Towards A Connected History of Equine Cultures in South Asia: Bahrī (Sea) Horses and "Horsemania" in Thirteenth-Century South India

Abstract

This article explores ways that the concept of equine cultures, developed thus far principally in European and/or early modern and colonial contexts, might translate to premodern South Asia. As a first contribution to a history of equine matters in South Asia, it focuses on the maritime circulation of horses from the Middle East to Peninsular India in the thirteenth century, examining the different ways that this phenomenon is recorded in textual and material sources and exploring their potential for writing a new, more connected history of South Asia and the Indian Ocean world.

THIS ARTICLE EXPLORES ways that the concept of equine cultures, developed thus far principally in European and/or early modern and colonial contexts, might translate to premodern South Asia. As a first contribution to a history of equine matters in this region, it focuses on the maritime circulation of horses from the Middle East to Peninsular India in the thirteenth century, examining how this phenomenon is recorded in textual and material sources and assessing their potential for writing a new, more connected history of South Asia and the Indian Ocean world.

In Noble Brutes: How Eastern Horses Transformed English Culture, Donna Landry remarks upon "the imbalance between the equestrian saturation of early modern culture and today's marginalization of matters equine."¹ To many in the "First World," the horse's overwhelming association with elite sport and leisure, from polo to pony clubs, has obscured its profound importance in the history of daily life, both into the very recent past and in many parts of the world today. For its meat, its milk, its use as a draft animal, as a mount, and as a luxury object, the horse has been essential to many cultures and subcultures, and without these diverse uses the horse would not have become a global animal. Landry's book has joined a small but growing number of studies redressing this imbalance through an imaginative combination of textual, visual, and material sources that convey the breadth and depth of this forgotten "equine saturation," as it is termed by Landry.² Equine cultures as they emerge through this scholarship are an entangled complex of animate and inanimate things decidedly unbounded by subject-object oppositions: from grooms, horse merchants, veterinarians, and those who rode or worked horses, to horse furnishings and stables, writings about horses (such as veterinary and military treatises), visual and sculptural representations, and of course horses themselves.

Where earlier scholarship approached horses as commodities in a larger economic history, as military technologies in a broader history of warfare, or again as veterinary subjects in a wider history of premodern medicine, the new focus on equine cultures places horses at the center and brings these different disciplines together to write deeper cultural and intellectual histories. As Landry's title suggests, "Eastern" horses were an important locus of encounter between the English and the Orient: that encounter produced the English thoroughbred and substantially changed English equestrian practices in the eighteenth century, but also impacted national identity. Not only were new riding styles seen to differentiate the English from their Continental European neighbors, the "free forward movement" experienced by the gentry during cross-country riding became associated with ideas of liberty, control, and adventure that were profoundly entangled with contemporary political thought and the British imperial enterprise.³ Horses, it transpires, are "good to think with." This approach is, of course, part also of a larger movement toward writing non-anthropocentric histories, histories that see things other than humans as equally valid subjects of inquiry.

Yet paradoxically, given the largely non-European origins of the modern horse (equus caballus), the new interdisciplinary approach to matters equine continues to marginalize the non-European and, in particular, the premodern. Notwithstanding the importance of "Eastern"—that is, Arabian, Persian, and North African—horses to the development of European thoroughbreds, the wider equine cultures of Eurasia and Africa remain sparsely represented in the new research, and there is little dialogue with existing, though often more disciplinarily bounded, literatures on horses in Eurasia and Africa.⁴ In the gap between these two scholarly tracks lie important questions about the premodern and non-European equine cultures that contributed to the "equestrian saturation" of the early modern period and that made horses a truly "global interest" even up until the mid-twentieth century.⁵

In premodern South Asia, the study of matters equine is still a marginal enterprise. Aside from a small body of specialist literature scattered across economic, military, religious, and art history,⁶ horses most often figure as "extras" in the background of other narratives. As a first contribution to a new history of equine matters in South Asia, this essay focuses on the importation of horses by sea from the Middle East to Peninsular India—the area comprising the modern-day states of Karnataka, Kerala, Andhra Pradesh, Telangana, and Tamil Nadu—in the thirteenth century. In Arabic and Persian sources, horses imported by sea, bahr in Arabic, were often referred to as bahrī, "from or of the sea," and this is the term I adopt here. Thirteenth-century sources describe a near-bankrupting consumption of expensive imported horses that I have chosen to call "horsemania," in reference to "tulipmania": the consumer passion for the newly-introduced Ottoman tulip that consumed seventeenth century Holland.⁷ If horses had been transported to South Asia by sea since at least the early centuries of the Common Era, the thirteenth century saw a boom in the volume of this trade and established patterns of circulation and consumption that continued into the early modern period and the Portuguese entry into this same trade.

Thus far, the boom in the maritime horse trade to South India in the thirteenth century has been written primarily as a commodity history or as part of Mongol history, since a significant part of this trade took place within larger networks of commercial and diplomatic exchange between Il Khanid Iran, the polities of South India, and Yuan China.⁸ This scholarship has largely been built around three contemporary, non-Indic sources: passages from the two Iranian geographical histories of the Il Khanid technocrat Hazrat-i Vassaf⁹ and the Il Khanid vizier Rashid al-Din,¹⁰ and the travel account of the Venetian merchant Marco Polo, itself heavily dependent on Il Khanid data.¹¹ There is no denying the critical importance of these three sources, but this essay proposes that they have perpetuated an Irano-centric history, overly focused on one South Indian dynasty, that of the Pandyas, which had particularly close political and economic ties with Il Khanid Iran; this narrative is thus little interested in the subsequent lives of bahrī horses in India and the Indic equine cultures they joined, or in the wider importation of horses to the region. This essay accordingly argues for a more complex, multi-directional, and interconnected history of the horse in thirteenth-century South India, built on a broader reading of existing Il Khanid sources and the introduction of new textual and visual material from both indigenous and external sources. The sources presented here have all been edited and translated and have already contributed to other histories—of commodities and tax regimes, military technologies, chivalric cultures, or animal husbandry—but they have not previously been combined in a history of equine cultures.

This essay proceeds in two steps. It begins by broadening and complicating the history of the horse as a maritime commodity through the integration into the existing narrative of data from Yemeni and Indic textual sources, as well as Indic visual sources, to give a more rounded picture of the volume, patterns, and politics of the horse trade to Peninsular India in the thirteenth century. Prominent among these sources is a collection of late thirteenth-century documents from the customs house of Aden, documents which were only published in 2003, and visual representations of horses and their equipment on Indian hero stones and in temple sculpture. The second part of this essay moves beyond this revised commodity history and into the realm of equine histories proper, by arguing that the facts and figures pertaining to the horse trade as given in the influential accounts of Vassaf, Rashid al-Din, and Marco Polo were never primarily "about" trade but were instead part of larger discourses surrounding civilizational differences and values, the encounter between the Middle East and South Asia, or (in Polo's case) Europe and non-European Others. The study of "equine cultures" rather than "horses" therefore underlines the extent to which, much as in eighteenth-century England, horses mediated cross-cultural encounters and were vehicles for larger discourses. Recognizing animals as a fundamental component of the material and intellectual exchanges that took place around the Indian Ocean opens new interpretive possibilities beyond their history as trade commodities. The second section also broaches the challenge of following bahrī horses into their new "Indian lives" and the problems that South Asian sources and South Asian historiography pose to this task. I explore the potential of inscriptional and iconographic evidence from the Hoysala and the Kakatiya polities for future research into the place of horses within military networks and their translation into distinctive Indic chivalric cultures (see Map 3). I conclude by tracing the ways that these newly forged Indic equine cultures made reverse journeys back across the western Indian Ocean, entering Middle Eastern hippiatric texts as a distinct body of Indian equine knowledge.

The Horse in South Asia

The ubiquity of the horse today tends to obscure its limited original natural habitat and the very active human interventions that made it a global animal. Domesticated and wild horses are now found on four of five continents, excluding only Antarctica, but at around 6000 BCE the wild horse populations that seeded the first domesticated horses were found only on the steppes of eastern Europe and central Asia. It is from this relatively circumscribed area that the modern horse has spread globally in just eight millennia,¹² a remarkably rapid and powerful expansion in which premodern non-European actors played a central role. Horses are indigenous, and best adapt to semi-arid zones or steppe climates, dry but not desertic conditions that tend to support a short or scrubby vegetation of grasses or shrubs.¹³ Only a limited number of regions across the globe are able to sustain wild horse populations, and the horse's wider global spread would have been impossible without human initiatives to export it beyond, and sustain it outside, its original natural habitat—in effect, to domesticate it.

Archaeological evidence points to the introduction of the modern horse to the north of the Indian subcontinent from at least the first millennium BCE,¹⁴ yet the

Map 3.

South India Showing Select Locations and Features. Map created by the author.

story of the horse in South Asia is far from being a neat narrative of gradual north-south diffusion. Finds of horse-bits in South Indian Megalithic (Iron Age) burials, broadly datable to around 800 to 300 BCE,¹⁵ as well as representations of horses and riders in Megalithic rock art, offer clear indications of the early presence and high status of these animals in the Peninsular south.¹⁶ Detailed understanding of the horse's diffusion across South Asia and its reception by, and translation into, the subcontinent's hugely varied cultures and environments remains a task hindered by a neglected base of textual sources, scattered iconographic evidence, and all too little secure archaeological data. This neglect is particularly puzzling given the seminal position of the horse within Hinduism and Indic kingship; the horse sacrifice or aśwamedha was one of the most important religious rites described in the Vedas, the foundational texts of Hinduism, and was performed by kings into the medieval period to celebrate their paramountcy. Nevertheless, even at this early stage it is clear that the story of the horse in South Asia is multi-directional and often paradoxical, profoundly intertwined with that of the elephant, India's royal mount par excellence, and particularly complex in Peninsular India, where maritime supply routes from the Middle East and Bengal made the horse an Indian Ocean commodity from an early period.

South Asia largely lacks the extensive grasslands needed to sustain stable populations of strong and healthy horses. A variety of sources—from the Kautilya Arthaśastra (a treatise on statecraft now agreed to date from the fourth century CE) to a fourteenth-century Sultanate history—affirm that the Punjab, Sind, and Saurashtra were the principal regions of the subcontinent with environments that could sustain stable horse populations and large-scale breeding.¹⁷ Not surprisingly, these are the very areas designated as semi-arid (BSk and BSh) in the Köppen-Geiger climate classification system.¹⁸ The climate beyond these areas was not suited to horse-keeping, fostering environments with little natural grass or fodder for equine diets while monsoonal humidity adversely affected horses' health, resulting in both a shortened lifespan and small, sickly offspring. Throughout the history of South Asia, any "indigenous" horse populations which became established were repeatedly supplemented by imported animals driven overland or shipped by sea. Horses driven overland from their native habitats in central Asia or the eastern Himalayas were referred to in medieval Arabic and Persian sources as barrī or "horses from the land," while (as noted above) those shipped by sea were bahrī (Arabic) or daryā'ī (Persian): literally "from the sea." It is not uncommon in medieval Indian texts to find horses discussed alongside pearls as one of the bounties of the ocean.¹⁹ Within this broader scenario, Peninsular India appears to have suffered a double disadvantage, in that its climate is especially challenging to horses, and it was often cut off from terrestrial supplies of barrī horses by polities in central India. South Indian polities therefore found themselves much more dependent on animals imported by sea. We have records of Persian horses reaching Sri Lanka by sea as early as the sixth century CE,²⁰ but some of the earliest evidence suggests that horses were also being shipped from the head of the Bay of Bengal to the Tamil south in the third century CE.²¹ Nevertheless, epigraphic and textual references to Arab or Persian horses, many of which must have arrived by sea, only begin to appear regularly from the tenth and eleventh centuries onwards.²²

Horses never entirely displaced the indigenous elephant as the choice mount of Indian rulers, and elephants remained crucial in South Asian warfare; but in spite of the cost and difficulty of maintaining horses in much of the subcontinent, it is generally agreed that cavalries had become a staple of Indian warfare by the early medieval period. South Asian rulers took titles that expressed their command of diverse armies: the king was Aśvapati or Advapati (Lord of Horses), Gajapati (Lord of Elephants), and Narapati (Lord of Men).²³ The Kautilya Arthaśastra (fourth century CE) already includes advice on the administration of royal stables and the care of horses,²⁴ and the first specifically hippiatric texts—Śalihotra's Aśvayurvedasiddhanta (Complete Ayurvedic System for Horses) and Nakula's Aśvacikitsita (Therapeutics of Horses)—are generally held to have been composed in the ninth and early eleventh centuries, respectively.²⁵ Yet in the popular reception of horses in Indian folklore, as Wendy Doniger has remarked, the horse was always perceived as non-Indian and ephemeral, the mount of conquerors from the Kushanas to the British, an animal that entered but could never truly establish itself in India.²⁶ It was always a little otherworldly, a semi-divine creature.

The Bahrī Horse Trade

Horse-Producing Regions and Supply Networks

The thirteenth century saw a major rise in the volume of the horse trade between the Arabian Peninsula, the Gulf, and South India. The fact that horses were circulated into and around India by sea has long fascinated Western observers, not least Marco Polo; however, the shipping of large animals is not surprising in itself. Horses and animals far larger and more dangerous, such as elephants, were regularly transported around and across the Indian Ocean, and had been since at least the early first millennium CE. What is surprising are the multiple, corroborating references to this trade in sources from around the Indian Ocean rim, from Syria and the Yemen to China.²⁷ For the first time, horses emerge as a major commodity of maritime trade, circulating in apparently very large numbers.

In the 1290s, the Il Khanid historian and erstwhile fiscal administrator Vassaf al-Hazrat reports that 1,400 horses were due to be exported annually from the Gulf island of Kish (Qa'is) to Ma'bar, the Tamil region of South India then under the Pandya dynasty, together with as many as could be sourced from other Il Khanid tributaries elsewhere in the Gulf: areas such as Qatif, Lahsa (al-Ahsa), Bahrayn, Hormuz, and Qalhat. The price per head of horse was 220 dinars of red gold.²⁸ In his contemporary Ta'rikh al-hind (History of India), the Il Khanid vizier Rashid al-Din counts a total of ten thousands head of horse exported annually to Ma'bar and purchased at a cost of 2,200,000 dinars of red gold.²⁹ Marco Polo, whose journey to China via India is one of the best known accounts of the European mercantile encounter with Eurasia at this time, gives similar quantities, reporting that two thousand or more horses were purchased by each of the five Pandya kings each year, for a total of ten thousand horses.³⁰ Elsewhere in his account, Polo specifically mentions Qalhat in northern Oman as an export depot for good destriers (warhorses): "because know that from this country [...] large quantities of beautiful horses are taken to India, so many that you could hardly count them."³¹ The consistency of the figures across these two last sources is remarkable and suggests that Polo had access to informants with reliable knowledge of the Il Khanid economy. A surprising aspect of the Il Khanid trade in horses with South India, given the hugely variable traits of individual animals, was the operation of a fixed price per head of horse. For the Il Khanid trade, both Vassaf and Rashid al-Din give the rate at 220 dinars of red gold per horse,³² Polo gives 500 saje d'or, which he states are worth more than 100 silver marks (mars).³³ According to both Il Khanid sources, the price was set on the condition that every horse was paid for, whatever its condition upon arrival, whether or not it survived the journey.

In Vassaf's text and Polo's, the Pandya kingdom (in what is now southern Tamil Nadu) emerges as a key market. As Polo observed during the course of his return west via the Pandya ports in the early 1290s, "in this kingdom no horses are born, and therefore all the treasure from the income (rende) that they have each year, or the greater part, is consumed in buying horses."³⁴ The final destination of the majority of these horses was the Pandya cavalry; as Vassaf observed, these horses were ridden by the troops of the Indians (junūd-i hindū) and used for war and equestrian exercises (li-l-harb wa furusiyya al-mayd ān).³⁵ Of course, these figures might be disputed as conventional medieval hyperinflation, and Simon Digby has wisely cited the advice of an early thirteenth-century Sultanate author Fakhr-i Mudabbir, to the effect that "the king or the commander of an army may know the numbers of horse and foot, he must say two or three times this number" when boasting about his troops.³⁶ Nevertheless, Rashid al-Din, as vizier, was well positioned to obtain fiscal information; while Vassaf, as a fiscal administrator in Fars for the Il Khanids, could have accessed, and perhaps even compiled, such figures. Vassaf was, furthermore, personally acquainted with members of the Tibi family who dominated this Il Khanid trade from their island emporium of Kish.³⁷ Moreover, these figures are altogether modest when compared to the hundreds of thousands of horses reportedly mobilized by Delhi Sultanate armies,³⁸ and not unreasonable when we consider, for example, that Jamal al-Din Tibi was able to present 1,000 geldings to the Mongol Il Khan Gazan in 1296, along with Chinese and Indian gifts.³⁹ The figures may even turn out to be remarkably consistent with subsequent trends, since the Portuguese horse trader Nuniz reported, in the first half of the sixteenth century, that the ruler of the large South Indian polity of Vijayanagara imported thirteen thousand head of horse per annum via the port of Hormuz.⁴⁰ Environments can only sustain animal populations of a certain size, and it may be that the surplus available for export was only ever in the region of ten to thirteen thousand head of horse per annum, after internal Iranian demand for horses was met.

Nevertheless, these figures represent only a portion of the total number of horse exports to South India in this period: horses were bred elsewhere than in Iran and the Gulf, and were also exported to other South Indian polities besides the Pandya. Departing from the heavily Irano-centric foci of Vassaf and Rashid al-Din, Marco Polo reports that the Pandya port of Qa'il (modern Kayalpattanam, on the Coromandel coast) received ships "laden with merchandise and horses" coming from Hormuz and Kish but also Aden, then under the Rasulid dynasty (r. 1229–1454), and "all of Arabia."⁴¹ A remarkable corpus of Rasulid customs documents from the port of Aden and dating to the 1290s confirm this, with one document explicitly underlining the scale of Pandyan purchases. Referring to the inhabitants as the Sūliyān—literally "Cholas," since the Chola polity (r. ca. 850–1279) had until recently been the dominant power across the region—this document repeatedly notes this people's desire for horses: "the Sūliyān, they are the people who wish to buy horses," and again, "in all of India few desire to buy horses as much as the Sūliyān desire it."⁴² The triliteral root used in both cases, RGHB, can be translated neutrally as "to wish" or "to desire" but might also be translated more forcefully as "to crave" or "to covet." Whichever the translation, it is the particular purchasing enthusiasm of the Sūliyān that is underlined. One ultimate destination of these horses was undoubtedly the Pandya polity, as confirmed by Marco Polo. "All of Arabia" may also have included Syria, since a geographical treatise finished in 721/1321 by the Ayyubid prince and governor of Hamah Abu al-Fida' includes the fact that horses from various places were brought to the capital city of Ma'bar,⁴³ while an account of the plunder of the Pandya royal stables in 1311 (by Delhi Sultanate forces) reports that the five thousand horses found there included bahrī, that is, Yemeni and Shami or Syrian animals.⁴⁴

Importantly, the Rasulid customs documents also point to other horse-purchasing polities in South India. One document lists the prices that horses fetched at the northern Malabari ports of Mangalore and Barkur (Faknur in the Arabic), the principal ports of the small coastal polity of the Alupas in what is now south Karnataka. At Mangalore, a horse fetched 400 mīnī (equivalent to 200 Egyptian gold mithqāls) and, at Faknur, a healthy horse without blemishes was worth 420 mīnī.⁴⁵ As Eric Vallet has noted, these two ports were most likely the transit points for horses destined for the cavalries of the Hoysala kingdom based on the Deccan plateau to the east, with its capital at Dwarasamudra (located on the western edge of the Deccan plateau, near modern-day Halebid). The northern Malabari ports offered the Hoysalas the most direct access to goods coming from the Indian Ocean system. The same pattern of horse supply, sometimes via the same ports and passes through the Western Ghats, operated under the Hoysala's Vijayanagara successors and was eventually taken over by the Portuguese.⁴⁶

It is unclear what pricing system operated in the horse trade with Rasulid Yemen. The sources from Yemen are contradictory: some customs house documents describe an annual auction of horses at Aden and thus suggest that prices were set by market demand,⁴⁷ and the northern Malabari evidence suggests different prices at different ports, while another document fixes the price of a horse at 800 dinars, stating that "horses worth 2,000 dinars, 1,000 and 500, 600, 400, 300, 200, all are sold for 800 [dinars], which has become a customary rule."⁴⁸ Nevertheless, the precise mechanics of exchange for the Yemen are well understood thanks to Eric Vallet's work on the Rasulid customs documents. These indicate that the dynasty operated a highly centralized system, at least in theory: the kingdom's only sanctioned horsefair took place once a year in Aden during the month of August, a period known as mawsim al-khayl, the "horse season."⁴⁹ Those purchasing horses on behalf of Indian polities are described simply as nakhuda al-hind, the ship-owners and great merchants of India, and they settled their purchases and the taxes and dues, half in coin (using silver dirhams) and half in unsewn silk textiles (fuwwat harīr).⁵⁰ Vallet has also demonstrated how problematic the payment of such large sums of silver was for Indian buyers and has mapped the hugely complex chain of interregional exchanges, from China through to Egypt, which underlay these purchases.⁵¹ A comparable network analysis of the horses-for-gold system operating between Ilkhanid Iran and Pandya south India is long overdue.

The Rasulid documents do not record the numbers of horses exported annually to India, or the annual revenue from this trade, but they do specify where some of the animals shipped from Aden were raised. Notwithstanding the fame of the "Arabian" horse, only select regions in the Arabian Peninsula are suitable for horse-breeding, and one document suggests that Yemeni horses were mainly bred on the great plains around San'a and Damar to the north of Aden, and on the high coastal plateau of Hasi to its east.⁵² These regions are comparatively small compared to the semiarid regions found across Iran and Central Asia; so small, in fact, that they are not represented on current Köppen-Geiger maps of the Arabian Peninsula, and they cannot have sustained horse populations on the same scale.⁵³ Although the modern notion of "breed" does not translate back to these prescientific, premodern milieus, it is clear that these areas produced a variety of horses. Polo gives by far the most detailed account of horse exports from the Arabian Peninsula, noting that merchants from Aden carry to India "many beautiful Arab destriers [war-horses] of great value, from which the merchants turn a great profit."⁵⁴ From Shihr "merchants carry many good destriers and many good mixed-breeds [chevalz de II selles] to India, which are very expensive and very valuable."⁵⁵ Further north, at Dhofar, "they bring many good Arabian destriers from other countries," which profit merchants greatly.⁵⁶ It is unclear at present where the horses exported from Shihr and Dhofar were raised, but it seems clear that Rasulid Yemen would have been unable to supply Peninsular India on the same scale as Iran.

Unfortunately, Indic sources for Peninsular India are almost totally silent about the horse economy and seldom specify where horses originated, let alone in what numbers.⁵⁷ We are left with a string of general references to a contemporary trade in horses all along the western seaboard of the subcontinent, from Solanki Gujarat (where a Jain text describes the merchant minister Vastupala supervising the unloading of horses from ships at the port of Cambay)⁵⁸ to Sri Lanka (where a mid-thirteenth-century Chinese source, the Description of the Barbarous Peoples, mentions horses traded at the island's ports).⁵⁹ However, we should not forget to look east for evidence. Polities such as the Pandyas and their northerly neighbors the Kakatiyas controlled ports on the Coromandel coast that were more immediately tied into Bay of Bengal maritime networks. While horses from the Arabian Peninsula and Iran certainly reached Pandya ports via the maritime route, we should not discount the possibility that these polities also turned to other, closer horse-breeding areas. Important research by Ranabir Chakravarti and Bin Yang has suggested that horses from the eastern Himalayas and southwestern China entered the Bay of Bengal trading system.⁶⁰ The region of Karambattan in the eastern Himalayas (likely corresponding to modern-day Bhutan or the fringes of Tibet) raised small hardy horses, what we would now term "ponies," known simply as kohī (mountain) horses.⁶¹ The unique climate of Yunnan in southwestern China had allowed horse husbandry to develop there from as early as the sixth century BCE. Horses from both regions were led into northern India via Bengal and, according to Chakravarti, were often re-exported to the eastern Deccan and even onwards to eastern Chinese ports.⁶²

While the strongest evidence for this trade begins in the fifteenth century, there is scattered earlier evidence for similar patterns, notably for the thirteenth century, when Marco Polo again provides important testimony by observing that Amu (Yunnan) produced horses and oxen that were traded into India via Pagan and Bengal.⁶³ Although it is likely that the larger horses from the Middle East were considered better suited to warfare than Himalayan and Yunnanese "ponies," we cannot exclude the Bay of Bengal system as a source of horses for the Kakatiyas and even for the Pandyas to their south. When the armies of the Delhi Sultanate conquered the Kakatiya capital of Warangal in 1309, the Kakatiya ruler Rudradeva reportedly surrendered twenty thousand horses, a mix of kohī ponies from the Himalayas and bahrī horses, in this case almost certainly Arabian and Persian horses that had reached South Asia by sea.⁶⁴ A proportion of these animals would certainly have been acquired as booty in engagements with neighboring polities—this was always the fastest and cheapest way to obtain new horses, as the very surrender of these animals to the conquering forces of the Delhi Sultanate illustrates—but we should not dismiss the possibility that a proportion of these animals had also reached the Kakatiya polity as direct imports via the western and the eastern Indian Ocean.

As all these examples underline, the large number of horses needed to supply Peninsular India had to be sourced from a wide area and through networks running deep into the hinterlands of the actual shipping points. When Vassaf mentions horses from Qa'is and "the islands of Fars, Qatif, Lahsa [al-Ahsa], Bahrayn, Hormuz, Qalhat," these were simply assembly points, perhaps formal horse fairs, for animals that had been raised across the grasslands of Arabia and Iran. Indeed, "sea horses" first enter Polo's narrative in his description of the kingdom of Persia and the importance of Kish and Hormuz in the export of good Persian horses to India.⁶⁵

The Birth of the Bahrī Trade

Multiple sources, from Syria and the Yemen to China, point to the emergence of the horse as a major international commodity in Indian Ocean trade to Peninsular India by the later thirteenth century, an expenditure so important, according to some accounts, that it consumed the greater part of state revenues. While this trade certainly did not start ex nihilo, fixing a more precise "when" and "why" to this horsemania is by no means easy. Premodern sources for the Middle East and South Asia usually yield quantitative data only reluctantly—hence the attraction of the much repeated Il Khanid figures—but long-term fluctuations in the trade of a particular commodity are even more difficult to capture. Of all the sources noted above, only Vassaf gives some historical perspective on the horse trade from Fars and the Gulf, based on his knowledge of trade under the Salghurid Atabek of Fars, Abu Bakr (r. 1226–60), who was, for the latter part of his reign, a Mongol tributary.⁶⁶ Vassaf notes that according to "authentic writers," ten thousand horses a year were already being exported to Ma'bar as well as to Kambayat (Cambay, the principal port of the Solanki kingdom) and other areas of India at this time for the same price of 220 dinars and under the same terms.⁶⁷ This is the clearest suggestion that the Tibi merchant princes of Kish may have taken over an existing trade agreement. If Vassaf is correct in this, this large-scale trade was already several generations old by the time he described it in the early fourteenth century. The fact that the various polities of Fars were never conquered by the Mongols but entered tributary relationships with them would certainly have encouraged continuity in local economic patterns.

If a large-scale maritime trade in horses does go back to the mid- or even early thirteenth century, what had perhaps changed by the end of that century, at least with regards to the Gulf, was the distribution network. The mid-thirteenth-century Gulf trade, as described by Vassaf, supplied ten thousand horses per annum to a variety of locations in India, including Gujarat; however, by the early fourteenth century Vassaf and Rashid al-Din appear to describe a trade of around the same volume but this time apparently focused exclusively on supplying the Pandya polity. If such a southwards shift is confirmed, it may reflect the growth of the maritime trade route from the 1270s onwards due to Chagadai disruptions in central Asia, increased contact between Yuan China and Il Khanid Iran via this maritime route, and South India's growing importance as a participant in these networks.⁶⁸ Although the Mongol conquest of Iran eventually encouraged and facilitated the export of horses to South India by sea, that does not explain in itself the rise in demand. A number of historians, from Simon Digby to Ralph Kauz, have suggested a clear correlation between the general growth of the bahrī horse trade in the thirteenth century and the challenge posed by the large cavalries of the Delhi Sultanate, which led to a "general structural militarization of the Indian subcontinent."⁶⁹ Bahrī horses, it seems, were part of a wider South Asian "arms race" precipitated by the Ghurid conquest of northern India in the late twelfth century. Digby carried out probably the most thorough assessment of military technologies and supplies for late twelfth-century India and concludes that there existed a general parity in military technologies between the Delhi Sultanate forces and their Indian opponents; for Digby, the sole factor that explains the success of the Delhi Sultanate's forces is the larger numbers of mounted troops, particularly archers, they could muster.⁷⁰ Digby's thesis is strongly supported by evidence from a variety of South Asian sources and would explain why horses became the single most important military import for other South Asian polities, notably those of the Peninsular south, which were disadvantaged by both climate and distance from horse producing areas.

Figure 5.

View of the Lower Level of the Kesava Temple at Somanathapur in Karnataka, Hoysala, ca. 1268 CE. In a distinctive Hoysala iconographic innovation, a frieze of mounted horsemen has been added to the more traditional repertory of elephant, goose, and makara friezes.

Photograph reproduced courtesy of the American Institute of Indian Studies, Accession No 15366.

Nevertheless, horses alone do not win battles: they are surrounded by larger assemblages of people and things, and it is this complex of factors that makes winning cavalry units. There is much we still do not know about the chronology of militarization in the Peninsular south and the place of horses within this. Indic sources will have an important role to play in clarifying this even if, at present, their use poses many challenges. Epigraphy, the mainstay of so much premodern South Asian history, has not yet been combed systematically for references to horses in the Peninsular south, while visual evidence is unevenly documented: some sites have been meticulously photographed and recorded, others are known only through brief and often passing textual references, and nearly all are poorly studied. In the future, systematic surveys of both types of evidence may be able to uncover patterns that substantially clarify when and where in Peninsular India bahrī horses first became a major imported commodity, and when new equine technologies and perhaps skilled humans accompanied them. For the moment, the evidence we have is suggestive rather than conclusive.

Perhaps some of the strongest evidence for understanding changes to the flow of horses in the thirteenth century is iconographic: Hoysala temple sculpture offers rich material. Friezes of prancing horses or mounted combat appear suddenly on the basement walls of Hoysala temples from the early twelfth century onwards, where they are integrated with a more traditional repertory of elephant, lion, or makara friezes (Figure 5). Horse friezes were an iconographic innovation unique to Hoysala sculptors, a pointer to the high status of the horse and cavalry within that society. However, the first half of the thirteenth century stands out as a particularly intense period of horse representation; of the twelve temples with equine friezes studied by the military historian Jean Deloche, ten date between 1219 and 1268 CE,⁷¹ suggesting that horses had become, or were becoming, far more prominent and valued in Hoysala society at this period. Within this timeframe, the middle of the century stands out as particularly significant. As Deloche demonstrates, Hoysala sculpture began to represent a number of distinctively Turkic-Islamic equine technologies, such as stirrups, horseshoes, horse armor, and (most importantly) a new type of saddle with pommel and cantle that held the rider more securely than previous designs. The historian Philip Wagoner has suggested that these representations show "the first step in the adaptation of Turkic-Islamicate military technologies by local Indic military elites," resulting from the Turkic conquest of northern India.⁷² Yet the Hoysala had not yet engaged the Delhi Sultanate in battle directly at this period, suggesting different mechanisms of exchange. In many scholarly discussions, direct military engagement is understood to be a necessary prequel to the transfer of military technologies, via captured horses, horse equipment, or fighting men; and yet in the Peninsular south these changes appear to have occurred prior to direct military engagement.

Figure 6.

Hero Stone Showing a Single Mounted Horseman in Battle, from Tripuran-takam, Kurnool District, Andhra Pradesh (Kakatiya Period, thirteenth Century CE). This imagery is more commonly seen in western India and indicates the distinctive regional equine culture of the Andhra region at this period.

Photograph reproduced with kind permission of Professor R. Chandrasekhara Reddy.

It is important to bear in mind that incidental evidence for economic and technological phenomena such as this can never be entirely reliable. Visual representations follow their own conventions, fashions, and timescales: they do not represent the "real" world, let alone aim to supply economic historians with usable quantitative data, and it is impossible to gauge at present the speed at which Hoysala sculptors reacted to social and technological changes.⁷³ The introduction of a new equine iconography and its increased use arguably tell us more about the changing status of the horse in Hoysala society or its changed visibility than about the numbers of animals imported or about the timing of technological transfers. Deloche also did not systematically examine evidence from Hoysala hero stones, which are clearly iconographically related to temple friezes, and such a study might help refine or revise his thesis (for a beautifully preserved Kakativa example, see Figure 6).⁷⁴ Nevertheless, if this interpretation of the Hoysala evidence is correct, it suggests a rise in the presence and status of the horse during the first half of the thirteenth century with increased adoption of certain Turkic-Islamic military technologies from the middle of the century. This timeframe correlates broadly with Vassaf's information about a large-scale horse trade from the Gulf to South Asia under Atabek Abu Bakr.

The combined evidence further suggests that knowledge of military technologies circulated within South Asia and between South Asia and surrounding regions before any military confrontation, indicating a phenomenon of preemptive rather than reactive technological change. Wagoner's work on the fifteenth-century Deccan has certainly underlined the mobility of military elites, notably Turkic Muslim nobles, across political and religious frontiers, and it is perhaps thanks to similar mobilities that Turkic-Islamic equine technologies traveled south through the Peninsula.⁷⁵ We should also not forget that a wider Turkic-Islamic equine culture had established itself across much of the central and eastern Islamic world since the first large-scale use of mounted Turkish slave soldiers by the Abbasid Caliphs in the ninth century. Subsequent waves of Turkic invasion consolidated the Turkic contribution to Islamic culture. In the thirteenth century, the "new" horse furnishings represented in Hoysala sculpture were as common in Iran and the Arabian Peninsula as in the Delhi Sultanate. It is entirely possible, therefore, that these technologies were also transmitted to the Hoysala through the trade in bahrī horses from the Gulf or Yemen. Indian nākhūdas at Aden's annual horse fair certainly came into direct contact with Turkic-Islamic equine cultures. The diplomatic exchanges that accompanied the wider Indian Ocean trade may also have assisted in the circulation of equine technologies, as it was common to send fully caparisoned horses as diplomatic gifts. Although we have as yet no concrete examples of gifted horses from thirteenth-century Middle Eastern or Indic sources, the diplomatic activity of the period is well attested, and such a scenario is entirely possible.

This is fragile data, but it points a way forward. Nevertheless, it remains important to calibrate both epigraphic and visual evidence carefully, and to be aware of their fundamentally recalcitrant natures. One of the most paradoxical aspects of horsemania in thirteenth-century South India remains the fact that, in spite of the Pandyas' clear preeminence as importers of bahrī horses according to Yemeni, Iranian, and European sources, neither epigraphic nor visual sources hint at that dominance. References to horses as commodities and taxable items appear to be almost nonexistent in Pandya inscriptions: the historian Nilakantha Sastri is able to signal only one reference to a horse trader from southern Kerala in his history of the polity.⁷⁶ Nor did horses enter Pandya visual culture. They are not represented in temple sculpture or on hero stones, the memorials carved and often inscribed to commemorate dead warriors. Without the Middle Eastern sources and Marco Polo, there would be little trace of the Pandya consumption of bahrī horses.

Interconnected Equine Cultures

Equine Knowledges and Civilizational Encounters

The facts and figures on the bahrī horse trade provided by Vassaf, Polo, and others are rare for the thirteenth century, and it is perhaps no surprise that historians have clung to them. In so doing, however, they have largely ignored the discursive contexts in which these facts are situated. Restoring this data to its original context offers a remarkably effective means of using these sources as part of a broader history of equine cultures and illustrates the extent to which, well before the eighteenth century, horses were vehicles for other ideas. In the case of both Vassaf and Marco Polo, the fate of horses in South India offers each the opportunity to develop complex and contradictory discourses that attribute to horses an almost emblematic civilizational role.

In Polo's narrative, observations about horses are woven into a larger narrative fabric intent on pointing to Oriental, sometimes specifically Muslim, excess and ignorance. Although horses feature only briefly in his description of the main shipment point of Hormuz, the context in which they feature is highly significant: horses are included in a longer, disparaging description of the coir-sewn boats so typical of the western Indian Ocean. Because "they," the Muslim merchants or perhaps the Indian shipmasters, do not have iron nails, the boats are sewn, and they are "very bad and many founder"; the loaded merchandise is simply covered with hides and onto this "they put the horses which they carry to India to sell."⁷⁷ According to Polo, Indian Ocean boat-building techniques are primitive, and understanding of lading methods is similarly backward. This passage sets the tone for the longest discussion of the bahrī trade, embedded in Polo's description of the kingdom of the Pandyas. Having told his readers that this kingdom consumes most of its revenues on buying horses, and that these are good, valuable animals, he continues:

they all die, because they [the Indians] do not have ostlers [marreschaus],⁷⁸ nor know how to care for them, but they die though neglect. And I tell you that the merchants who carry these horses for sale do not allow to go, nor send, any ostlers, because they [the merchants] want these kings' horses to die.⁷⁹

Indian ignorance of horse care is here compounded by the exploitative practices of merchants from Muslim lands who deliberately impede knowledge circulation in order to ensure the early deaths of these animals. Eastern horses were rare luxuries in Europe, even into the sixteenth century, and to squander rare animals on such a scale—furthermore, transporting them in primitive ships—must have been shocking to Polo's readers, perhaps to Polo himself. Polo's narrative anticipates many themes familiar from later Orientalist discourse, and this very familiarity has perhaps aided an uncritical acceptance in later Western scholarship of the idea that the bahrī horse trade was impractical and even foolhardy, and that Indians did not know how to care for horses.

The Persian chronicler Vassaf offers a very different perspective on the problem of horse mortality in South India and one that, again, must be situated within its original context. Vassaf's focus is not the bahrī horse trade per se; rather, his discussion centers on the Tibi merchant family and is itself situated within a highly florid account of the "marvels and excellences" of the country of Hind—a discourse that betrays a clear lineage in the well-established tradition of the 'ajā'ib, the "wonders," of non-Islamic lands.⁸⁰ With the stock phrase "it is a strange thing," he turns to peculiarly Indian practices of horse management, depicting what Finbarr Barry Flood has described elsewhere as a "tospy-turvy" Indian world where horses are neither fed, exercised, or ridden according to expected Middle Eastern practices.⁸¹ The result is that "in a short time the most strong, swift, fresh, and active horses become weak, slow, useless, and stupid."⁸² Yet whereas Polo understands this to be a symptom of a wider Oriental backwardness, for Vassaf the fate of horses is ultimately explained as part of the divine natural order: it is predominantly "this climate"—literally, "water and air" (in āb va havā)—which renders horses "exceedingly weak and altogether worn out and unfit for riding,"⁸³ and so explains the constant demand in South Asia. Vassaf concludes that

their loss is not without its attendant advantages, for it is a providential ordinance of God that the western [world] should continue to want eastern products, and the eastern world western products, and that the north should with labor procure the goods of the south, and the south be furnished in like manner with commodities brought in ships from the north.⁸⁴

Horses are one of the commodities that mediate communication between the quarters of the earth, and the peculiar practices of their trade and "consumption" in south India are part of the "marvels and excellences" of the country of Hind.

In both Polo and Vassaf, then, and yet in radically different ways, horses are vehicles for larger discourses about civilizational differences and values: about the encounter between Europe, the Middle East, and South Asia; and between Hindu, Muslim and Christian cultures.

Finding "Matters Equine" in South Indian Sources

The seemingly obvious military destination of bahrī horses and their apparently brief Indian lives—mentioned by Polo, Vassaf, Rashid al-Din, and again in a later Yemeni treatise⁸⁵—have hitherto limited the discussion of the reception and translation of horses into Peninsular India. Yet archaeological and visual evidence going back to the Megalithic period in South India indicates that, in the thirteenth century, these animals supplied already ancient regional equine cultures, even if the precise features and chronologies of these await further study. The problems that Indic sources pose for writing commodity histories of horses are understandably compounded in the attempt to move beyond this to broader equine histories. As noted above, epigraphic sources have not been systematically explored across the Peninsula, and little is known about the local circulation and eventual translation of the classic Indian hippiatric treatises of Śalihotra and Nakula, while literary sources that might write the Indian lives of bahrī horses have similarly not been systematically sifted for equine references. Such a project would of necessity be collaborative, demanding at the very least expertise in Sanskrit, Kannada, Malayalam, Tamil, and Telugu.

One example from the Tamil literature illustrates the potential of such material. A thirteenth-century narrative of the life of the ninth-century Tamil saint Manikkavacakar, the Tiruvilaiyatal puranam (The Divine Plays of Lord Shiva), sets the story of his entry into the religious life in a markedly contemporary, thirteenth-century context.⁸⁶ Manikkavacakar was chief minister to the Pandya monarch and as such was sent to a port to purchase horses for the royal stables. En route, the future saint encountered Śiva in the form of a guru and decided to devote himself to the Śaiva cause. Returning to court without money or horses, Manikkavacakar was duly punished. Later, however, Śiva arrived at court "in the guise of a foreign horseman leading a herd of fine horses to the Pantiyan's stable."⁸⁷ The "horses," however, were really wild jackals, and later they reverted to their original form and caused havoc within the palace. The story offers a subaltern view of both the wastefulness of the Pandya ruler's spending on horses and the foreign dominance of that trade, but it also reprises the idea that horses were alien, terrifying creatures. Jackals in India are closely associated with death because of their scavenging of corpses.

Material sources are even more patchily documented and studied than such informative anecdotes. Surveys and excavations of elite architectural complexes and urban spaces have rarely looked for evidence of the material culture of horse riding and horse keeping (such as stables and training or parade areas). Also, because monsoon climate conditions complicate the survival of all but the most stable materials (such as stone and ceramics), faunal remains are rarely collected and researched during archaeological excavations. Compounding this problem, archaeologists have displayed a clear bias in favor of the Megalithic and Early Historic periods. Yet the work carried out at a regional or dynastic level provides models that might be transferable to the wider canvas of the Peninsular south. Although Deloche's work has focused on using visual sources to index military-technological change, it has also traced the development of a unique visual culture of equine representation among Hoysala craftsmen and, we must presume, their patrons. Further work on Hoysala hero stones and on equine references in Hoysala epigraphy would not only refine Deloche's principal thesis but provide a working model for transfer to other regions and polities—in the process revealing regional interactions, convergences, and divergences. Even more significant is the work of historian Cynthia Talbot on the Kakatiya polity, its military networks, and what she terms the "martial ethos" in medieval South Indian society. Together with the work of a small number of other historians, this work challenges prevailing models of medieval Indic kingship by emphasizing the place of military action and training in South Asian society.⁸⁸ For Talbot, the models of so-called "dharmic kingship" favored by Western scholars fail to take account of the truism that "political power has a physical basis in armed might."⁸⁹ The central importance of cavalry units in the Kakatiya army and the elite status of those warriors inevitably makes horses a recurring leitmotif of Talbot's study. Although she works solely from Kakatiya epigraphic sources and does not engage with the rich body of equine representations on contemporary Kakatiya hero stones (Figure 6), Talbot provides valuable new models for the writing of broader equine histories across the Peninsular south.

Horses, Horse Merchants, and the "Web of Military Associations"

Talbot's research has underlined the importance of thinking of polities as military networks:

because a king's own martial skills and his ability to recruit and retain the loyalties of other fighting men were essential to his survival, the web of military associations underlying every state system should be prominent in our models of [...] medieval polity.⁹⁰

Her analysis focuses on the personal networks forged among the Kakatiya ruler, regional governors, local chiefs, and their supporters, but it is clear that other human networks supported the maintenance of power, notably through access to essential military supplies. If understanding of the place of horses and horse merchants in the Kakatiya's "web of military associations" must await further study, Talbot's model translates well to the more abundant data we have for some of their contemporaries. The Indian nākhūdas who operated as middlemen in the trade between Rasulid Yemen and the Hoysalas can be seen as fundamental participants in the Hoysala's own "web of military associations" through their participation in the purchase and transport of military supplies, notably horses. South Asian models of merchant activity (like models of kingship) have tended to neglect the prominence of armed conflict in merchant lives. The highly competitive and often violent character of Indian Ocean trade meant that ships regularly carried armed men or traveled in convoys with armed protection, while elite merchants also became courtiers and frequently fought with, or led, fighting forces—so it is entirely possible that some ship owners and merchants were also "fighting men."⁹¹

We have much more precise information about the place of horse suppliers within the Pandyas' "web of military associations." According to Vassaf, one family of merchant-princes, the Tibi family of Kish, controlled both the export of horses from Il Khanid tributaries in the Gulf and their importation to the Pandya kingdom after 1292. In Iran, the Tibis paid tribute to the Il Khanids and bought the right to tax farm the province of Fars and its coastal areas more or less autonomously; later in that decade, the area under their control was extended to the head of the Gulf and the port of Basra.⁹² Like many Iranian merchants under Il Khanid, rule the Tibis belonged to Mongol mercantile associations known as urtaqs which traded preferentially with and on behalf of the Mongol elite. What appears to have been unique about the Tibis—and perhaps explains Vassaf's inclusion of their family history within his "wonders of India" discourse—is the fact that they controlled both ends of the network and derived enormous wealth from it. In India, members of the family occupied key administrative roles within the Pandya polity. Vassaf records that Jamal al-Din's brother, Taqi al-Din 'Abd al-Rahman ibn Muhammad al-Tibi, was deputy (nā'ib), minister (vazīr), adviser (mushīr), and administrator (sāhib tadbīr) to the Devar Sundar Bandi, likely the Pandya subregent Jatavarman Sundara Pandya III (r. 1276–92/93). In addition to this, the three principal ports of Fatan, Mali Fatan, and Qa'il "were made over to his [Taqi al-Din's] possession,"⁹³ and Vassaf reports that in 1292 the agents and factors of the Tibi brothers had the first choice of any merchandise arriving at Pandya ports.⁹⁴ Vassaf's attribution of the ancient Sasanian title Marzubān al-hind or "Warden of the Marches of India" to Taqi al-Din reinforces the extent to which Il Khanid Iran regarded the Pandyan ports as an extension of its maritime frontier. If horses from other horse-breeding areas such as the Yemen were allowed to enter the Pandya market, it was patently only with Taqi al-Din's permission. A list of gifts to influential figures at Indian ports compiled at the port of Aden includes Taqi al-Din's name, and Eric Vallet has noted that such gifting was probably intended to smooth Rasulid access to the Pandya market.⁹⁵ The Tibis' influence was sufficiently embedded in India to survive and even flourish after the death of Jatavarman Sundar Pandya in 692 AH/1292–93 CE, and, according to Vassaf, Malik Taqi al-Din "continued as prime minister, and, in fact, ruler of that kingdom, and his glory and magnificence were raised a thousand times higher."⁹⁶

The appointment of Middle Eastern merchants to administrative positions in Indian ports was already a well-established practice in South Asia, but possession of equine knowledge may have become an increasingly important prerequisite for appointment. Horse merchants certainly feature prominently in the later political history of the Bahmani and Adil Shahi Sultanates in the fifteenth century, and in Mughal India in the seventeenth century; the example of the Tibis suggests that this may have been a much older practice and one that may have been particularly important in Peninsular India, away from terrestrial supply routes of barrī horses.⁹⁷ The Tibis continued as key powerbrokers in South India right up until the conquest of the Pandya kingdom by the Delhi Sultanate in 1311.⁹⁸ The considerable political and economic influence allowed to the Tibi family is evidence for the Pandya polity's ability to plan militarily on an international scale. This was not the exploitative relationship that Polo and even to some degree Vassaf suggest, but a symbiotic alliance that integrated Il Khanid merchants into the heart of the Pandya polity's "web of military associations," to the benefit of both parties. Horses were physical manifestations of the Pandyas' international "web of military associations." After centuries of importing horses, South Asian military and political elites would have been well aware of regional variations, and, in the Pandya kingdom, horses of different origins may have been recognized and understood within this framework of meaning as physical evidence of the king's ability to command resources internationally.

It remains to be seen how far this Pandya network of horse supply differed from that of their Chola predecessors and indeed from those of their Hoysala and Kakatiya contemporaries. The absence of horses from Pandya inscriptions contrasts with frequent references in earlier Chola donative inscriptions to horse merchants from Malaimandalam, present-day Kerala, and with references in contemporary Hoysala inscriptions to horses as traded items.⁹⁹ While further analysis is needed, this pattern appears to suggest that the Tibis did indeed participate in a monopoly that largely bypassed earlier west-coast intermediaries.

Horses and the "Martial Ethos" in Medieval South India

Another route into the South Indian reception and translation of bahrī horses is suggested by Cynthia Talbot's work on the "martial ethos" of the Kakatiya dynasty's court and army. In keeping with the importance of military action to the legitimacy of kingship and a wider military society, Talbot is able to demonstrate that military action is also a major, if neglected, subject in the epigraphic repertoire. As she observes, "fierce fighting between mounted horsemen is the main theme, for elite warriors were inevitably found in the cavalry contingents."¹⁰⁰ Her analysis of the Kakatiya "martial ethos" as it emerges from the epigraphic record currently offers our best insight into how horses were received into medieval South India, and does so in ways that suggest fascinating opportunities for the comparative study of this South Asian martial ethos and contemporary European chivalric cultures. The inscriptions Talbot analyzes are extremely varied in imagery and content, and here I discuss three examples, each one illustrating a different scale at which we might locate the translation of horses into Indic military culture.

In large numbers, at the scale of a panoramic battle scene, cavalry adds to the terrifying sound and sight (drums and banners) of the advancing army, as the clouds of dust raised by the galloping horsemen obliterate the sun in an apocalyptic fashion. It is worth quoting at length a passage from an inscription of the Kakatiya ruler Ganapati (r. 1199–1260):

When the thundering of the war-drums of his victorious army on the march pervaded the far corners, it was as if the echoes reverberating off the towering houses of his enemies were telling them, "Escape to the forest quickly, for King Ganapati, a master in the battlefield, is approaching!" Held up high on tall poles and waving vigorously in the wind, his army's battle colors seemed to signal to the many rival kings from a distance with the threat, "Run far away at once!" When the rays of the sun's light had been totally extinguished by the clouds of dust that rose up from the ground as the rows of sharp hooves of his throngs of horses tore it asunder, the astonished people thought the sun had gone away, observing the frightful heads of the hostile kings rise up [in the air] as they were cut off by his weapons and mistaking them for an army of Rahus.¹⁰¹

In the closer focus of battle engagement, it is the speed of the horse that is foregrounded and explicitly contrasted to that of elephants. An inscription celebrating a chief from the Telangana area of eastern India (r. 1262–89) describes an engagement featuring:

warriors [as savage as] lions and tigers, whose horses became [as swift as] thoughts and winds, and whose elephant-troops became [as towering as] mountains and clouds when Rudradeva's army exhibited its fearsomeness in the extermination, dispersal, or capturing of [enemy] kings without effort.¹⁰²

Finally, at the most intimate scale, the generic horse becomes an individual horse, belonging to a named soldier and with its own name. A donative inscription dated 1235 records how "the king's cavalier Vishaveli Masake Sahini," in effect the Master of the Royal Stables¹⁰³ "gave 25 cows for a perpetual lamp to the glorious lord Tripurantaka, having been victorious in the battle of Chintalapundi mounted on [the horse] Punyamurti."¹⁰⁴ One does not name an animal one does not value: Punyamurti means "Embodiment of Merits" and clearly was.

If the first example seems to resonate with outsiders' views of horses as significant only in, or because of, their large numbers, the second points to at least one of the qualities for which horses were prized above elephants—namely, their speed—while the third reminds us of the close bonds that developed between elite riders and their mounts. Horses may have lived brief lives, but they were no less important to their riders; and the close association between cavalryman and mount evoked in the third inscription is often illustrated on memorial stones. Talbot does not indicate whether Vishaveli Masake's inscription was also carved, but Figure 6 shows a roughly contemporary memorial stone from the same site, Tripurantakam in Kurnool District, depicting a single-mounted rider viewed in profile with smaller foot soldiers around him. Single-mounted horsemen are more usually represented on hero stones from Rajasthan, Sindh, and Kutch, areas with indigenous horse breeds and strong local equine cultures; and yet, alone in the Peninsular south, Kakatiya Andhra Pradesh developed its own distinctive tradition of human and horse representation.¹⁰⁵ The uniqueness of the Kakatiya iconography raises many questions about the deep embededness of equine cultures in the region and the place of the horse in the Kakatiya martial ethos.

It remains to be seen whether Hoysala, Pandya, and other sources yield evidence for largely similar discourses and tropes, or whether the final picture will be of interrelated but nevertheless distinctive regional equine cultures, as distinctive perhaps as the visual representation of horses and riders discussed thus far. Only with a better documentation of visual and textual sources about horses across South India will we begin to capture the more subtle differences between these regional or perhaps dynastic equine cultures and, in so doing, follow the Indian lives of bahrī and other imported horses.

Conclusions

Indian Equine Knowledges in Circulation

By the thirteenth century, Peninsular India had developed distinctive regional equine cultures born from a millennium or more of horse keeping and horse importation. Though a great deal of research remains to be carried out in this area, it is clear that South Asian equine cultures were already strong enough to impact the Middle East, a circular movement of equine cultures too rarely noted in the dominant focus on their export from the Middle East to South Asia. Vassaf's less judgmental view of Indian horse-keeping skills resonates with evidence for well-developed equine knowledge in South Asia by the thirteenth century. In his account of India, the Damascene geographer Ibn Fadl Allah al-'Umari (d. 1348) reports that one of his informants, a Bahrayni horse trader, said that the "people of India have got a knack to distinguish the horses of good breed. As [sic] they find the mark in a horse known to them, they may purchase it at any price."¹⁰⁶ Indian veterinary skills were another facet of this equine knowledge recognized outside South Asia: an Arabic language veterinary treatise written around 1331 by the Rasulid ruler of the Yemen, al-Malik al-Mujahid (r. 1321–62), remarks on the expertise of Indian doctors in identifying early symptoms of a "horse plague" that affected animals across the Arabian Peninsula and at Indian ports in 728/1327.¹⁰⁷ These examples point to a far greater interaction between Middle Eastern and South Asian traditions of equine knowledge than either Vassaf's or Polo's accounts suggest, stimulated very directly by the horse trade across the western Indian Ocean and the further interactions this necessitated between specialists from both regions.

The inclusion of this anecdote in al-Malik al-Mujahid's treatise reminds us of the openness of the Arabic hippiatric tradition to other equine knowledges. Although South Asia was to become the major zone of encounter between Middle Eastern and South Asian hippiatric traditions from the later fourteenth century onwards,¹⁰⁸ horses also mediated this cultural encounter beyond South Asia, across the rest of the Islamic world.

Several of the major Arabic hippiatric texts of the twelfth through fourteenth centuries, composed from Spain to the Yemen, include chapters on the interpretation of the colors and markings of horses and attribute this knowledge directly to Indian authorities.¹⁰⁹ As Housni Shehada observes, "the idea that markings or patches on the horse's body indicate the good or bad luck that it may bring to its master is characteristic of the Indian tradition."¹¹⁰ It is exactly this importance of markings (marks) that al-'Umari highlights in his account of India.¹¹¹ Yet attempts to map a textual chain of translation and transmission from South Asia to Spain via Abbasid Baghdad have largely failed, as these sources acknowledge only generic "Indian sages" or Indian viziers who cannot be traced historically. This vagueness contrasts with the care usually paid to chains of textual and/or oral transmission in Islamic scholarship and highlights the extent to which the hippiatric textual tradition emerged relatively late in the Middle East, and often from direct personal experience in the craft of horse care.

The most likely vector for the transmission of this Indian knowledge to the Middle East, therefore, seems to have been not textual at all but oral, transmitted perhaps through the direct interactions of horse merchants, grooms, and buyers at horse fairs in the Gulf and Arabian ports, or indeed in India itself. An understanding of South Asian equine cultures was essential for any merchant or court that maintained relations with South Asia, as Ibn Fadl Allah al-'Umari's Bahrayni horse-trader informant indicated, because in India horses were judged first on their color and the auspiciousness of their markings. These Arabic chapters on the Indian science of horse physiognomy, along with al-Malik al-Mujahid's account of the horse plague, can thus be read as evidence for the circulation between South Asia and the Middle East of a very practical equine knowledge. But we might also reflect that, when they were read in places far away from this trade—for example, in Islamic Spain—such chapters also perpetuated an older Middle Eastern discourse about the wonders of the non-Islamic world. Through the vehicle of horses, India emerges once more as a "topsy-turvy" world where Middle Eastern taxonomies and practices are upended.

An Agenda for "Matters Equine" in Medieval South Asia

As Wendy Doniger has underlined, the horse was never perceived as an Indian animal in indigenous folklore, and even the boom in the bahrī horse trade in the thirteenth century may have had little impact on domesticating the horse in the Peninsular south, at least beyond the courts of elites. Perhaps most telling is the manner in which command over horses became intrinsic to certain South Indian conceptions of the geo-political universe, and yet how little horses were ever identified as a South Indian animal. As Cynthia Talbot has demonstrated, from at least the early fifteenth century, in parts of South India, the idea developed that the subcontinent was in fact "divided into three realms, each ruled by a king laying claim to superiority in one contingent of an army."¹¹² The pattern that emerges over the centuries is that "the Lord of Horses was a designation that could signify any Muslim king," but apparently only a Muslim king.¹¹³ Non-Muslim kings, by contrast, were lords of elephants, the quintessential royal mount of Indian kings and Indian war animal, or they were lords of men. Horse representations never became commonplace in the south of the Tamil country, and, as the art historian Crispin Branfoot has shown, it is only in the sixteenth century (and largely in the Tamil north, on the border with Andhra Pradesh) that horses enter the sculptural repertoire.¹¹⁴ Even then, the horse was somehow always a little fantastical and certainly a little terrifying: large rearing brutes with diminutive riders and straining grooms join older iconographies of rearing griffins and other mythical animals in a new type of temple support known as the "mounted cavalryman" composite column (see Figure 7). As many outsiders remarked of South India, there are no horses native to that place, and the horse never became an indigenous animal: and yet the complex history of its reception and translation to South India magnifies the rich geographical and cultural diversity of this area and the ways in which horses became vehicles for the expression of other ideas.

The thirteenth century was not only a turning point in the diffusion of Middle Eastern horses to South Asia and beyond: by 1329, in northern Italy, the Gonzagas were cross-breeding Arab horses with local breeds, marking a symmetrical European fascination with "eastern" horses that later exploded, as Lisa Jardine and Jerry Brotton have shown, during the European "Renaissance".¹¹⁵ As Marco Polo's account shows, horses were already a truly "global interest" in the late thirteenth century, and had been for some time. However, this case study has demonstrated that the ubiquity of the modern horse at a global level hides a complex history of diffusion, and one that is substantially unwritten for much of this period across much of the globe. Moreover, at least for medieval South India, this is still far from being a history that is possible to write: a huge volume of primary source material exists in old, untranslated editions; archaeological excavations have been slow to collect faunal remains, let alone analyze them; and much architectural and iconographic evidence is inconsistently documented and published.

This situation is unlikely to change very quickly: there are few non-European medievalists compared to those specializing in Europe, fewer still being trained, and as long as this disproportion remains there will be tantalizing histories that can only be half told. Perhaps the frustration of not knowing how to tell these stories may spur more medievalists to study non-European history, or perhaps European medievalists will begin to see the possibilities for comparative and connected histories that these wider, global perspectives open. The medieval globe was always profoundly interconnected through the circulation of people, animals, things, and ideas, but its history can only be traced when seen from multiple perspectives, through both textual and material sources, and through sources from many regions in many languages and media.

Figure 7.

Rearing Horse Depicted on Composite Columns at the Sesaraya Mandapa at Srirangam, Tiruchchirapalli in Tamil Nadu (vijayanagara Period, Mid-Sixteenth Century CE).

Photograph reproduced courtesy of the American Institute of Indian Studies, Accession No 101330.