The Emperor's Carpet

A silk-foundation, wool-pile court carpet nearly 7.6 meters long and 3.4 meters wide, woven in Safavid Iran in the second half of the sixteenth century. At roughly 65 kilograms it takes more than one person to lift, and the Metropolitan Museum's own handling photographs show it held up by two staff members with a third steadying it from behind, the carpet still dwarfing all three. This is not a small or domestic object; it is a court-scale floor covering meant to be seen from a distance as a single field of color and read up close as an almost inexhaustible field of individual motifs.

The ground is a deep red on which a dense lattice of scrolling vines, composite palmettes, cloud bands, buds and blossoms carries a full menagerie: dragons, lions, tigers, leopards, a Chinese antelope-like creature, ducks, pheasants and other birds, some locked in combat, others moving quietly through the foliage. A wide, dark blue-green border of matching density frames the field, separated from it by a narrow band of continuous script. The whole composition reads, at scale, as a single controlled surface rather than a repeating pattern, which is the effect the finest sixteenth-century Safavid medallion carpets were built to produce.

The field is organized around a system of medallions and cartouches worked into the vine lattice rather than a single dominant central medallion; smaller rosette medallions recur at intervals along the long axis, each treated as a self-contained roundel of concentric bands. Confronted and combatant animals are placed through the field in pairs and small groups: a golden lion brings down a spotted, deer-like creature in one passage; leopards, tigers and dragons appear elsewhere locked in the same kind of encounter, drawn with enough anatomical control that species can still be told apart at eight hundred knots to the square inch and more.

Birds appear more quietly, often flanking the small medallions in pairs, sometimes with stemmed floral pendants hanging between them in a way that has been read, on comparable Safavid court hangings and carpets, as an evocation of a suspended lamp. This record follows the discipline the subject requires: the animal-combat and lamp-like pendant motifs are a genuinely documented feature of the Safavid court medallion-carpet design vocabulary of this period, discussed by the Museum's own department and in the wider literature on this class of object, but no source consulted for this record ties a specific doctrinal or devotional reading to this individual carpet, as opposed to the type. What can be said without overreaching is narrower and more useful: a court patron who commissioned a carpet of this size, in silk and wool, carrying real and imagined animals in combat around a field addressed in verse to 'the king of the world,' was commissioning an object about sovereignty and dominion over the natural world, not a decorative textile in the modern sense.

The inner guard band carries a continuous run of Persian verse drawn from the poetry of Zahir al-Din Faryabi, active in the second half of the twelfth century and dying in 1201 (the Metropolitan Museum's own material describes him as a thirteenth-century poet, a rounding this record notes rather than silently repeats). The verses are conventional courtly nature poetry, likening a ruler's spring garden to Paradise and his dominion to the freshness of the meadow after rain, and they address an unnamed 'king of the world.'

Putting a specific, older poet's verses on a court carpet rather than composing new text for the object was itself a choice: it borrows the authority of an established literary voice for the flattery being offered to whichever Safavid ruler the carpet was made for, rather than inventing praise that stood or fell on its own. No source consulted names the specific patron addressed by the inscription, and this record does not supply one.

The Metropolitan Museum dates the carpet broadly to the second half of the sixteenth century, a window of some fifty years that runs from the later part of Shah Tahmasp's long reign (1524-1576) through the brief reigns of Isma'il II and Mohammad Khodabanda and into the early years of Shah Abbas I, who acceded in 1587. Secondary accounts of the carpet sometimes state that it was made for Tahmasp specifically, but none of the sources consulted for this record trace that claim to a document, an inventory, or a named scholar's published argument; it appears to rest on Tahmasp's long reign occupying most of the dating window and on his documented role as the Safavid ruler most associated with the flowering of the court design studio, rather than on evidence tied to this carpet. This record treats the identity of the commissioning shah, if the carpet was a royal commission at all rather than court-workshop production for a wider elite market, as unresolved.

The Metropolitan Museum's own collection record commits only to Iran, marking the city of manufacture as unresolved. The wider scholarly literature has not settled the question either, and this carpet, together with its twin now in Vienna, sits at the center of the dispute.

Maurice Dimand, the Metropolitan's own longtime curator of Near Eastern art, placed this group at Tabriz on stylistic grounds, treating it as part of a broader family of early Safavid medallion carpets he assigned to that city as the dynasty's main artistic and court-design center. Against that stands an older, more traditional attribution to Kashan, which rests chiefly on Kashan's established fame as a center of fine silk weaving and on the city's documented production, a generation or two later in the early seventeenth century, of carpets of comparably luxurious materials and technique; the traditional Kashan attribution for this particular sixteenth-century group is not supported by direct sixteenth-century documentation naming Kashan as the place of manufacture. A third city, Herat, has also been proposed in the wider literature on Safavid court carpets of this general type, reflecting Herat's own status as a major court-art center under the Timurids and into the early Safavid period. No signature, no workshop mark and no court record consulted for this record names a city for this specific carpet, and this record does not adopt one of the three positions over the others.

The carpet is woven with a silk warp and weft and an asymmetrically knotted wool pile, the open or 'Persian' knot structure long associated with the finest Iranian court production, as opposed to the symmetric knot more typical of Anatolian weaving. A silk foundation under a wool pile of this size and knot density represents a very large investment in the most expensive locally available fiber before a single dye is applied; the choice of silk for warp and weft, where a coarser cotton or wool foundation would have served structurally, is itself a statement about the resources a court workshop was prepared to commit to an object meant to be walked on.

None of the sources consulted for this record report a published pigment or dye analysis specific to this carpet. General accounts of Safavid court weaving describe natural dye sources such as madder for red tones and other vegetable and mineral dyes for the remaining palette, but this record does not attach an unsourced generic claim about dye chemistry to this individual object; a technical dye study, if one exists in the Museum's own conservation literature, was not located in the sources consulted and is recorded here as an open question rather than filled with an assumption.

The carpet reached the Metropolitan Museum in extremely fragile condition and, in the Museum's own account of its recent conservation program, was shown to the public only twice in the roughly sixty years that followed, a direct consequence of how little handling and light exposure the textile could safely bear. Beginning in 2006 the Museum's Department of Textile Conservation undertook a three-year stabilization project aimed at making the carpet safe for regular display; published accounts of the project describe the removal of some seven hundred old patches accumulated over the carpet's working and museum life, along with the broader structural stabilization needed to support its own weight when hung or laid out.

The conservation project coincided with, and fed into, the Museum's 2006-2011 renovation of its galleries for the art of the Arab lands, Turkey, Iran, Central Asia and later South Asia, which reopened in 2011 with the carpet installed in Gallery 462 as one of the space's acknowledged highlights. In 2013 the Museum's exhibition 'Making the Invisible Visible' used conservation work of this kind, including on textiles from the Islamic Art department, to show the public what a conservator actually does to a fragile object of this scale before it can safely go back on view.

The carpet is called the Emperor's Carpet, in the singular, but the name belongs properly to a pair: this carpet and a companion, near-identical in design, now in the Museum fur angewandte Kunst (MAK) in Vienna. Both take the name from the same story, which the Metropolitan Museum's own published material states in careful, hedged language rather than as settled fact: the carpets are 'said to have' once belonged to Czar Peter the Great of Russia, who is supposed to have passed one or both to the Holy Roman Emperor Leopold I of Austria after 1698, after which they entered the Habsburg imperial collection and were used to furnish a Habsburg summer residence.

The wording matters. 'Said to have' is the Museum's own register for a tradition it has not independently verified against a Russian or Habsburg court document, not a claim it is making as established fact. This record follows that hedge rather than smoothing the Peter the Great and Leopold I story into settled history, because no primary Habsburg inventory or Russian court record confirming the 1698 transfer was located in the sources consulted. What is better supported is the carpet's presence in the Habsburg imperial collection by some point before the twentieth century and its use, at least by tradition, at an imperial residence; the specific mechanism by which it arrived there remains a named but unverified tradition, not a documented event, and this record grades it accordingly.

The Habsburg monarchy fell at the end of 1918, and the property of the former imperial house passed to the new Austrian Republic; carpets and other objects that had furnished the imperial residences were transferred into Austrian state museum custody in the early 1920s. Sotheby's own catalogue note on a later, unrelated reproduction of the design states the sequence plainly for the pair under discussion here: in 1925, with the Habsburg monarchy gone, the Viennese museum holding the pair sold one of the two carpets, to raise funds, to the London dealers Cardinal and Harford, while keeping its twin, which remains in Vienna today.

Cardinal and Harford held the carpet in London from 1925 until 1928, when it was consigned to Christie, Manson and Woods and sold at auction in London on 5 July 1928 as lot 146. This is a genuine, documented public sale in the carpet's history, recorded in the bibliography of a later Sotheby's catalogue note in the standard auction-house citation form; no source consulted for this record gives the hammer price or the identity of every bidder, and this record does not invent either. What the sources agree on is the buyer's side of the transaction that followed close behind the sale: the carpet passed, by or shortly after 1928, to the American collector Edith Rockefeller McCormick of Chicago, acquired for her by the dealer and scholar Arthur Upham Pope.

Edith Rockefeller McCormick (1872-1932) was the daughter of John D. Rockefeller and the wife of Harold Fowler McCormick, of the Chicago reaper and later International Harvester fortune; she was a major collector and patron in Chicago in the 1910s through early 1930s, with a country estate, Villa Turicum, in Lake Forest, Illinois. The Art Institute of Chicago's own exhibition records confirm that she lent the carpet to the museum under the title 'The Emperor's Carpet lent by Mrs. Edith Rockefeller McCormick,' on view from 4 April to 1 November 1931, a run that followed the carpet's appearance, earlier the same year, in the landmark International Exhibition of Persian Art at the Royal Academy of Arts in Burlington House, London, organized by Arthur Upham Pope and shown there from January to February 1931 under theatrical, tinted lighting chosen to intensify the carpet's colors for visitors.

Pope's role in this carpet's history is worth naming plainly, in the register this record's research standard requires for the people in a provenance chain, not only its owners. He was simultaneously a dealer who bought and sold Persian art, the organizer of the exhibition in which the carpet was shown, and the scholar who then published on its beauty, first in Art News and again, using his position as an advisor to the Art Institute of Chicago, in that museum's own Bulletin. He acquired the carpet for McCormick in 1928 and, after her death in 1932, was the dealer who arranged its sale onward, in 1943, to the Metropolitan Museum. A single figure functioning as agent, exhibitor and critic of the same object across fifteen years is a genuine conflict of roles by any modern standard, and this record states it rather than treating Pope's published praise of the carpet as disinterested scholarship.

Edith Rockefeller McCormick died on 25 August 1932. The carpet does not appear to have been part of the general household-contents sales of her estate held in Chicago and Lake Forest in January 1934, which were dominated by French furniture, books, linens, porcelain and silver rather than major carpets; this record did not locate the carpet in the sale catalogues checked, and the object's custody between 1932 and 1943 is not documented in the sources consulted. In 1943 Arthur Upham Pope, again acting as dealer, sold the carpet to the Metropolitan Museum of Art, which purchased it with the Rogers Fund; it was accessioned as 43.121.1. A second, related carpet, described by the Museum as a Vase carpet of the late sixteenth or early seventeenth century, was accessioned at the same time as 43.121.2, also through the Rogers Fund, which is consistent with both carpets having reached the Museum through the same 1943 transaction with Pope, though this record has not located a single acquisition document confirming that explicitly.

One secondary account states that the carpet arrived at the Museum in 1941, two years before the accession date the Museum's own record gives; this record has not been able to resolve that discrepancy and states it rather than silently adopting one date. The Museum's own accession number and credit line, verified against its own open-access catalogue record, are given above as the primary facts of the acquisition.

The carpet's companion remains in Vienna, in the Museum fur angewandte Kunst (MAK), part of a carpet collection built substantially from the former Habsburg imperial holdings and numbering some two hundred pieces, of which roughly thirty are normally on public display. The two carpets were designed and, in all likelihood, woven as a related pair or at least a closely matched set, sharing the medallion, animal and border vocabulary this record describes above; MAK's own published material was not fetched directly for this record, and this record therefore does not state Vienna's current inventory number, having found more than one candidate number attributed to the piece in different indirect sources without being able to verify which is correct against the museum's own database. That verification is recorded here as an open task rather than guessed at.

Separating the two carpets in any account of this object matters for the same reason the site's own past errors on paired and twinned works matter: a caption, a dimension or a conservation detail that belongs to the Vienna carpet must never be silently presented as belonging to the New York carpet, or the reverse. This record describes only the Metropolitan's carpet, 43.121.1, in its technical and provenance detail, and treats the Vienna piece strictly as a named comparandum.

The weight of the documented record here is real: a public sale at Christie's in 1928 with a lot number, a dated loan exhibition in Chicago in 1931 with a lender's name on the wall label, a named dealer, a Rogers Fund purchase with an accession number the Museum's own database confirms. Set against that, the earlier and later ends of the chain remain genuinely open. The 1698 Peter the Great and Leopold I story is a named tradition, stated by the Museum itself in hedged language, not a documented transfer. The city of manufacture is a live, three-way scholarly disagreement that this carpet's own catalogue record declines to resolve. Eleven years of the object's history, between Edith Rockefeller McCormick's death in 1932 and its sale to the Museum in 1943, are undocumented in the sources this record was able to consult. This record states each of those gaps as a gap, in keeping with the standard the rest of this archive holds itself to, rather than bridging any of them with an invented owner, date or city.