03 POLITICAL CULTURE IN THE PERSIANATE WORLD

If Rajendra Chola’s raid on Bengal had operated within the world of the mandala, the digvijaya and South Asia’s sacred geography, Mahmud of Ghazni’s 1025 raid on Somnath – like his sixteen previous raids on northern India – was driven by very different ideas. The core of Mahmud’s forces was composed of Turkish slaves or mamluks, who, as young men typically captured in war and separated from their kin groups, had been recruited to eastern Afghanistan from Central Asia on account of their exceptional military skills. They were ‘Turks’ inasmuch as their native language was Turkish, dialects of which in the eleventh century were spoken from the western frontiers of China across Central Asia to the Oxus River (Turkish-speakers would not occupy Anatolia – that is, most of present-day Turkey – until several centuries later). And they were slaves inasmuch as they were attached neither to their natal kin nor to land, but to their masters, who were typically state officials. But, because they were entrusted with weapons and lived in close quarters with their masters, this type of ‘elite’ or ‘military’ slavery differed fundamentally from the plantation slavery typical of the early modern Atlantic world.

The kingdom that Mahmud inherited, the Ghaznavid sultanate (975–1187), had arisen from its declining parent Samanid kingdom in Central Asia (819–999) at a time when this sort of military slavery was already a well-established institution. In 962 the Samanid commander-in-chief, Alptigin, abandoned the court in Bukhara and carved out a semi-independent state with its capital in Ghazni, in eastern Afghanistan. He was succeeded in 975 by his slave Sabuktigin, who, after consolidating his rule as an independent sovereign, launched a number of raids on the ruler of Peshawar in the north-western corner of the Indus valley. On his death in 997, Sabuktigin was succeeded not by another slave but by his own son, Mahmud, thereby launching a new dynasty at Ghazni.

Like their Samanid predecessors, the Ghaznavid sultans continued to recruit Turkish slaves and freemen from Central Asia, where cavalry warfare, together with the breeding and herding of horses, had been woven into the fabric of pastoral life. Even young boys acquired exceptional riding skills, including shooting arrows at full gallop. Writing in the ninth century, the Arab historian al-Jahiz captured the awe with which outsiders viewed these fighters:

If a thousand of their [Turks’] horse join battle and let off a single shower of arrows, they can mow down a thousand [Arabs’] horse. No army can withstand this kind of assault. The Kharajites and the Bedouin have no skill worth mentioning in shooting from horseback, but the Turk can shoot at beasts, birds, hoops, men, sitting quarry, dummies and birds on the wing, and do so at full gallop to fore or to rear, to left or to right, upwards or downwards, loosing ten arrows before the Kharajite can nock one … and if they do turn their back, they are to be feared as much as deadly poison and sudden death; for their arrows hit the mark as much when they are retreating as when they are advancing.¹ +++(5)+++

By contrast, Indian archers in Mahmud’s day were for the most part infantrymen, mounted archery not being widespread. Additionally, a shortage of extensive pastures and competition from sedentary agriculture in India drastically reduced the supply of war-horses.²+++(4)+++ This was never an issue in Central Asia. Moreover, by inhabiting the heart of the vast, intercommunicating zone between China and the Mediterranean basin, the peoples of Central Asia readily adopted both offensive and defensive military technologies that passed along the trade routes connecting the two ends of Eurasia. These included not only the most efficient technologies associated with horsemanship, such as iron stirrups or heavy saddles, but also siege equipment such as the trebuchet, which was used for hurling large missiles, or mortar for cementing masonry in the bastions and curtain walls of forts.+++(4)+++

Central Asian Turks and north Indian warrior clans also inherited very different conceptions of political territory. North Indian ruling lineages were organized in large, patrilineal kin groups which were dispersed on to ancestral lands and controlled the peasant society that produced the land’s surplus wealth. This link between land and kin inclined such clans to identify strongly with particular, ancestral territories. As pastoralists in Central Asia, Turks had also been organized into lineage groups, but their kin ties had been distorted by the institution of military slavery, which detached them from their clans and took them into unfamiliar households in eastern Afghanistan or Khurasan – that is, the Persian-speaking region embracing today’s north-eastern Iran, western Afghanistan and the territory up to the Oxus. Whether as mobile pastoral nomads in Central Asia or as uprooted slaves serving sultanates in Khurasan or Afghanistan, Turks had little or no attachments to ancestral lands.+++(5)+++ This inclined them to envision political space as open and unbounded, which helps explain the elastic, expansive nature of sultanates, in India or elsewhere. By contrast, states of eleventh-century north India, such as Rajendra Chola’s empire in the south, were rooted in the ideology of the mandala, with its fixed centre based on a maharaja’s palace or royal temple, surrounded by concentric circles populated by allies and enemies.+++(4)+++

What gave the Ghaznavid Turks their special character, and perhaps their clearest contrast with contemporary Indian states, were the geostrategic forces driving their continued raids on north India in the early eleventh century. Those raids aimed not at appropriating territory but at plundering wealthy cities and their temples, especially for gold or silver. Taken back to Ghazni, this bullion was typically melted down into coins to finance campaigns in Central Asia and Iran, where the annexation of land was very much the objective. Cash was also needed for purchasing war-horses, slaves and manufactured goods, and for paying the salaries of Mahmud’s army and administrative hierarchy. As this hierarchy was elaborated, and as the size of the army grew with the addition of more mercenaries or slaves, ever more cash was needed to pay them. This in turn required still more raw treasure, readily acquired by launching more raids, for which still more troops were needed. The result was a self-catalysing cycle that was inherently expansive and predatory, based above all on mobile wealth.+++(4)+++ The raids also reversed historic patterns in the transregional flow of precious metals.³ Whereas for centuries such metals had poured into India, mainly in payment for textiles and spices, after the early eleventh century the bulk of precious metals began flowing from India to Central Asia and the Middle East, both for trade and for maintaining Ghaznavid armies in those regions. In 1009 alone Mahmud seized seventy million minted coins, amounting to 136 metric tons of silver, from the mountain fortress of Kangra in modern Himachal Pradesh.⁴

The transregional circulation of wealth through Central Asia, the Iranian plateau and north India was the material counterpart to a growing canon of Persian texts that spread through those same regions. By elaborating distinctive norms of kingship, governance, courtly etiquette, social comportment, Sufi piety, poetry, art, architecture and so on, these texts provided the ideological scaffolding that sustained an emerging Persianate world. At the same time, royal courts, regional political centres, the lodges or shrines of Sufi shaikhs (venerated religious leaders) and schools (madrasas or maktabs) provided the institutional bases from and through which such texts circulated.+++(4)+++ From the days of Mahmud of Ghazni on, these networks spread across ever greater stretches of territory. And, as this happened, an urbane, literate and transregional Persianate culture defined by an evolving literary canon was superimposed over a number of vernacular ones.⁵ The prestige associated with this culture also attracted non-Persian-speaking peoples into its field of influence. In particular, from the ninth century confederations of Turkish-speaking pastoralists migrated westwards from the fringes of western China into Central Asia and the eastern rim of the Iranian plateau, which included Khurasan and Afghanistan. As these nomadic or semi-nomadic peoples encountered Persian-speaking peoples of Central Asian oasis towns and their rural hinterlands, they rapidly assimilated both the Persian language and the broader Persianate culture associated with that region’s agrarian and urban classes. Whenever they achieved positions of power – thanks to their equestrian and fighting skills – these Persianized Turks lavishly patronized the entire gamut of Persian culture, not least in order to earn political legitimacy for themselves in an expanding Persianate world.+++(4)+++

It was the Ghaznavid sultans, successors to Mahmud, who initiated the diffusion of Persian culture into north India. In 1040 Mahmud’s son Mas‘ud (r. 1030–40) launched an expedition to recover the city of Merv (in modern Turkmenistan) from another confederation of Persianized Turks that would soon dominate the Middle East – the Seljuqs (1037–1194). He failed to do so, and the defeated sultan was forced to flee across the Indus before being assassinated by one of his own men. Having lost their realm in Iran and Central Asia to Seljuq power, Mas‘ud’s successors then focused their energies on south-eastern Afghanistan and north-west India, with Lahore serving as their base in the Punjab. After several decades of instability, the second half of the eleventh century saw the relatively steady reign of Sultan Ibrahim (r. 1059–99), who resumed the dynasty’s earlier policy of raiding deep in the Gangetic plain, even capturing Agra in the late 1080s.⁶ But eighteen years after his death his dynasty’s old nemesis, the Seljuqs, struck from the west, sacking Ghazni in 1117 and reducing the Ghaznavid sultan, Bahram Shah (r. 1117–57), to a tributary vassal.+++(4 Mandala system here too!)+++ In 1135 the Seljuks struck again, forcing Bahram Shah to take temporary refuge in Lahore. The final blow to the Ghaznavids’ splendid capital came in 1150, when a deadly feud broke out between Bahram Shah and one of the Ghaznavids’ former vassals – the ruler of Ghur, in mountainous central Afghanistan – who burned the city to the ground. For seven days Ghazni was plundered, with a reported 60,000 slain and many splendid palaces, schools and mosques destroyed, justly giving the Ghurid chieftain the sobriquet Jahan-suz, ‘one who sets the world ablaze’. Also lost in that attack was the library of the great philosopher and polymath Ibn Sina (Avicenna, d. 1037), which had been brought to Ghazni from Isfahan in 1034.⁷ +++(5)+++ The city never regained its former glory. A severely weakened Ghaznavid state now fell back on Lahore, which became the dynasty’s capital until 1186, when it too was overrun by the rulers of Ghur.

Deprived of their direct ties with Central Asia – and with it their access to Turkish slaves, mercenaries and war-horses – the later Ghaznavids lost their wider, imperial vision and acquired the character of a regional, north Indian state. They were certainly not seen as menacing aliens who might have posed a civilizational threat to Indian culture. Contemporary Sanskrit inscriptions refer to the Ghaznavids not as Muslims but as turushkas (Turks), an ethnic term, or as hammiras, a Sanskritized rendering of amir (Arabic for commander), an official title.⁸ +++(4)+++ For their part, in the eleventh and twelfth centuries Ghaznavid rulers in India issued coins from Lahore bearing the same legends that had appeared on those of their Indian predecessors, the Hindu Shahi dynasty (c.850–1002). These included Śiva’s bull Nandi and the Sanskrit phrase śri samanta deva (Honourable Chief Commander) inscribed in Devanagari script.⁹ +++(4)+++ Such measures point to the later Ghaznavids’ investment in establishing cultural and monetary continuity with north Indian kingdoms. Moreover, despite the dynasty’s rhetoric about defending Sunni Islam, religion posed no bar to military recruitment, as Indians had always been prominent in Ghaznavid armies. In 1033 Mahmud of Ghazni gave the command of his army stationed in Lahore to a Hindu general, and in Ghazni itself Indian military contingents had their own commanders, inhabited their own quarter of the city, and were generally considered more reliable soldiers than the Turks.¹⁰ +++(4)+++

Crucially, the Ghaznavids brought to the Punjab the entire gamut of Persianate institutions and practices that would define the political economy of much of India for centuries to come. Inherited from the creative ferment of tenth-century Khurasan and Central Asia under the Samanid rulers of Bukhara, these included: the elaboration of a ranked and salaried bureaucracy tied to the state’s land revenue and military systems; the institution of elite, or military, slavery; an elaboration of the office of ‘sultan’; the courtly patronage of Persian arts, crafts and literature; and a tradition of spiritually powerful holy men, or Sufis, whose relations with royal power were ambivalent, to say the least.+++(5)+++

The first of these, the institution of a salaried bureaucracy, was based on the principle of state-run revenue assignments known as *iqta‘*s, which were defined units of land whose revenues were collected by the assignee, or iqta‘dar. From these revenues the iqta‘dar was required to recruit, train, equip and command a stipulated number of troops who would be available to the sultan on demand. The state’s revenue and military systems were thus tightly integrated. Iqta‘ lands were assigned to free nobles as well as to high-ranking slaves who enjoyed the special confidence of the sultan.+++(4)+++ Although the iqta‘ system had evolved in Iraq and western Iran in the ninth and tenth centuries, Ghaznavid rulers do not appear to have used it until the late eleventh century, when they could no longer rely on war booty to finance government operations, as they had done in the dynasty’s early days.¹¹

The institution of military slavery was also inherited from earlier practice in Iraq. From the ninth century, rulers in Baghdad had recruited Turks in Central Asia to serve the Abbasid caliphate (750–1258), entrusting them with both military and administrative responsibilities. As young men uprooted from their natal communities and recruited for service in a distant court, military slaves embodied a deep paradox. Having no traceable genealogy in a Persian-ordered universe where purity of blood translated into high status, they were lowly non-persons. But as well-trained elite soldiers given arms and close proximity to a ruling dynasty, they possessed power, wealth and the opportunity for advancement. For courts plagued with internal factionalism, it made strategic sense to stabilize central authority by recruiting powerful outsiders from Central Asia’s vast military labour market, and to place them under the tutelage of trusted state officers. These masters trained, equipped, fed and socialized their slave charges into a sultanate’s culture. As kinless aliens, they were rendered totally dependent on their masters, enhancing their presumed loyalty to the state. As the institution matured, ties of mutual trust and affection evolved; slaves close to a royal household were understood as fictive sons who might be praised even above biological sons for their loyalty and dedication.¹² For these reasons, from the tenth century to the fourteenth, thousands of Central Asian Turks were recruited to serve not only the Arab Abbasid dynasty in Baghdad but, in far greater numbers, native Persian dynasties such as the Samanids in Bukhara, and later the Ghaznavids and their client chieftains.

Overarching the entire political system loomed the sultan, an absolute sovereign whose political authority in the ninth and tenth centuries grew in tandem with the decline of that of the caliph in Baghdad. By the dawn of the eleventh century, sultanates evolved a de facto separation of religion and state. As the successor to the Prophet of Islam, the caliph still possessed religious authority, but sultans – or amirs or maliks, as they were called before 1002-30 – now held effective political authority.+++(4)+++ Moreover, since these political developments coincided with the revival of Persian culture in the Samanid court, Persian writers and theorists in Bukhara and elsewhere on the Iranian plateau endowed the office of sultan with the absolutist trappings of the ancient pre-Islamic Persian shahanshah, or ‘king of kings’, a title revived in western Iran by rulers of the Buyid dynasty (934–1062). In courtly discourse, the sultan was the state, while the world, as the political theorist Fakhr al-Din Razi (d. 1209) metaphorically put it, ‘is a garden, whose gardener is the state’.¹³

But the sultan was not the only claimant to worldly authority. Also emerging amidst a declining caliphal state in Baghdad and the florescence of Persian culture in Samanid Central Asia was a substantial body of literature that had grown up around vivid and charismatic spiritual personalities, Sufi shaikhs, who had played a key role in assimilating Turkish groups to Sunni Islam as the latter migrated through Central Asia into Khurasan and Iran. Sufis sought a direct experience of divine reality and postulated a vision of authority that sometimes complemented and sometimes opposed that of royal courts. Sultans certainly possessed power, reinforced by all the pomp and glory inherited from pre-Islamic Persian imperial traditions. But, in a discourse challenging such claims, Sufi texts suggested that rulers were entrusted with only a temporary lease of earthly authority, granted to them through the grace of some spiritually powerful shaikh.+++(4)+++ Possessing a special nearness to God, it was shaikhs, not princes or kings, who had the better claim to being God’s true representatives on earth.+++(5)+++ From this perspective, all things in God’s creation were understood as dependent on a hierarchy of spiritually powerful Sufis, or ‘God’s unruly friends’, as they have been characterized.¹⁴ Such a view, needless to say, was difficult to reconcile with the courtly vision of an all-powerful sultan and his flock of tax-paying subjects.

Whether arriving as invaders or immigrants, then, Persianized Turks brought to India two competing visions of legitimate authority and power:+++(4)+++ a Sufi discourse that circulated mainly among Muslims, and a courtly discourse that claimed validity across all communities. Both of them, however, sharply contrasted with India’s chessboard-world of the mandala and the digvijaya, which so preoccupied north India’s warrior clans as to blind them to the storm clouds that, by the end of the twelfth century, were gathering beyond the Khyber Pass.