This is the html version of the file
Google automatically generates html versions of documents as we crawl the web.
To link to or bookmark this page, use the following url:
(downloaded Sept. 2002)


Pillars, Palimpsests, and Princely Practices:
Translating the Past in Sultanate Delhi

Prof. Finbarr B. Flood
Dept. of Fine Arts, New York University

In the traditional master narratives of South Asian history, the conquest of northern India by the army of the Ghurid sultan Ghiyath al-Din Muhammad (r. 1163-1203) in the last quarter of the twelfth century figures a profound rupture in the cultural fabric of the region. Many of the monuments erected in the wake of the conquest appear to confirm this, for they make extensive use of architectural elements that predate the conquest and are assumed to have been purloined from temples destroyed in its wake. The idea of the trophy looms large in published discussions of these monuments, largely on the basis of a practice of reuse that has never been subjected to any serious analysis. Analysis has been obviated, in large measure, by the widespread perception that such reuse offers support for the lurid and highly formulaic tales of looting, spoliation and desecration found in the medieval texts that have been privileged in the construction of histories (and even art histories) of the period.

      In addition to the circularity of such an approach, the disciplinary divisions written into Orientalist discourse on South Asia at its inception frustrate the assumption of a diachronic approach to the material culture of the region. Yet some of the objects and buildings in which the past is instantiated for us today were already antique in the early pre-modern period. Enduring through time, such artifacts often ensured continuity in the process of imagining and re-imagining the past, providing a focus for the "necessary sedimentation of meaning that accumulates as part of the process of historical change." The palimpsest nature of such artifacts was intrinsic to their role as sites for an ongoing process of "translating" the past, a conceptual process that frequently encompassed a physical translation. This was a past that inhered not only in objects, however, but also in the practices and rituals associated with them. The act of physical appropriation might itself be palimpsest upon earlier re-uses of the same or similar objects, thus serving to construct a dynamic continuity between contemporary practices and their historical antecedents.

      This is the case with a number of pre-Islamic commemorative pillars (stambhas or lats) that were re-erected in Delhi during the thirteenth and fourteenth centuries. Essentialist notions of "Islamic" cultural practices have combined with traditional disciplinary divisions to obscure the transcultural nature of the pillars, which were central to the self-conscious articulation of an imagined relationship between the sultans of Delhi and the Indian past. The relocation of the pillars appears to reference a past that encompassed both a dimly perceived Indian antiquity and the immediate predecessors of the Delhi sultans, who provide a precedent for the reuse (that is, re-inscription and/or re-location) of such pillars. The existence of indigenous precedents for the re-use of the past suggests that there was much greater continuity in the characteristic cultural practices of Indian rulers between the pre- and post-conquest periods than has previously been acknowledged. What have been read as trophies at first glance, on closer inspection appear to offer evidence for transculturation in the ritual practices of the earliest Delhi sultans.

      The earliest instance of reuse following the conquest is evidenced by a seven and a half meter high iron pillar standing in the courtyard of the Friday Mosque built after the Ghurid conquest of Delhi in 1192, the mosque known today as the "Quwwat al-Islam". The pillar can be counted among a number of commemorative columns often referred to in inscriptions as pillars of fame (kirtistambhas) or pillars of victory (jayastambhas). A Sanskrit text inscribed upon it tells us that was originally dedicated as a standard (dhvaja) to a Vishnu temple by the fourth- or fifth-century ruler, Chandra, whose military prowess the inscription celebrates.

      It has generally been assumed that the erection of the pillar was contemporary with the foundation of the mosque by Qutb al-Din Aybak, the slave general of the Ghurid sultans, in 1192. According to the fourteenth-century Tarikh-i Firuz Shahi, however, the pillar was set up in this position not by Qutb al-Din Aybak, but by Shams al-Din Iltutmish. Iltutmish was a former Turkish slave who had risen through the ranks of the army in the service of Aybak, before acceding to the Indian sultanate that had emerged after the death of the last of the Ghurid sultans with effective control over the empire, in 1206. The date at which the pillar was installed is unknown, but it was presumably after the accession of Iltutmish in 1211, and possibly around or before 1229, when the area of the mosque was more than tripled, one of numerous architectural projects sponsored by the sultan.

      The reuse of the Delhi iron pillar in the early thirteenth century inspired the actions of later Indo-Islamic rulers, who similarly relocated antique pillars as part of their architectural patronage. The Tarikh-i Firuz Shahi makes it clear that it was the precedent set by Iltutmish that inspired the reuse of as many as ten antique pillars by one of his successors, Firuz Shah Tughluq (r. 1351-88). It seems likely that the re-erection of the iron pillar by Iltutmish was also a factor in the later reuse of an Ashokan pillar at Allahabad by the Mughal emperor Jahangir (r. 1605-27).

      Iltutmish's erection of the iron pillar in a mosque constructed in the wake of the Muslim conquest using an abundance of temple spolia has usually been assumed to reflect its value as a trophy, and its consequent ability to memorialize "Muslim victory" over the conquered Hindu population of Delhi. This assumption may be influenced by the European practice of re-erecting ancient obelisks looted from colonial possessions in metropolitan capitals, but it is also part of a broader belief in an essentially Muslim penchant for triumphal gestures involving the reuse of artifacts and monuments identified as "non-Islamic." The attribution of static sectarian identities to medieval objects and buildings is standard in traditional discourse on South Asian art and architecture. Thus the iron pillar in Delhi is often referred to as the "Hindu" iron pillar, an object in direct opposition to the adjacent "Islamic" minaret, ignoring the fact that the terms stambha (pillar) and minar (minaret) were used interchangeably to refer to the same objects by medieval Indians, Hindu and Muslim. Within such a paradigm, the presence of "Hindu" materials in a "Muslim" context is necessarily ascribed to the promulgation of sectarian victory rhetoric.

      Although the Delhi mosque was constructed using large quantities of spolia, the inclusion of the iron pillar was merited on something other than utilitarian grounds, for it fulfills no structural function. The re-situation of stone and metal Hindu icons looted during Iltutmish's campaigns of conquest in Ujjain (1233-34) along the approach to the mosque, where they could be trampled by those entering it, might support the idea that the iron pillar was intended to commemorate "Muslim victory". The display of the Ujjain loot harks back to earlier Islamicate precedents (recalling Mahmud of Ghazni’s treatment of the Somnath linga, for example), and can be ascribed to the ability of the looted icons to index the expanding frontiers of Iltutmish's empire very publicly in the first mosque of the imperial capital. However, it should not be assumed that all the objects garnered within the mosque had the same semantic function. There is nothing to suggest that the iron pillar was seized during one of Iltutmish's military campaigns, and it is unmentioned in contemporary sources, which are quite clear in associating the installation of looted icons in the Delhi mosque with the theme of imperial victory.

      Neither does the idea that the iron pillar (or any of those reused in later imperial monuments) was a trophy find support in the medieval sources. Ibn Battuta, writing after a visit to the Delhi mosque in or around 1333, gives the most extensive description of the iron pillar:

In the center of the mosque is the awe-inspiring column of which [it is said] nobody knows of what metal it is constructed. One of their learned men told me that it is called Haft Jûsh, which means 'seven metals' (sic), and that it is composed of these seven. A part of this column, of a finger's length, has been polished, and this polished part gives out a brilliant gleam. Iron makes no impression on it. It is thirty cubits high, and we rolled a turban round it, and the portion which encircled it measured eight cubits.
In the Tarikh-i Firuz Shahi we are told that fifty years later the Central Asian conqueror Timur was similarly awed by two of the pillars re-erected in the sultanate monuments of Delhi by order of Firuz Shah Tughluq. A reference in Elliott and Dowson's translation of the same work to these pillars being moved to Delhi "as trophies" is not found in the published Persian text, where their value as wonders (´aja’îbān) is instead stressed. Awe and mystery then, rather than triumph and victory, are the keynotes in the reception of these fragments of the Indian past in the fourteenth century.

      In a rare dissension from the tendency to ignore the transcultural nature of the pillars reused in Delhi, William McKibben notes that Iltutmish "may have appropriated the iron pillar for the Quwwat al-Islam mosque in part to glorify the achievements of past civilizations and affirm the ideological beginnings of Islamic rule in India by aligning himself with pre-Islamic sovereigns." A similar interpretation of Jahangir's later reuse of two pre-Islamic pillars was offered by Catherine Asher, who very plausibly read the inscription of Jahangir's royal lineage on the Allahabad pillar as an attempt to link Mughal rule "to both the Timurid tradition and to deeply rooted Indian traditions."

      In both cases, while the meaning of the reused pillars is considered in relation to their connection (however vague) with the pre-Islamic history of India, the practice of reuse itself and its cultural antecedents remains unexamined. If one considers the antique pillars reused in sultanate Delhi diachronically, however, it quickly becomes apparent that many of them had already acquired complex genealogies through reuse and re-inscription in preceding centuries. Just as Egyptian pharaohs or Byzantine emperors inscribed their names on columns and pillars that were already antique, medieval Indian rajas were apparently prone to re-inscribing existing commemorative pillars in order to commend their own glorious deeds to history. Some of these inscriptions correspond with the re-erection of the pillars during different phases of reuse.

      In the case of the Delhi iron pillar, although attention has been focused on the Gupta temple in which it first stood and the Ghurid mosque in which it eventually came to rest, in fact these only mark the beginning and end of its history. Among a number of less well-preserved inscriptions recorded on the column in the nineteenth century was one that referred to the foundation of Delhi by the Tomar ruler Anang Pal in Samvat 1109 (AD 1052). On epigraphic grounds it has been assumed that this is a contemporary inscription, carved at Anang Pal's behest more than six centuries after the column had first been dedicated. Some of the temple material reused in the Delhi mosque is of similar date, and it has been suggested that the iron pillar was taken from its original location and re-erected within a temple built by the Tomar ruler in 1052, at the time that the pillar was re-inscribed. On the basis of numismatic evidence, however, the date of Anang Pal's reign has recently been put at circa 1130-1145. Since no attempt seems to have been made to correlate the epigraphic and the numismatic evidence, this re-dating requires further research. If the later date is correct, however, the error in the date inscribed on the column suggests that the inscription is anachronistic and cannot therefore be used as evidence for the reuse of the pillar by Anang Pal in the eleventh century.

      The complicated genealogy of the Delhi column is by no means unusual. On the contrary, inscriptions on a number of reused antique pillars tell the same story. The Allahabad pillar re-erected by Jahangir in the early seventeenth century illustrates just how complex the life histories of such pillars could be. Reconstructing its travails from the multiple texts that it bears in different languages and scripts, the pillar was first inscribed in the third century BC by the Mauryan emperor Ashoka (although it may even have been erected earlier), then reused in the late fourth century AD by the Gupta raja Samudra, when it was carved with a list of his accomplishments, and those of his ancestors. The same pillar was re-erected once again in 1605, and carved with a Persian text celebrating the lineage of Jahangir. Such documented instances of reuse may be but the tip of the iceberg, since those reusing these pillars may not always have inscribed them, and even inscriptions are liable to wear and erasure over time.

      While the reuse and re-inscription of antique columns by later Indian rulers provides a general context in which to locate Iltutmish's appropriation of the iron pillar, the reuse of an antique pillar in the vicinity of Delhi in the decades preceding the Ghurid conquest points to more immediate precedents.  The pillar in question is one of five surviving from the architectural program of Firuz Shah Tughluq, and still stands where it was re-erected in 1357, at the center of an extraordinary pyramidal structure in the heart of Firuz Shah's new capital, Firuzabad, just north of Ghurid Delhi. The Firuzabad pillar predates the monument which it now graces by over fifteen hundred years, for it may already have been standing in the third century BC, when it was inscribed with an edict of the Buddhist emperor Ashoka, written in Prakrit in brahmi script. Just below this original dedication is a Sanskrit inscription carved in devnagari script a millennium and a half later, in 1164. The inscription records the conquests of prince Visala Deva, Vigraharaja IV of the Chauhan dynasty, which had taken Delhi from the Tomars a few years previously and still ruled most of northwestern India at the time of the Ghurid conquest three decades later. Among the victories mentioned in the pillar inscription is the defeat of a Mleccha (presumably Ghurid) army; in an ironic twist, plays concerning Visala Deva's battles against the Turushkas (including one written by the raja himself) were found inscribed on stones later reused in the Ghurid Friday Mosque at Ajmir. Since it seems unlikely that this pillar remained in one spot for fifteen centuries, it is probable that it was re-erected when it was reused in the late twelfth century by Visala Deva to commemorate his military victories.

      It is thus clear that some of the pillars that came to grace the monuments of the Delhi sultans had been transported around the north of the subcontinent over the course of a millennium or more. During this time they were erected, re-erected, inscribed and re-inscribed, by rulers of different dynasties and different faiths. That this fact has been overlooked is largely due to the neglect of (or selective quotation from) the inscriptions on these remarkable artifacts within an academic tradition partitioned in ways that make it difficult to deal diachronically with objects crossing rather arbitrarily defined cultural or taxonomic boundaries. Yet its importance for understanding the apparently anomalous reuse of such artifacts by Indo-Islamic sultans can hardly be overstated. The cumulative weight of the evidence discussed above suggests that the ritual practices of medieval north Indian kings encompassed not only the erection of commemorative pillars, but also the appropriation of those erected by their ancient predecessors.

      The existence of a prior tradition of reuse casts serious doubt upon the idea that the iron pillar in the Delhi mosque memorializes cultural rupture. Based on the evidence of the Firuzabad pillar, we can be certain that the Chauhan rulers of Delhi were reusing antique pillars less than three decades before the Ghurid conquest of the city. Seen in this light, Iltutmish’s re-erection of the iron pillar in the Friday Mosque of Delhi has little to do with cultural rupture. The gesture may have been intended to commemorate the victory of Islam through the deeds of the Ghurids and their successors, but it did so in a language and idiom adopted directly from indigenous Indian rulers. This was no mere appropriation of spolia designed to suggest a symbolic continuity with pre-Muslim kingship, but the actual continuation of a practice associated with medieval Indian kings. Culturally and chronologically, the relocation of the iron pillar by Iltutmish can be seen as a pivotal act, linking the actions of later Islamicate rulers with a pre-conquest tradition of reuse.

      The peregrinations of the antique pillars bear comparison with those of the icons and insignia that were ritually appropriated and re-appropriated by medieval Indian rulers. As Richard Davis has shown, the meaning of these objects was directly related to their possession of a history, to the existence of a genealogy memorialized in contemporary narratives (both oral and textual), or even inscribed upon the objects themselves and the buildings that housed them. Such relics were inalienable by virtue of an indexical relationship with kingship and their consequent possession of the power to define the historical identity of those who deployed them. Possessing the ability to "attract new meanings, fictitious memories, altered genealogies, and imagined ancestors," inalienable objects are ideally suited to confer legitimacy on those associated with them.

      The association of the pillars with earlier Indian kings (historical and mythical) was not only stressed in medieval Persian sources, but in some cases was literally legible. A keen interest in the original cultural context of the reused pillars is suggested by attempts to decipher the inscriptions on the pillars reused in the fourteenth century by Firuz Shah Tughluq, an interest in antique epigraphy that recalls the response of earlier Persian rulers to the relics of the pre-Islamic past. Sanskrit inscriptions were evidently read with a high degree of accuracy, for the Sirat-i Firuz Shahi, the Persian text that records the removal of the Firuzabad pillar, reports quite correctly that the writing on the column commemorates its re-inscription by prince Visala Deva two centuries earlier, and notes the fact that it had once been associated with a temple. By contrast, the Prakrit inscriptions on the Firuzabad pillar remained elusive, connoting a mythologized antiquity. Even here the inscriptions were integral to the perceived meaning of the pillars, since monumental texts are capable of evoking power not only through their content, but also "through their location in space and the way they look." That the only two pillars to be inscribed when reused in an Islamicate context were inscribed with genealogical texts very similar in content and nature to those that Gupta and Chauhan rajas had earlier carved on similar pillars is strong evidence for a continued association between reuse, kingship and legitimacy.

      The act of epigraphic translation prefigured a physical translatio that was no less relevant to the issue of legitimacy, for it was not only the pillars that were palimpsests, but the acts and practices associated with them. The very ability to move and re-erect these extraordinary relics of the Indian past echoed the original act of creation, conveying significant messages about patronage and power in a manner determined by royal precedent. Such heroic endeavors were memorialized in contemporary Persian texts, which prefigure European treatises on similar topics by several centuries. The reported failure of earlier kings to move the pillars transported by Firuz Shah Tughluq, and the demise of later attempts to repeat such feats, only serve to underline the extraordinary skills needed to move pillars, which could weigh up to thirty tons or more.

      The attempt to foster a sense of legitimacy by forging an association with a distinguished political lineage, and with the glories of a real or imagined past, was a common concern of those engaged in the business of state formation in medieval Iran and South Asia. In both realms, the construction of genealogical histories relating the present to the perceived glories of the past finds a visual counterpart in the patronage of archaizing art and architecture, or the reuse, recontextualization, and reworking of carefully selected relics of the past. Occasionally the textual and artifactual coincided in the articulation of genealogical claims, as when Mughal emperors re-inscribed royal artifacts bearing the accumulated names of illustrious predecessors, or when epic passages from the Shahnama, the Iranian book of kings, were quoted on the walls of cities and palaces, in an endeavor "to legitimize the present through identification with the past."

      The latter case is particularly interesting, evidencing as it does the ability of the mythologized pre-Islamic past to serve as an instrument of legitimation. The Shahnama was reportedly inspired by an ancient Indian king's patronage of historical and allegorical texts. Like the "Mirrors for Princes" literature of the eleventh and twelfth century, it provided a paradigm for a re-framing of the pre-Islamic kingly past within an Islamicate matrix that emphasized the shared experience of kingship. Just as the appeal to authority in contemporary Persianate histories was "linked to the appropriation of prior authoritative narratives," the epic past enshrined in texts like the Shahnama came to provide a legitimizing genealogy for medieval Persianate rulers, a vehicle by means of which the present was "instantly incorporated into a victorious tradition." The use of past precedent to frame the contemporary encounter with India can be seen in the commissioning of the Shahriyar-nama, a text that celebrates the Indian exploits of the great grandson of Rustam, one of the heroes of the Shahnama. The text was commissioned by the Ghaznavid sultan Mas'ud III, who also campaigned in India, and who was earlier celebrated in an epic poem inscribed on the walls of his palace, which gives the genealogy of the Ghaznavid sultans in the meter of the Shahnama. Slightly later, the Ghurid overlords of Iltutmish devised a lineage that related themselves both to the kings of pre-Islamic Iran and to the Arabic-Islamic past embodied in the caliphate, invoking (like many Persian kings before and after them) various material relics of that past to bolster their claims to a noble lineage.

      The paradigmatic role of the pre-Islamic past in conferring legitimacy on parvenu Persianate dynasts offers one potential model for the co-option of the Indian past and its material traces by Iltutmish and the newly emergent Delhi sultanate. Contemporary evidence for the role of material remains within such a paradigm may be found in Seljuq Anatolia, geographically remote from sultanate India, but culturally contiguous by virtue of its shared Persianate Turkic cultural milieu. Seljuq participation in the pre-Islamic past of Anatolia was orchestrated by means of spolia, historical and mythological tales, and textual quotation. Scott Redford notes of the architectural program undertaken at Sinop after its conquest in 1214 that, “"Sultan Izzedin Keykavus seems to have placed himself in a mythic context, keeping company with the kings of yore, whether Caesar or Khusraw." Similarly, the walls of Konya erected by Alaeddin Keykubad in 1219-21 incorporated Classical and Byzantine spolia along with epigraphy, which included both religious texts and quotations from the Shahnama. The latter functioned as a vector of appropriation that extended the mythic age of pre-Islamic Iran to Anatolia and permitted the assimilation of a pre-Islamic past instantiated in the material remains of cultures which, strictly speaking, lay outside the cultural ambit of the text.

      This extrapolation from the mythologized past of pre-Islamic Iran to the instantiated past of pre-Islamic Anatolia represents the transcultural extension of a type of isnad paradigm, by means of which the past lent its authority to the present by virtue of identification with a valorized precedent. European encounters with India similarly attest to the fact that, "symbolic representations of power could be 'translated' on the basis of cross-cultural analogy." Both undertakings bring to mind Walter Benjamin' classic formulation of the translator's art as "a somewhat provisional way of coming to terms with the foreignness" of its object. Although no equivalent for the stambhas and lats of India existed in Iran, the evidence from Seljuq Anatolia suggests that they were potentially assimilable as a variant of the known. Such pillars may even have been familiar from pre-existing descriptions of the region, for a tale preserved in the eleventh-century Egyptian compendium, the Kitāb al-Hadāyā wa al-Tuhaf refers to a mysterious iron pillar encountered by an Arab army in India during an earlier phase of conquest. According to this tale, the 'Abbasid governor of Sind encountered an iron pillar seventy cubits long in 'Kandahar' (probably the region of Gandhara in north-eastern Afghanistan and north-western Pakistan) during an attempt to conquer India in 768. The governor was told that the column was a victory monument erected by Tubba', the celebrated pre-Islamic ruler of Yemen, and fashioned from the metal of weapons used in gaining the victory that it commemorated. The association of the Kandahar column with a known figure from pre-Islamic Arabia sets the 'Abbasid campaign of conquest in an epic context, while demonstrating how familiar figures of antiquity could come to be associated with the obscure relics of Indian antiquity. In similar fashion, the pillars reused by Firuz Shah Tughluq came to be seen as relics from the time of Alexander the Great.

      Such identifications suggest that what was at stake in the appropriation of the antique pillars was neither the assimilation of a living adversary, nor a past characterized by a single antecedent regime, but the palimpsest of cumulative heroic pasts. It is this that distinguishes the re-erection of the pillar in the Delhi mosque from the contemporary display of recently acquired Indian loot in the same monument. The commemorative value of the iron column lay not in its specific associations with the historical kings of India, but in its ability to represent  "the broader notion of Indian kingship, regardless of whether specific names or deeds were known.” As “visual substitutes" for history, inalienable objects such as insignia, regalia, or royal pillars of fame are ideally suited to "bringing past time into the present, so that the histories of ancestors, titles, or mythological events become an intrinsic part of a person’s identity." The construction of identity through the incorporation of such objects is directly relevant to the re-erection of antique pillars by the Delhi sultans, for as William McKibben notes:

The ‘translation’ of the Indian past into the Sultanate present may have been read as a retrospective "Islamicization" according to existing paradigms developed in relation other pre-Islamic pasts. The manner in which this was achieved, however, reveals a dependence on indigenous Indian models of legitimation. If the iron pillar itself was a fragment of the distant mythic past, the reuse of similar objects during the more recent Chauhan past was conceivably within living memory when Iltutmish incorporated it into the Delhi mosque in the early thirteenth century. The potential for legitimation lay not just in the pillar itself, therefore, but also in the act of translating it, which belonged to the normative realm of Indian kingly self-representation.

      The benefits of conforming to type were no doubt obvious to the Ghurid conquerors of India and their parvenu successors such as Iltutmish, sultan of an Indian empire within which Muslims were a minority. Through the manipulation of inalienable objects, Iltutmish and his successors encompassed and incorporated an Indian history within which they were in their turn accommodated. That the process of articulating legitimate sovereignty was a bilateral one is clear from the well-known Palam inscription, written in 1276, eight decades after the conquest of Delhi. Enumerating the dynastic changes in the region during the eleventh to thirteenth centuries, the author of the Sanskrit text writes:

The land of Haryana was first enjoyed by the Tomaras and then by the Chauhan. It is now ruled by the Shaka kings (i.e. the sultans). First came Sahabadina (i.e. Shihab al-Din Ghuri), then Khudavadina (i.e. Qutb al-Din Aibak), master of the earth, Samusdina (i.e. Shams al-Din Iltutmish), then Pherujsāhi (i.e. Firûz Shāhi), lord of the earth.
This seamless integration of Ghurid and early sultanate rulers into South Asian history as just the latest in a long line of conquering monarchs betrays little sense of the religious, political, and cultural rupture so often highlighted by historians and art historians alike. Such assimilation may be a product of a pragmatic urge to legitimize effective political authority, but it is surely just as much a product of the ability of early Indo-Islamic sultans to engage with existing traditions, to project their authority in the expected manner. In part at least, the incorporation of the Delhi sultans into indigenous Indian histories and royal genealogies reflects the successful manipulation of semiotically charged artifacts whose power resided in the fact that they were already possessed of a distinguished history.

Finbarr Barry Flood

Department of Fine Arts, New York University.