As I argue in this paper, not only is the evidence on this subject extremely scanty, it is also deliberately evasive. The discourse of the fourteenth-century Persian historical narratives (tawarikh) carried their author's vision of an ideal public order tempered by their class, cultural and ethnic prejudices. This was transcribed into an idealized history of the court of Sultan Mahmud Ghaznavi and held as a touchstone of good governance to be followed by future generations. These social and intellectual precommitments meant that the lords and masters of Delhi could not be reported as frontiersmen or ex-servants of the loathed Chinggisids. To have done so would have conveyed the sense of a Sultanate in crisis and decline.Footnote ⁶ To communicate the sense of stability and order, on the other hand, frontier commanders like Ghiyas al-Din Tughluq had to be creatively reinvented as paradigms of virtue, a veritable ‘saviour of Islam’.

Although the narratives of Sultanate histories were selective in their inclusion of information, there were other more episodic records that provided incidental information on the backgrounds of the Delhi Sultans. I use these not just to detail the frontier origins of the monarchs of Delhi, or study the impact of their arrival as Sultans on a world that had only recently treated them as ‘rustics’. As I argue in the concluding sections of the paper, it is crucial for historians today to draw attention to the ways in which the chronicles and eulogies of the Delhi Sultans ignored crucial aspects about their protagonists, how their silences and elisions (and sometimes their ignorance) misrepresented the character of their patrons. It is absolutely vital that we foreground the discursive intervention of the Persian chronicles in their representations of Sultanate history because without a sensitivity to their objectives and their prejudices we will never be able to disengage and texture their representations of a monolithic Islam, a hegemonic state and a timeless Persianate culture. Retrieving the history of the frontier military commanders allows us to recall the role of a vital, if marginalised, group of people involved in the framing of Sultanate history. That these marginalised groups happened to be the political elites and Sultans of Delhi is in itself a telling commentary on the state of the evidence and the dire need to renarrativise the history of the Sultanate. To that end my paper starts with an analysis of the reportage on the Khalajis and the Tughluqs during their deployment as frontier commanders.

The Origin of the Khalajis and Tughluqs as Frontier Commanders

Trying to follow the early history of the founders of the Khalaji and Tughluq regimes is easier said than done. Despite the importance of Jalal al-Din Khalaji (1290–96) and Ghiyas al-Din Tughluq (1320–24) in the history of the Sultanate, Persian chronicles are remarkably silent about their frontier backgrounds. It is from a fifteenth-century chronicle that we know the name of Jalal al-Din Khalaji's father, Yughrush.Footnote ⁷ We have to turn to Il-Khanid chronicles to learn that Jalal al-Din was the Mongol commander (shahna) of Binban, just west of the Indus.Footnote ⁸ Amir Khusrau quotes him on his exploits against refractory Mongol and Afghan tribes in the Salt range. He does not provide any context to these events which, if not hyperbole, might have occurred before he joined service with the Delhi Sultans.Footnote ⁹ An incidental reference in Juzjani informs us that the son of Yughrush (Jalal al-Din?) visited Delhi with a Mongol embassy in 1260.Footnote ¹⁰ It is not clear when Jalal al-Din started serving the Delhi Sultans but it must have been a few years later, sometime during Balban's reign. ‘Isami mentions that he was in the service of Balban's younger son, Prince Bughra Khan.Footnote ¹¹ The account is chronologically unclear but this must have occurred before 1280 while the Prince was still located in Samana, a town which would later become Jalal al-Din's headquarters. Jalal al-Din's political influence increased as commander of the north-west marches until Sultan Kaiqubad invited him to the capital as a counterweight to the old Balbani elites and the new parvenu commanders entrenched in Delhi. Although given the exalted title of Shaista Khan and the military assignment of Baran, Jalal al-Din found it difficult to integrate himself in the politics of the court especially with the murder of the young Sultan. Faced with political marginalization he acted against the clique who controlled the capital and seized the throne in 1290. Even after his accession, insurrection led by members of the old regime continued and it was not until 1292 that his reign approximated some degree of stability.Footnote ¹²

As obscure as Jalal al-Din's early history is the past of Ghiyas al-Din Tughluq, the other military commander on the north-west frontier who went on to become Sultan. Although Amir Khusrau provided a eulogy of Ghiyas al-Din's campaign against the usurper Khusrau Khan Barwari (1320), we are informed by the Moroccan traveller, Ibn Battuta, and not a chronicler of the Delhi court, that Ghiyas al-Din was a Qara'una Turk.Footnote ¹³ Aubin had clarified years ago that the Qara'una epithet was used to describe the followers of the Mongol commander Negüder who belonged to the Jochid-Golden Horde dispensation. Negüder and his followers were marooned in the Afghanistan-Khurasan region as the territories under the sons of Chinggis Khan hardened into antagonistic regimes.Footnote ¹⁴ In the 1260s the Qara'unas found themselves sandwiched between the Il-Khanid and the Chaghatayid realms on the one hand and the territory of the Delhi Sultans on the other. They were eventually scattered amongst the other Mongol groups but their identities were not completely erased. In his description of the areas near Ghazni in the early-sixteenth century, the Mughal Emperor, Babur, noted, ‘in the mountains of Ghazni are Hazaras and Negüderis, amongst some of whom Mongolian is spoken’.Footnote ¹⁵

Ghiyas al-Din was not a great Negüderid amir when he migrated to India. According to Ibn Battuta he worked for a merchant as a humble keeper of the horses, perhaps a cattle driver (gulwaniya > guala), and received patronage first from Ulugh Khan, the brother of Sultan ‘Ala al-Din Khalaji.Footnote ¹⁶ Amir Khusrau was less explicit. He recalled Ghiyas al-Din's statement about his early years as a nomad (awara mardi) when the patronage received from Sultan Jalal al-Din Khalaji (not Ulugh Khan in this version) raised him to high status.Footnote ¹⁷ The discrepancies amongst his early patrons notwithstanding, there is no dispute about how Ghiyas al-Din's military activities on the frontier improved his fortune until by the second decade of the fourteenth century he was commander of Dipalpur, with some respect as a successful general against the Mongols.Footnote ¹⁸ His frontier background and his successes did not endear him with the Khalaji military commanders and Amir Khusrau details their refusal to join him in his effort to remove Sultan Khusrau Khan Barwari from power.Footnote ¹⁹ Like Jalal al-Din Khalaji, Ghiyas al-Din was also outside the charmed circle of military confreres in the capital and, as an outsider, not regarded worthy enough to be a potential candidate to the throne of Delhi.

The progenitors of both, the Khalaji and the Tughluqid regimes shared many features. Both had served the Mongols in the early part of their careers and both had risen to power as ‘wardens of the marches’. In trading their old Mongol associations for opportunities present in the service of the Delhi regime, the two protagonists were representative of the larger social and political milieu on the north-west frontier of the Sultanate that we have already described. Unlike many other migrant Mongol commanders who had made their way to Delhi, both Jalal al-Din and Ghiyas al-Din stayed on the frontier. They had greater success here than their compatriots in Delhi many of whom were indicted in conspiracies and purged in ‘Ala’ al-Din Khalaji's reign. And yet, when they tried to capitalize on their achievements on the frontier and seize power in the capital, they were opposed by the military elite. Their indifferent networks of political support in Delhi certainly weakened their cause in the capital. But as I will show in the following sections, the arrival of frontiersmen in the capital was also the cause of considerable unease, an awkwardness that the seizure of political authority was not quite able to erase from the narratives of the court chroniclers.

Frontiersmen, the Tughluqs and Delhi's Persian Literati

Although all the records relating to Jalal al-Din Khalaji and Ghiyas al-Din Tughluq underline the distance of the two frontier commanders from the Delhi-based elites, the narratives of the Persian chronicles remained quite neutral, if not eulogistic, about their personal qualities. Their early histories were quite ignored and, as we noticed, only Ibn Battuta recalled Ghiyas al-Din Tughluq's origins as a keeper of horses, perhaps a cattle driver (gulwaniya). But the future Sultans also brought to Delhi their contingents and associates recruited from the frontier tracts. And their qualities were sometimes expatiated without as much reservation. To place some of these sentiments in context it might be useful to reflect upon some of the early instances that we have of Sultanate recruitment of frontier groups.

In 1260, 6 years before his accession, the future Sultan Balban had deployed Afghans in areas just south of the capital. These were a new group of people not mentioned as a part of any Sultanate military contingent prior to this date. The impact of these warriors on the Delhi literati was evident in Juzjani's awe-struck description of their fearful, strange presence:

In marked distinction to Turkish slaves brought to the Sultanate from the Eurasian steppes, whose fighting qualities and abilities to adapt in their host societies were applauded by early-thirteenth century chroniclers, Juzjani underlined how the Afghans were a strange, unfamiliar body of soldiers.Footnote ²¹ Although Balban went on to deploy the Afghans in armed camps around Delhi, they remained socially and culturally distanced from the world of the urbane literati. Sometime around 1280, a decade before Sultan Jalal al-Din Khalaji's arrival in Delhi, Amir Khusrau penned a letter to Ikhtiyar al-Din Begtars complaining about his plight at being forced to reside near Afghans. Amir Khusrau noted:

Amir Khusrau was more ambiguous when it came to a description of Ghiyas al-Din Tughluq's armed forces as they marched to the capital. In the Tughluq Nama, Amir Khusrau's eulogy to the monarch, the author noted ‘his troopers were mainly from the upper lands (iqlim-i bala, a euphemism for Khurasan and Transoxiana) and not Hindustanis or local chieftains. They included Ghuzz, Turks and Mongols of Rum and Rus and some Khurasani Persians (tazik) of pure stock (pak asl)’.Footnote ²³ To this motley crowd, ‘Isami detailed the presence of Khokars, a body of frontier pastoralists, forever in conflict with Sultanate armies and at least one Afghan commander.Footnote ²⁴

The Turks and Mongols mentioned by Amir Khusrau are of some interest. If we take the latter first, the reference to Mongols of Rum and Rus referred to the Mongols who occupied the Eurasian steppes in pre-modern Russia, the dasht-i Qipchaq, at this time under the hegemony of the Golden Horde. The Negüderids/Qara'unas (of which Ghiyas al-Din Tughluq was a part) belonged to this larger body of people. Amir Khusrau was actually quite familiar with them; he was briefly captured by a Negüderid in 1285. In his elegy written at the death of Balban's son, Amir Khusrau had described his captor as a Qara'una, a coarse detestable being:

Harsh sentiments, perhaps, and brought about by the experience of captivity no doubt; Amir Khusrau certainly never repeated these sentiments in the Tughluq Nama dedicated to his Qara'una patron.

Amir Khusrau had mentioned Turks as well. The first were the Ghuzz, Turkoman nomads, present in the Afghanistan region. These were fragmented, pastoral groups who nomadized in the Khurasan, Transoxiana, Afghanistan belt and had a long history of conflict with regimes as diverse as the Seljuqs, the Ghaznavids and the Ghurids. Sultan Mu'izz al-Din Ghuri had managed to secure Ghazni only after clearing the Ghuzz out of the region in the 1170s.Footnote ²⁶ So far as I know there is no prior reference to their movements in the subcontinent beyond the Punjab.

In his other reference to Turks in Ghiyas al-Din's retinue, Amir Khusrau conflated them with the Mongols of Rum and Rus presumably because they belonged to the same dasht-i Qipchaq region. This would be the large, loosely organized confederacy of the Qipchaq Turks lately feeling the impact of Chinggisid invasions. Many Qipchaq tribes came under the dominance of the Jochids and were absorbed in their military retinues. Some of them had already made their way into Hungary where they were known as the Cumans.Footnote ²⁷ Others were enslaved and sold in Egypt and India. In Egypt they formed the ruling elite of the Mamluk (Bahri) Sultanate and in India they were the dominant Turkish group in the political dispensation of Sultan Iltutmish. Their presence in the Kabul region into the sixteenth century—or certainly the memory of their presence—is suggested by Babur who mentioned a Qipchaq road, a Qipchaq pass on the Andarab River in the vicinity of Kabul, even a Qipchaq gate in the city of Herat.Footnote ²⁸ Like the Ghuzz, the Qipchaqs were a fragmented body of people seeking sanctuary from Mongol incursions in enclaves of sanctuary between large state formations. They congealed into war bands and, when not predators themselves, were an accessible body of recruits for military commanders like Ghiyas al-Din Tughluq on the north-west marches of India.

How did Delhi's Persian literati respond to the arrival of these people in Delhi? The level of discretion displayed by the Persian chroniclers in their accounts of frontiersmen-turned-Sultans and their retinues does not mean that they knew very little about the Turk and Mongol tribes or their dispersal in Central Asia. Indeed, they often used the ethnic term ‘Turk’ in a very generic sense to signify a military slave or a military commander.Footnote ²⁹ And yet the occasional usage of very precise markers clarified that Persian chroniclers possessed a relatively clear knowledge of Turko-Mongol ethnicities, tribal affiliations and dispersal in the lands adjacent to the Sultanate.

In his Tabaqat-i Nasiri (completed ca 1260) Minhaj-i Siraj Juzjani provided the provenance of the Turkish slaves purchased by the Delhi Sultan. These range from fairly precise identification of clans and tribes like the Ilbari/Ölperli and Qipchaq to vaguer references to a ‘Turk from Rum’ or ‘[Qara] Khita'i’. There is nothing particularly exceptional about this information. It is when Juzjani enlarged on Sultan Balban's background that he displayed the extent of his knowledge regarding the eastern Qipchaqs. He mentioned that Balban was an Ilbari/Ölperli Turk, a tribe associated with the Qipchaq, Qanqali and Yimak.Footnote ³⁰ In the concluding verse to his chronicle he eulogized Balban as ‘Khan of the Ilbari/Ölperli and Shah of the Yimak [Khan-i Ilbari ast wa Shah-yi Yimak]’.Footnote ³¹ That they were all associated tribes in close proximity to each other is clarified in his account of the campaigns of Batu, Khan of the Golden Horde, in the region of the dasht-i Qipchaq. Juzjani listed Batu's victories against (amongst others) the tribes of the Qipchaq, Qanqali, Yimak, Ölperli and Rus.Footnote ³²

Details mentioned by Juzjani echo in Amir Khusrau's incidental remarks concerning the spread of the Turkish language in north India. In the Nuh Sipihr (completed 1318), the author observed that there were three pearl-like languages in the world: Arabic, Persian and Turkish. While Arabic was used by those in the religious sciences for scholastic purposes [‘ilmwari], Turkish possessed a grammar and dictionary [ahl-i hunar sarfeh-yi surf wa lughat-i zir o zabarwan] but was used (only) by the administrators and the military personnel [sahib-i ‘amilan; saran-i sipah].Footnote ³³ Amir Khusrau further noted that the Turkish language came from the Qanqali, Uyghur, Irti (? = Irtyush?) and Ghuzz tribes, from the lands of the Qipchaq and Yimak. It then spread through the world.Footnote ³⁴

Although in the Nuh Sipihr Amir Khusrau never clarified whether Turkish was spoken in India, suggesting instead that Persian or the vernacular subcontinental languages were more popular, in the Dibacha-i Diwan Ghurrat al-Kamal (compiled 1286–93) he made some other interesting observations. Here he exclaimed at the unique linguistic abilities possessed by Persian scholars of Hindustan in mastering foreign languages. ‘I have seen many Persians, not Turks [chandin tazik na Turk]’, commented the Persian poet, ‘who have learnt Turkish studiously and industriously [ba-ta'allum wa kasab], in Hindustan. And they speak [Turkish thus] that (when) the eloquent speakers of that language, (fusaha'yi an ta'ifa), come from Turkistan/‘upper lands’, they are astonished [furu mandand].Footnote ³⁵

According to Amir Khusrau, Persian litterateurs in India—individuals like himself—had mastered Turkish because it was the language of governance, and they had learnt these languages without ever visiting the ‘upper lands’. The possibility that this was no idle pastime but a consequence of the need for patronage was incidental to his fulsome remarks about the superlative intellectual capacities of the Persian secretaries to learn foreign languages. At any rate, an aspiring courtier such as Amir Khusrau, searching for patronage amongst frontiersmen of Turko-Mongol background, displayed a remarkably cogent knowledge of Turkish tribes and their relationships with each other. This was obvious from the confident connections he made between the Qipchaq and Yimak/Kimak tribes.

In the tenth century the Qipchaqs were dependent but distinct from the northern Yimak/Kimak people from whom they had already separated. By the twelfth century the two groups were in association again, but the Yimak/Kimaks were the dependent, depleted group who had migrated south to settle near the Qipchaqs in the Khwarazm region. Both Juzjani and Amir Khusrau remained sensitive to these associations. This is further clarified by the connections they made between the Qanqali, the Irtyush and the Ölperli tribes located on the eastern belt of the dasht-i Qipchaq, all of whom, as P. Golden has shown, were a part of the Qipchaq confederacy.Footnote ³⁶ Chinggisid invasions had devastated these tribes, pushing survivors westwards. Nor were they secure here: Juzjani described how in the western lands of the dasht-i Qipchaq, the Golden Horde under Batu, conquered, enslaved and dispersed these groups.

While the information on the Turkish tribes is extremely fragmented in Sultanate chronicles it is also unexpectedly accurate; the Persian literati based in Delhi were remarkably well informed of geopolitical developments in the steppe ‘upper lands’. This information may have reached them from a variety of intermediaries: geographical texts and travelogues perhaps, as also travellers and merchants, and most definitely their kin—Amir Khusrau himself was of Turkish descent. Their sources notwithstanding, it was certainly germane information to carry as the literati searched for patrons amongst immigrants from the ‘upper lands’.

And yet this was a knowledge that the Persian secretaries chose, as it were, not to wear on their sleeves. As we have already noticed, their scanty remarks concerning the Turko-Mongol provenance of the Tughluqs and their frontier retinues were quite remarkable for their silences. Equally noteworthy was their actual cognizance of the background and languages of these new migrants. The silences and elisions were deliberate, a difficult task because frontier commanders who went on to become Sultans were also the subjects of their elaborate eulogies. These eulogies had to reinvent their protagonists in ways that displaced their troublesome ethnicities. Ghiyas al-Din Tughluq is a good example of this phenomenon

The Qara'una military commander who arrived to sedentary habits only as an adolescent is remembered in Persian historiography primarily as a warrior in the cause of Islam. According to the mid-fourteenth-century historian, Ziya' al-Din Barani, Ghiyas al-Din stood as an impenetrable wall against the onslaught of the Mongols. This service on the frontier was further embellished by his actions against the reigning usurper, Sultan Khusrau Khan Barwari. Persian chronicles were not only depreciative of Khusrau Khan's slave origins, his recent conversion to Islam and his apostasy after becoming Sultan, they also condemned him for murdering and despoiling the harem of his master, Mubarak Shah Khalaji (1316–20). If Islam was threatened by the Mongols across the frontier, it was challenged in the capital by the usurper-Sultan under whose malevolent influence all appropriate norms governing service and loyalty were forsaken and idol worship had commenced in the palace. In these circumstances, Barani narrated, Ghiyas al-Din fulfilled the promise carried in his title: by thwarting Mongol invasions and overthrowing Khusrau Khan Barwari, he was indeed the ‘saviour of religion’.Footnote ³⁷

Amir Khusrau's text complicated Barani's narrative by recollecting sufficient details about the frontier commander's retinue and the degree of animosity he faced in Delhi.Footnote ³⁸ But this also served a useful purpose. Despite the odds faced by the frontier commander, and at the brink of disaster, Ghiyas al-Din's fortune changed because of divine benediction. Even though rectitude was by his side he was almost bested in battle until a God-sent opportunity turned fortune in his favour.Footnote ³⁹ As the account of his reign unfolded in these histories, Ghiyas al-Din proceeded to rule according to the norms of governance of his more worthy predecessors. The Qara'una frontier commander together with his retinue of Ghuzz, Turks and Mongols of Rum and Rus, and a smattering of Khokars vanished as if they had never existed.

Persian Historiography and the Turks and Mongols

Imbedded in the narrative of Ghiyas al-Din's rise to power and accession to the throne were all the tensions faced by the Persian literati as they wrote their histories of the Delhi Sultanate. The literati were a fairly composite body of people with some training in Islamic theology and jurisprudence and considerable facility in Persian prosody and poetry. Their juridical training, their skills in diplomatics, accountancy, as scribes and raconteurs made them a very valuable body of people in the administration of the Sultanate. Their intellectual training, acquired skills and professional preferences are an important reason why I use the euphemism a ‘Persian Secretarial Class’ to describe them. Although a small number of these individuals had appeared in north India at the end of the twelfth century, their large-scale immigration into north India occurred during the 1220s following the invasions of the eastern Iranian lands by Chinggis Khan and his commanders. By training and disposition, by their shared aristocratic backgrounds and, not least of all, through their shared traumatic experiences of displacement and immigration, the literati were an extremely class conscious, conservative body of people. They linked their interests with the state and produced histories (tawarikh) that valorized the stable and safe universe that hazrat-i Dehli, the sacred city of Delhi, provided to Muslims while the rest of the Islamic world was in ruins.Footnote ⁴⁰

These literati presented a synthetic image of an ideal Muslim-Persianate civilization whose finest protocols were present in the court of the Delhi Sultan. And yet, it is important to note they did not all speak with one voice. We need to distinguish the early-thirteenth century authors from the mid-fourteenth, two ends of a continuum which was bisected, so to speak, by the maverick Amir Khusrau.

Thirteenth-century chronicles were produced during the rule of the Turkish Slave regime, a time when the rulers and dispensers of patronage were of unfree origin. As I have detailed elsewhere, Turkish military commanders were rich, powerful and could be generous patrons but, in the context of the aristocratic pretensions of the litterateurs and their violent experience of the Turks and Mongols, writing a panegyric for Turkish slaves demanded considerable ingenuity. But these Turks, Fakhr-i Mudabbir and Juzjani argued, were quite different from their marauding compatriots in the steppe. Turkish slaves did not carry any lingering affection for their ‘hearth and homes’ in the steppes.Footnote ⁴¹ Instead, once they were reborn as servants of Islam they brought glory to their master. Juzjani sometimes ruminated on how the protagonists in his work were not ordinary Turks. They were of aristocratic families in the steppes (thus, Iltutmish and Balban), removed from the steppes and their families to serve a greater destiny.Footnote ⁴² Their passage into slavery was God's way of rescuing them from the Mongol holocaust and casting them on a path where they were fated to be rulers and protectors of Islam in India. The quality of being a Turk, per se, was not a problem in these texts; it was their slave origin that posed the conundrum. Elaborate attempts were therefore made to discuss one without the other and when the two could not be resolved it was left to the mystery of divine providence where their positive contributions as Sultans in the service of Islam remained as proof of supernatural intervention.Footnote ⁴³

We need to keep in mind that these records were produced in the early years of the Sultanate and its chroniclers struggled to place the history of the regime within a constellation of Muslim Sultanates. This was a Herculean enterprise because it seemed at that time that the Mongol holocaust marked the end of history itself. While Juzjani's chronicle mimed the traditions of ‘universal’ history it carried no grand theorization regarding the arrival of Islam in India or of the nature of the political regime. Juzjani's narrative, and those of his contemporaries, rarely shifted beyond the annals of events, battles and literary tropes recording the victory of Muslim arms against infidels.Footnote ⁴⁴

By contrast, mid-fourteenth-century histories were far more complicated. They were produced by authors of subcontinental provenance, descendents of émigrés who had prospered in their new homes. By the middle of the fourteenth century they had family histories that boasted of aristocratic, urbane accomplishments. These authors were also distant from many of the events that they described and the hindsight allowed them the perspective to contextualize their world within a larger political experience of Muslim governance on Islam's eastern frontier. In the 1350s, scholars such as ‘Isami and Barani were the first to suggest that the prehistory of the Delhi Sultanate lay in the Ghaznavid state and that its ruler, Mahmud Ghaznavi, provided the foundation and inspiration integral in the making of the Delhi regime. The Mongols and infidel Hindus were the great ‘Others’ in these narratives and the Persianate and class conscious, aristocratic virtues of the ideal state were creatively memorialized in the Ghaznavid state, now the templates for the Delhi Sultanate.Footnote ⁴⁵ Cast within a historical narrative it allowed for a more self-reflective, linear rooting of the Sultanate in the great traditions of Muslim statecraft. But it also left little space in these narratives for frontier commanders and their retinues—with ‘breath that smelt like an arm-pit’ with ‘voices hoarse and shrill as a jackdaw’—whose origins lay amongst the Mongol hordes or frontier tribesmen who were the great threat to the Sultanate. These people were reinvented like Ghiyas al-Din Tughluq as the ‘saviour of religion’, and their more problematic qualities were blithely ignored.

Bisecting the two worlds was the maverick figure of Amir Khusrau: the son of a Turkish slave, born and brought up in Patiali and Delhi, whose personal experience never went beyond the subcontinent. He was a court eulogist and a poet, comfortable and proud of his background and of his skill as a litterateur, able to reproduce the most difficult literary styles and innovate with great ease. His grand paeans extolled the virtues of respective patron-Sultans setting new standards in form, style and rhetoric. And yet it was when he wrote about his domicile, his craft, and his friends that a rare sensitivity and eloquence crept into his work. He received patronage from different Sultans but rendered homage only to his pir Shaykh Nizam al-Din Auliya, near whom he was eventually buried. His historical works were constructed around specific themes and episodes: the victories of ‘Ala al-Din Khalaji, the seizure of power by Jalal al-Din Khalaji and Ghiyas al-Din Tughluq, the meeting between father and son, the rulers of Lakhnauti and Delhi.Footnote ⁴⁶ These were not large texts and the constant shifts in themes, styles and patrons also meant that the author never theorized on the role and character of the Delhi Sultanate. If there was an overarching, comprehensive ideology regarding its political manifestation, it was to suggest that the subcontinent—more so than the Sultanate—housed a vibrant and unique Persian culture.Footnote ⁴⁷ This allowed the incredibly successful court poet the freedom to discourse and insert random insights regarding the conditions of his age. These were fleeting remarks and observations but, as we have already noticed, they covered subjects that were often erased in the master narratives of the fourteenth-century grand histories.

Through this paper, it is the information provided by Amir Khusrau that often led to a further interrogation of other literary materials. But it has also left us with a host of unanswered questions. It is clear from the information already discussed that the Sultanate was not a cultural monolith, that its rulers came from social and cultural backgrounds quite distinct from that of its Persianate secretarial class who provided the histories and eulogies of the state. These differences generated some discomfort especially when these ‘frontiersmen’—a useful term to capture their marginal status in the representations of urbane, Persianate Sultanate society—emerged as the lords and masters of the realm.Footnote ⁴⁸ In the next section, I interrogate the Persian materials further to see how some of these tensions were manifest in their records.

Turko-Mongol practices and a Persianate Muslim Order

Delhi was the ‘Sanctuary of Islam’ for a large number of immigrants of different social backgrounds, ethnicities and regions. A poet such as ‘Isami writing in the mid-fourteenth century dutifully recorded the diverse backgrounds of these people but also went on to suggest that the common denomination of being ‘Muslim’ bonded immigrants as they sought asylum from the Mongols.Footnote ⁴⁹ Penetrating this discourse is extremely difficult; it was certainly a subject that the Sultanate chroniclers did not want to accommodate in their narratives. Once the frontier commanders became Sultans, however, ‘difference’ was harder to ignore and their new patrons tested the creativity of their eulogists in interesting ways. They introduced traditions that were clearly outside the experience of the Persian literati and yet many of their novel practices were seamlessly assimilated in the records of the chroniclers without a comment. There were other Turko-Mongol traditions of the émigrés that were far harder for the Persian literati to gloss. In their efforts to work out resolutions—not always very effective ones—literary materials of the time left spaces that can be usefully enlarged for an insight into the dialectical processes through which a Persianate Muslim order was rather uneasily fashioned. The evidence on the subject is quite dispersed and rather than pursuing random and decontextualized instances I have selected three examples through which I can develop the range of cultural interaction more precisely in my discussion.

The first example concerns an administrative system used by ‘Ala’ al-Din Khalaji and Ghiyas al-Din Tughluq. Concerning the vital communication system established by Ghiyas al-Din to knit his expanding empire, Barani refers to ulagh/ulaq on several occasions as a ‘horse-post’ without any need for further elaboration.Footnote ⁵⁰ The same institution was described by the Morrocan traveller Ibn Battuta, who found the system novel enough to merit a description. He distinguished ulagh/ulaq, a horse-post, from dawa, a courier who travels on foot.Footnote ⁵¹ Clearly the system was new; but it did not have a long life. At the end of the sixteenth century the Mughal historian Nizam al-Din was not at all confident that his audience would know the meaning of the term. When he narrated the events of Ghiyas al-Din's reign and mentioned the ulagh/ulaq system he explained the term with a synonym: ‘During this time, the dak chowki, which in the language of those people was called ulagh/ulaq, arrived from Delhi and brought orders’.Footnote ⁵²

Barani did not need to explain the meaning of ulagh/ulaq because his audience was familiar with it. There is no evidence to suggest, however, that they knew of its history and that it was an administrative innovation authored by Chinggis Khan, a universally hated figure in Sultanate historiography. The Mongol historian, Juwaini, clarified that ulagh/ulaq was a courier system that connected the vast Mongol domains of the Great Khan. His history had explained how it was originally one of the qubchur taxes, a contribution levied upon the Mongols for providing mounts and other sustenance for the couriers. This changed through the thirteenth century until ulagh/ulaq came to be a tax levied on the peasantry. Certainly by Ghazan Khan's reign (1295–1304) it was a part of the qalan taxes levied on non-Mongols.Footnote ⁵³

It was clearly in its late-thirteenth century form that Barani used the term ulagh/ulaq when he was describing the postal relay system in ‘Ala’ al-Din Khalaji and Ghiyas al-Din's reign. Under Ghiyas al-Din ulagh/ulaq was a courier system maintained through local provisions/taxation, connecting Delhi with its frontier regions. The mutations in the ulagh/ulaq system could have occurred independently in the Sultanate and Il Khanid territories; otherwise its transmission into Khalaji and Tughluq administration was very quick—‘Ala’ al-Din Khalaji and Ghazan Khan were contemporaries. In Ghiyas al-Din's reign this courier system had broken down while Ulugh Khan, the future Sultan Muhammad, was besieging Arangal in the Deccan.Footnote ⁵⁴ It is unlikely that ulagh is of an earlier Oghuz inheritance: the Seljuqs (of Qiniq, Oghuz background) much to the wazir Nizam al-Mulk's dismay, paid scant attention to the barid (postal, spy system) network within their dominion and much of the Siyasat Nama was spent marketing its virtues.Footnote ⁵⁵ Certainly, by the time that the Qara'una military commander Ghiyas al-Din used the system, it had reached a stage of development roughly coterminous with the Chaghatayid and the Il-Khanid political dispensations (post-1227 and post-1256, respectively).

Although the usage of the ulagh system by Ghiyas al-Din Tughluq does show Sultanate administrative proximity to a Mongol source, it did not carry with it any implication of large-scale social and political reorganization as it had for the Mongols under Chinggis Khan. In other words, ulagh/ulaq did not challenge the structures of early fourteenth-century Sultanate society. For a narrator such as Barani it was an efficient administrative system and, since its provenance was without any apparent consequence, it provoked little interest. For all that it seemingly mattered, ulagh/ulaq was a new name grafted on to old administrative procedures that were revamped and made more efficient during dynastic change.

On the other hand, the response of the Persian chroniclers to the titles used by the Delhi Sultans and their important military commanders indicates a more complex response to steppe traditions. The Khalajis and early Tughluq Sultans used fairly conventional titles couched in Arabic and Persian. But they also defined membership in the imperial kin-group by the Turko-Mongol honorific khan, designating brother, son, or honoured kinsman of the monarch. This was in contrast to the usage of malik and amir, a title used by military commanders. While khan designated kinship within the imperial lineage, it was also deployed by the Khalajis and Tughluqs as a title of privilege through which some fictive kinsmen were honoured while others clearly excluded. The nuances in status become evident from the following example narrated by Barani at the time of Ghiyas al-Din Tughluq's accession when he honoured his kinsmen and collaborators. Barani reported:

In other words, the son-in-law, and the adopted son were not honoured by the title of Khan, but the non-kinsman and honoured associate, Bahram Ai-aba, was treated like a brother and hence given that title. Furthermore, seniority between the sons of Ghiyas al-Din, all Khans, was further clarified by the grant of the title Ulugh Khan to Malik Fakhr al-Din Juna, the future Sultan Muhammad Shah (1324–51), and his public acknowledgement as wali ahd (heir-apparent). Ulugh which literally means ‘great’ in Turkish distinguished the future Sultan from his peers.

Turkish titles were also used by the Khalajis. The father of Sultan Jalal al-Din Khalaji (1290–96) was Yughrush and Mahmud Kashgari explained that the title Yughrush meant wazir;Footnote ⁵⁷ his nephew ‘Ali Garshasb was the future Sultan ‘Ala al-Din (Garshasb was an ancestor of Afrasiyab). On seizing power in 1296, ‘Ala al-Din's lineage became the imperial one and the monarch could appropriately dispense the honorific Khan upon his kinsmen. This appellation was absent in their earlier titles and he now proceeded to distinguish his brother Almas Beg with the title Ulugh Khan (Great Khan); another brother was entitled Qutlugh Tegin (Qutlugh = auspicious, and Tegin = prince, hence an auspicious prince); his brother-in-law Malik Sanjar received the title Alp Khan (Alp = powerful).

Turkish titles were not a novelty for the Tughluqs or the Khalajis; they were systematically given by Iltutmish to his bandagan-i khass, Turks and non-Turks, alike. Iltutmish's actions were, however, quite innovative; the title of Khan was never used by his predecessors. Perso-Arabic titles were in currency under the Ghaznavids, the Ghurid Sultans and in the usage of his own master, Qutb al-Din Ai-beg. The use of the royal honorific khan—far be it for slaves—was unprecedented and the Persian secretaries were aware of its novelty. This is apparent from Juzjani's long anecdote of an embassy sent by Ulugh Khan (the future Sultan Balban) to the court of the Il-Khanid monarch Hulegu. The letter to the Mongol monarch was in Persian and when it was translated into Mongolian the emissary replaced Khan in Balban's title with Malik. Juzjani clarified: ‘the custom of Turkistan [qa'ida-i Turkistan] is this that there is but one Khan, no more, and all the others have the title of Malik.’ In Juzjani's narration, Hulegu Khan knew of Balban's usage of the Khan title and honoured him sufficiently to protest its omission when his letter was read out with the honorific missing. He asked for the title of Khan to be restored. Juzjani concluded his report: ‘all of the Khans from the lands of Hind and Sindh who went to the presence of the Khans, their titles were altered [tabdil kard] in all of the documents proffered to the Mughal and they were referred to as Malik. But they confirmed the title of the great Ulugh Khan without change [as in] the original’.Footnote ⁵⁸

The assumption of titles of steppe provenance served as effective communicators of status for a local as well as a distant audience. It is therefore interesting to note the emergence of specific protocols relating to the kind of titles that monarchs and their subordinates could carry. Some of the early Sultans—Qutb al-Din Ai-Beg, Shams al-Din Iltutmish and Ghiyas al-Din Balban—carried ‘compound’ titles, with both Turkish and Perso-Arabic elements.Footnote ⁵⁹ Barring these three examples no Sultan after 1286 ever took on a Turkish title, not even Khan. As princes they had either taken Turkish titles or ‘compound’ ones, but these were given up for formal Perso-Arabic ones when they ascended the throne. From the individuals already cited note the examples of ‘Ali Garshasb who became ‘Ala al-Din Khalaji and Ulugh Khan who later took the title of Sultan Muhammad Shah Tughluq.

It would appear from the evidence cited thus far that administrative practices of steppe provenance that promoted efficient governance were not a subject of concern to the Persian literati. They were as candid in reporting the usage of Mongol and Turkish titles and comprehended their manipulation at the time of dispersal of new honours. As Juzjani's report suggests, the Persian secretaries understood the cultural value attached to these titles and were sometimes involved in adjusting them according to the demands of the situation.

At least as far as the Persian literati were concerned the ethnic backgrounds of the ruling elites were not a problem as long as they abided by the larger templates of a Persianate Muslim order. The Turkish slaves of Iltutmish, all of whom were given Turkish titles by their master, were applauded for their abilities to forsake their homes and families for their new world. This fiction allowed for the acceptance of the military slave as a loyal servant of his master and the realm. In the long duration the logic of this discourse was useful enough in persuading frontier commanders to abandon their old titles that carried significant Turkish elements and aspire to high status in their new world as ‘Ala al-Din (Glory of Religion) Khalaji or Ghiyas al-Din (Saviour of Religion) Tughluq.

Transitions from one world to another, however, were never quite as definitive, especially when moving from a frontier environment to the core territories embraced a large entourage and not just the individual military commander. The Persian secretaries were familiar with many details of the inhabitants of the Eurasian steppes but there were other specifics about the lives and traditions of the steppe peoples that were completely unfamiliar to them. On their part, frontier commanders like Jalal al-Din Khalaji and Ghiyas al-Din Tughluq must have retained Persian secretaries to help them keep accounts and manage their diplomatic life. Our ignorance of their service notwithstanding, it would not be presuming too much to assume that these secretaries were important members of the governor's household. Their influence—and through them, the influence of urbane Persianate traditions—were always tempered in the frontier camps by the large-scale recruitment of mobile warriors that constituted the retinue of the commanders. Bonds of ethnicity, common natal origins, and the contingencies of marginalization on the frontier created forces of social and cultural cohesion that Persian secretaries could not always penetrate.

At Ghiyas al-Din Tughluq's accession, the Ghuzz, the Turks and Mongols of Rum and Rus were the new political elites of the Sultanate and the monarch was as reliant on their continuing support as he was on the Persian Secretarial class. The new monarch had to remain sensitive to the contrasting bodies of people and lexicon of associations that he had knitted together in the making of his war band. Acculturation into the Persianate world of the secretarial classes, even if such transitions could be accomplished quickly by commanders who were only recently awara mardi, would have meant distancing himself from the very fraternity that had made his rise into political prominence possible. Since this was not a feasible alternative, when Ghiyas al-Din Tughluq marched into the capital, he actually brought the frontier with him to Delhi.

The hegemonic narratives of mid-fourteenth-century Persian chroniclers are quite impoverished on subjects concerning cultural plurality within the Muslim community. On the other hand, the amateur ethnography of the Moroccan traveller Ibn Battuta, a travelling jurist, is far more direct in recording the unfamiliar rituals and traditions of his Muslim coreligionists. He arrived in north India during the reign of Muhammad Shah Tughluq and described the ritual procession of the king in ways that was very different from the Persian chronicles.

This was the ritual procession of the king where his ornately decorated saddle cover, ghashiya, was carried before him. Ibn Battuta provided descriptions of the ritual from the reign of Muhammad Shah Tughluq, which was apparently celebrated on the occasion of major festivals and whenever the Sultan returned to the city. Concerning feast days (‘id) the traveller noted:

Ibn Battuta added further details regarding the ritual at the time of the Sultan's entry into the capital:

While ghashiya has an Arabic etymology, meaning to cover, veil,Footnote ⁶² the origin of the ceremony lies in the accession and ceremonial rituals of the early Turks where the ‘Lord of the Horse’ would be identified with the newly enthroned leader, and the procession would celebrate the conquest of the four quarters by the Universal Emperor.Footnote ⁶³ Although the paths of its transmission into the central Islamic lands are unclear the tradition was followed in some of the major steppe-descended polities: by the Seljuqs, the Zangids and the Bahri Mamluks of Egypt (with a military elite of Qipchaq origin).Footnote ⁶⁴ At least in Syria and Egypt it was accepted as a ritual associated with royalty and performed by the Kurdish Ayyubids, who learnt of it from their Turkish patrons the Zangids. With the Ayyubids it was integrated as a part of their accession ceremony together with the ritual pledge of allegiance, bay'a, and the investiture from the Caliph.Footnote ⁶⁵

Detailed descriptions of the ghashiya ritual exist from the Mamluk Sultanate of Egypt where Ibn Taghribirdi clarified that it was a part of the accession ceremonies of the monarch and repeated on major festivals. Its performance in Egypt mirrors Ibn Battuta's description of the ceremony from Muhammad Shah Tughluq's court and al-Qalqashandi gives us the following description:

An important common feature between the Mamluk state in Egypt and the Delhi Sultanate were the common reliance upon Turko-Mongol personnel from the dasht-i Qipchaq for their respective armies. The Sultanate's link with the Eurasian steppe already present in Iltutmish's reign continued into the reign of Ghiyas al-Din Tughluq who was of Negüderid background, and his retinue of ‘Turks and Mongols of Rum and Rus’.

Just as most of the Persian chronicles ignored the composition of Ghiyas al-Din's retinue they paid no attention to his royal procession ceremony. Since Ibn Battuta's observations remained largely ‘unsubstantiated’ in the accounts of the Persian literati they did not draw the attention of modern scholars. Yet, Barani's description of ‘Ala al-Din's triumphant march to Delhi after Jalal al-Din's murder (1296) does possess some of the elements present in Battuta's description although completely different motives to the discharge of gold coins (panj-man akhtar, five mans of gold stars) amongst the crowds observing the Sultan's march are ascribed by the author.Footnote ⁶⁷ Equally selective was Yahya Sirhindi's early-fifteenth-century account of Muhammad Shah Tughluq's celebratory procession after his accession. The narrative is close enough to Ibn Battuta's description of the ghashiya ritual for us to follow its main features but the elisions are important as well. Sirhindi noted:

In Sirhindi's account, as in Barani's, the Sultan's triumphal processions receive due recognition but there is no reference to the ghashiya. Was the omission deliberate or was it an aspect of Turko-Mongol practice quite unfamiliar to Persian secretaries? Were they, in other words, just inadequate historians reifying the practice of their subjects either through ignorance or because of their own class and cultural prejudices?

This is a difficult question to answer and it might help if we disaggregated the two reports of Barani and Sirhindi; different factors influenced the production of the two texts. Barani was a contemporary of Ibn Battuta and both authors were in Delhi during Muhammad Shah Tughluq's reign. If the Moroccan visitor could notice and learn about the ghashiya during the brief period of his visit, so, theoretically speaking, could Barani. He noticed the ulagh and the manipulation of the Khan title to enunciate hierarchy within Ghiyas al-Din's new political dispensation without any problem. But these details did not disturb the larger point that the author wanted to make about the monarch in his history. In Barani's narrative Ghiyas al-Din was a ‘Saviour of Islam’, a morally righteous Muslim, renowned for his combat with the infidel Mongols. Now he was waging a war against a different heathen menace located in Delhi, a neo-convert slave, an apostate, who had killed his master and his heirs. The conflict between Ghiyas al-Din Tughluq and Khusrau Khan Barwari was over the future well being of the Muslim community. Incorporating details about the Turko-Mongol composition of Ghiyas al-Din and his retinue, or the practice of steppe rituals by the frontier commander would have complicated the binaries on which Barani had framed the qualities of his protagonist—the Muslim versus the non-Muslim—and his narration of the triumph of rectitude over evil. The author preferred not to tread these waters.

Writing a century later Yahya Sirhindi, a litterateur himself, was familiar with the writings of his predecessors. His history of the Sultanate, from the late-twelfth into the early-fifteenth century, is an interesting piece of synthesis. For the better part the narrative is reliant upon the histories of Juzjani and Barani but there are significant additions and omissions.Footnote ⁶⁹ Sirhindi followed ‘Isami in suggesting personal proximity between Yaqut and Sultan Raziyya and, while otherwise staying close to Barani's text, omitted any mention of ‘Ala’ al-Din Khalaji's price regulations. While Barani's rhetorical statements were deleted, the author inserted contemplative passages in verse and prose regarding destiny's stranglehold on humans. And yet none of these insertions and deletions shifted the larger narrative framework of Sirhindi, which remained entirely dependent on Juzjani and Barani. Sirhindi ignored alternative narratives, such as ‘Isami's, or details from Amir Khusrau that disturbed Barani's conclusions. This independent line of investigation would have complicated the author's work considerably and, in forcing him to question and depart from Barani's reportage, required the author to write the history of the Delhi Sultans afresh. Instead Sirhindi was satisfied with paraphrasing, pruning and collating material, shifting details about individuals without actually rewriting the history of the Delhi Sultans. As he excavated Juzjani and Barani for their information, he also treated them as artefacts that needed to be dusted off and preserved. The elisions and silences regarding frontier commanders, Turko-Mongol traditions and customs present in the earlier master narratives were then transported into the history of the other.

Conclusion: Frontier Commanders in a Persianate Milieu

In the beginning of this paper, I had clarified my intent to study the ‘ignored elites’ in the early Delhi Sultanate. This was an unusual project given that modern historiography on the Sultanate has ignored many subjects, but alas, not elites. Elites, of course, are never ignored; they get to be reinvented by their narrators in various ideological hues. Frontier commanders who seized power in the Sultanate in 1290 and later, were grandiloquently panegyrized by the Persian literati. The extent to which these patrons challenged the skills of their eulogists, however, should not be minimized. As the case of Ghiyas al-Din Tughluq clarifies, for a variety of reasons relating to his background, the geopolitics of his age, and the social and ideological precommitments of his eulogists, it was necessary for the Qara'una military commander to receive a brand new profile. There was a lot else about these frontiersmen which it was safer to simply ignore. Once ignored, the salient characteristics of these military commanders and the politics that made them important participants in the events of the Delhi Sultanate simply passed out of the realm of history. It is hardly surprising then, that many historians mark 1290 as the termination of Turkish dominance in the history of the Sultanate and the arrival of new ‘plebeian’ forces.Footnote ⁷⁰ As Barani explained, these plebeians were the new Indian converts to Islam that Sultans like ‘Ala al-Din Khalaji and Muhammad Shah Tughluq started patronizing. In modern historiography this started the process of ‘rooting’ the Sultanate in the subcontinent, processes through which Islam came to have the unique features so admired today in the practice of its sufis and bhakta sants.

Much of my paper was occupied in recovering the pasts of frontier commanders and discussing the ways in which their presence was elided in the records of the Persian literati. Reintegrating them in the histories of the Sultanate implies charting new genealogies for the state, its ruling elites and the processes that shaped the pasts of the Muslim community in the subcontinent. This would imply that rather than focusing only upon processes of indigenization and vernacularization of Islam in the regions of the subcontinent, historians need to be sensitive to the processes through which Muslim society in Delhi and its adjoining regions were constantly reconstituted through the infusion of immigrants from the Afghanistan-Punjab frontier. Barani, for example, gestured to the presence of these migrants in his chronicle. He referred to them quite derisively as nau-Musulman, or new Muslims. Although these Mongols were apparently Muslims, Barani used the epithet of nau-Musulman to communicate their alien, novel character since their politics and social and cultural practices were so abhorrently different from people of his upbringing. Some of these Mongol migrants gained patronage in the short duration but, if we follow Barani most of these people remained segregated in Sultanate society and were executed. And yet, despite Amir Khusrau's information on the composition of Ghiyas al-Din's retinue and the resistance that they faced from the elites of Delhi, the military commander and his contingent were never ascribed the epithet nau-Musulman. One of the reasons why this did not happen, of course, is because Ghiyas al-Din and his retinue were rulers and patrons and it would not be politic to refer to them as new-Muslims.

Another reason for the hesitation to refer to these frontiersmen-turned-Sultans as nau-Musulman arose from the fact that the new rulers were themselves sucked into the structures of power and social hierarchies present in Delhi. All the information that we have on Ghiyas al-Din suggests that he moved very quickly to install conventional modes of governance once he became Sultan. This included collaborating with the Persian literati in the preservation of the social and moral order familiar to them. Ibn Battuta communicated the extent to which Ghiyas al-Din Tughluq's son, Muhammad Shah Tughluq (1324–51) was invested in this project. Muhammad Shah Tughluq issued orders that ‘all were required to show a knowledge of the obligations of ablution, prayers and the binding articles of Islam. They used to be questioned on these matters; if anyone failed to give correct answers he was punished and they made a practice of studying them with one another in the audience hall and the bazaars and setting them down in writing’.Footnote ⁷¹

Records of this nature suggest ways in which an Islam that abided by a rigorous interpretation of its rituals was reproduced in the core territories of the Sultanate. It consolidates the dominant historiographical image where creative ferment in Islam came through sectarian and doctrinal controversies amongst Muslims in the core territories of the Sultanate in contrast to the more heterogeneous populations on the frontier where there was interaction between Muslim and non-Muslims.Footnote ⁷² And yet the ‘creative encounter’ that occurred in Ghiyas al-Din Tughluq's capital was not just between Muslims and non-Muslims, it was between old and new Muslims, between the urbane elites of the capital and the frontier commanders, the Sultan and his entourage. In this interaction, however, hierarchies were reversed. It was the new, rustic Muslim, recently arrived from the frontier who was the Sultan. The ‘saviour of Islam’ may have patronized the Persian literati and the learned jurists, but he also continued to practice his steppe rituals in Delhi.

It is in this context that we need to remember the ghashiya ritual performed by Muhammad Shah Tughluq, a ritual that Sultanate chroniclers ignored, but a public ceremony that was performed during ‘id. In the retinue that followed the procession to the festival ground were the great qazis of the city. Should it happen to be bakr-i ‘id, Ibn Battuta noted, the Sultan himself did the honours of sacrificing the camel. Commingled in the rituals of the ghashiya and ‘id were strands from multiple backgrounds but the elements that reminded observers of the frontier origins of their masters were not the ones that were transmitted to posterity.

It is hard to determine how long royal rituals like the ghashiya were practiced in Delhi because no Persian chronicler wished to record its performance in the first place. We know of it only by accident through Ibn Battuta's travelogue. But we can gauge the slow loss of comprehension of some steppe traditions if we recall the example of the ulagh/ulaq. By the end of the sixteenth century the term was equated with an alternative system and its specificity was erased. Alternatively, while Mongol titles like Khan were seamlessly incorporated within a Persianate tradition and persisted over the long duration, there were other Turkish titles like Iltutmish whose meaning, as Simon Digby's research has brought out, was already confused by the end of the sixteenth century. Digby also points out that the manuscripts copied in the 1700s altered the name of the Sultan according to a false identification with a word (altamish) current at that time.Footnote ⁷³ Copyists thereafter were less sure how to spell the more exotic Turkish titles and errors crept in. As Peter Jackson's efforts have shown, restoring Turkish titulature to their correct form is now an incredibly laborious task.Footnote ⁷⁴

It may appear commonsensical to note that what is understood today as a part of ‘Muslim’ tradition is quite removed from the ways in which it was understood in 1700 or 1324. But having said that, researching the contours of Muslim society in 1324 means that we have to step out of teleological modes of analysis and remain sensitive to the ways in which the culture and politics of that moment constructed or elided their complex aspects. The study of Jalal al-Din Khalaji, Ghiyas al-Din Tughluq, or the much later Bahlul Lodi—all frontier commanders who became Sultans—stands as a salutary reminder that the great traditions of Islam were not just the product of interaction amongst social groups in the Gangetic plains or the Deccan; a variety of frontier traditions brought by periodic migrations of military commanders also impacted on this world. What kind of responses did these intrusions raise? Is it possible to locate the contexts in which they were produced? It is through a study of this dialectic—sometimes only fleetingly visible in our historical narratives—that we can understand the construction of the social and cultural lineages of Muslim society and structures of authority in the history of the Sultanate. Otherwise the complexities introduced by the presence of frontier traditions and their importance in the formation of Muslim societies and politics will remain where the Persian literati of the age sought to consign them: on the periphery of our narratives, on the frontier. That would be an ironic location to place many of the Sultans of Delhi.