|Cross-shaped tile 13th century|
by Margaret S. Graves
The town of Kashan, as a city associated with high-quality ceramic production in the medieval period, became a distinct point of scholarly attention in the twentieth century. It appears to have been a major site for the manufacture of fine wares between the 1170s and 1220s (565-620s H.) as well as in the later 13th and early 14th centuries.
The association between premodern Kashan and ceramic production is attested in the use of terms derived from the site’s name (kāši/kāšāni, or qāši/qāšāni) as the general designation for glazed tilework in Arabic and Persian sources. Yāqut Ḥamawi (d. 626/1229) notes in Moʿjam al-boldān (written in the 1220s) that Kashan’s ceramic wares were widely exported (Yāqut, IV, p. 15; Watson, 1973-75, p. 3). The town was well situated for the development of this industry, being located near sources of some of the materials necessary for manufacturing the so-called stonepaste or fritware, a fine, hard, and light-colored ceramic body of ground silica with additions of glass frit and clay. The two major decorative techniques found on the high-status ceramics commonly associated with medieval Kashan are luster painting and the polychrome enamel decoration frequently termed mināʾi in art historical literature (although this term is not normally associated with ceramics in medieval sources). While entirely distinct, these two techniques both involve fixing overglaze painting onto previously glazed and fired stonepaste bodies through a secondary firing at a lower temperature, and they were at times practiced by the same painters, occasionally even appearing on the same vessels. A wide variety of forms are included in the corpus attributed to Kashan, from closed vessels such as bottles through plates and bowls to tilework of various dimensions.
|Kashan.Bowl Abu Zayd al-Kashani|
The high quality wares produced in the late twelfth to fourteenth centuries in Iran, particularly those decorated with luster, are notable for the loquacity of their inscriptions. From these, much documentary information has been gleaned. Foremost amongst the evidence for Kashan’s role in the production of fine ceramics are the names of potters with the nesba “Kāšāni” or “Qāšāni” (from Kashan) or inscriptions that name Kashan as the site of creation.
Most notable amongst the potters’ names associated with Kashan is that of Abu Zayd. It should, however, be noted that while the nesba “Kāšāni” was once thought to appear after this name on two enameled bowls , more recent research by Oliver Watson has demonstrated that the relevant inscriptions were misread and the Kāšāni nesba does not in fact appear with this name on any known works (Watson, 1994). Therefore, while the name “Abu Zayd Kāšāni” continues to circulate in scholarship on Iranian luster, there no longerappears to be any justification for this appellation. To date, thirty-two surviving vessels, tiles, and fragments executed in both luster and enameling techniques, and bearing dates that span thirty-four years, have been associated with “Abu Zayd” , although many of these show only a part of the signatory formula associated with this figure (be-ḵaṭṭehi, “in his own hand”) and do not include the name “Abu Zayd” itself. The earliest dated piece signed “Abu Zayd” is an enameled bowl decorated with a central scene of seated and standing figures before a pond and dated 4 Moḥarram 582/27 March 1186 . “Abu Zayd” is the only name so far known in this milieu to appear on both vessels and tiles, the latter appearing in the shrines at Qom (dated 602/1205-6) and Mashadalongside tiles signed by Moḥammad b. Abi Ṭāher of the Abu Ṭāher family, another celebrated potter associated with Kashan.
|Kashan.Fritware with lustre decoration|
The Abu Ṭāher family of luster potters can be traced through four generations that span the Mongol conquest of Iran, the last two generations of which used the Kāšāni nesba on occasion (Watson, 1973-75, n. 18). In the case of family member Yusof (active in the early 14th century), Watson records the only instance of a documented signed luster potter, signing a work a work made in the cheaper medium of underglaze-painted ceramic. Yusof’s signature, ending with the epithet “al-Qāšāni,” can be seen on a tile bearing a foundation inscription in the Masjed-e Qalʿa in Qohrud, which Watson (1975, pp. 62-63) dated to 716-27 (1316-27). Yusof’s brother Abu’l-Qāsem ʿAbd-Allāh Kāšāni, a historian at the court of Oljāytu, leftan important document for the history of Iranian ceramic production within his treatise on precious stones, ʿArāʾes al-jawāher wa nafāʾes al-aṭāʾeb (700/1300-01). The concluding section of this text details the materials and techniques of Iranian potters, including a well-known description of the luster-painting process, as well as mention of potting practices at Kashan. These passages also include a reference to “seven-colored” (haft-rang) vessels, a term presumed by James Allan and others to refer to enameled wares. The probable inspiration for Abu’l-Qāsem’s text was an earlier work, Jawāher-nāma of Moḥammad b. Abi’l-Barakāt Jawhari Nišāpuri (Porter, 2003), dated 592/1196. Jawharidoes not state that his knowledge of ceramic production was acquired in Kashan, but he does mention the town as a site of the manufacture of fine ceramics (Porter, 2004, p. 345).
|Kashan.Dish with seated figure|
Various other names, some of them with the nesba Kāšāni or Kāši, have also been associated with medieval Iranian luster production. Yedda Godard and more fully Watson discuss further named potters, while Sheila Blair has documented a family of luster potters who claimed descent from ʿAli b. Bābawayh and worked alongside members of the Abu Ṭāher family in the early fourteenth century (Blair, 1986). More recently ʿAbd-Allāh Qučāni has suggested that an inscription on a late twelfth- or early thirteenth-century gilded and overglaze-painted faceted dish in the Robert Lehman collection should be read as the signature of a painter called Borhān or Mardān Kāšāni (reading cited inKomaroff, p.355); if so, this would add some weight to the hypothesis that Kashan was a major site of production for other types of fine ceramic beyond luster. Qučāni has also identified several new names of tile-workers within the tilework of the shrine in Mashad, to be discussed in a forthcoming publication.
Beyond nesbas, there are also some circumstantial inscriptions that lend weight to the case for Kashan as the major center of luster potting. In the best-known example, Moḥammad b.Moḥammad Nišāpuri, working in a style comparable to that of dated wares from the first two decades of the 600s/1200s, signed a bowl found at Gorgān with the formula al-moqim be-Qāšān. The suggestion that this was the signature of a potter displaced from Nishapur by the Mongol invasions and keen to stress his current association with a town renowned for its fine ceramics is certainly plausible (Watson, 1973-75, p. 4). From the twilight of luster production in the Mongol period survives a group of star tiles dated to 738/1338 and 739/1339 bearing inscriptions that state that they were made in Kashan. A rather plaintive inscription on one of these late tiles pleads for Kashan to be protected from the ravages of time, perhaps a sign of the site’s declining fortunes in the fourteenth century .
Luster production at Kashan
The creation of luster-painted ware is a costly process that requires considerable expertise. Compounds containing metal oxides, sulphur, and a refractory medium, such as ochre, are painted onto a previously glazed and fired surface. On the wares associated with Kashan the glaze was normally opacified with tin, although a (presumably cheaper) clear glaze was sometimes used for the internal surfaces of closed forms such as bottles and jugs (Morgan, p. 162). Clear glaze, sometimes tinted with cobalt, was also used for the exterior surfaces of bowls. Once the glaze has been fired and decorated with the luster compound a second firing at a lower temperature with restricted airflow removes the oxygen from the metal oxide, fixing a layer of metal only a few molecules thick with a lustrous sheen upon the surface of the vessel or tile. The temperature and duration of the second firing must be very precisely controlled, and, when taken alongside the volatility of metal oxides in the kiln, it is unsurprising that faults of various types are quite common in lusterware. The technical knowledge required is complex enough to have led to suggestion of a “monopoly” on luster production in the late twelfth- and thirteenth-century Iranian lands, linked by Watson to the single site of Kashan (Watson, 1975, pp. 65-66; idem, 1985, pp. 37-44).
While Kashan’s supremacy in the production of widely distributed fine medieval luster vessels and tiles is now generally accepted, other major sites of luster production were proposed in earlier twentieth-century scholarship, principally Ray, Gorgān(see below) and Sāva. Countering these suggestions, Watson argued in his landmark study Persian Lustre Ware (1985, pp. 37-44) that stylistic differences seen amongst the surviving corpus of pre-Mongol Iranian luster wares should be read, not as evidence of different sites of manufacture, but as the result of three distinct stylistic modes of luster decoration created at Kashan between the 1170s and early 1220s (565-620 H.), before the Mongol invasions sent the industry into hiatus.He terms these the “monumental style,” the “miniature style,” and the “Kashan style”, and notes that the large number of surviving dated pieces has permitted the construction of a chronology.
Kashan lustre bowl with two seated figures, birds, and
Watson’s tripartite stylistic model of pre-Mongol luster production at Kashan has been generally accepted, although some scholars (e.g., Morgan; Mason) have suggested renaming and subdividing these three stages. Mason’s petrofabric analyses would appear to show a very similar fabric across all of the medieval Iranian lusterwares that he examined, tentatively strengthening the case for a single site of production. However, the hypothesis that Kashan was the only site of luster production within the medieval Iranian lands is still under discussion. Mehdi Bahrami proposed that the remarkable vessels uncovered at Gorgān in the first half of the twentieth century, stored inside large earthenware jars that had been left apparently undisturbed since the medieval period, were local products rather than imported Kashan wares (Bahrami, 1949, p. 16 and passim). This thesis never met with general acceptance in art historical scholarship, but it continues to circulate in some quarters. Bahrami and later Muhammad Yusuf Kiani reported finding luster sherds inside the kilns excavated at Gorgān and also the discovery of luster wasters at the site, but more information than has thus far been published isneeded before making a definite assessment (Kiani, p. 38; see also Watson, 2006a, pp. 247-48). By the same token, however, it should be noted that, although kiln sites have been uncovered at Kashan, no evidence of luster processes has been reported (Bahrami, 1947, pp. 225-29). More recently, Scott Redford and James Blackman (pp. 234-35) have raised the question of whether a decentralized model of luster production might not have held true in Iran as it apparently did in Syria, while some scholars working on Central Asian material have also argued for multiple sites of luster production (see Blair, 2008, n. 10). Luster wasters have reportedly been found at Marv, although these do not yet appear to have been published (Gascoigne and Bridgman, 2010, p. 118, n. 32). While the question remains open, at present substantial evidence for luster production at sites other than Kashan—in the form of inscriptions implicating other locations, scientifically documented kiln excavations showing evidence of luster production, incontrovertibleluster wasters, textual evidence from medieval sources—is yet to appear.
|Kashan.Fritware tile with a dragon in luster|
While luster-decorated tiles were certainly manufactured in pre-Mongol Iran, much of the production of lusterware before 1220s seems to have been vessels, of both open and closed types. The first of Watson’s proposed stylistic groups of Kashan lusterware, the so-called “monumental style,” is characterized by luster painting in reserve, normally on a white background, although sometimes on a transparent glaze. Painted vessels in this category show generally large figures and designs executed with fluent and relatively free brushwork. The moon-faced or Buddha-faced ideal of beauty, a notion also attested in medieval literary sources, is widely in evidence on these wares. This manner of painting, which does not appear on any dated pieces, is closest to that of the luster wares produced in Egypt and Syria in the eleventh and twelfth centuries. For this reason, the “monumental style” has been called upon as evidence for the so-called “migration theory” that explains the advent of luster technologies in medieval Iran through the geographical movement of potters escaping the economic collapse of the Fatimid state in Egypt (an idea first proposed in Butler, pp. 18-23 and expounded in Lane, pp. 37-38).
Watson’s second category of luster decoration attributed to Kashan workshops, the “miniature style,” had developed by 575/1179-80. This is the year recorded on the earliest surviving dated pieces of Iranian luster (a bottle in the British Museum, , discussed in Watson, 1985, pl. 37, and also a bowl formerly in the Plotnick collection,). He suggests that there
is no reason to suppose that the “monumental” and the “miniature” styles were not made alongside each other, with both ceasing production some time after 595/1198-89, the date inscribed on the latest dated piece of “miniature style” lusterware (Watson, 1985, pl. 54). Notable for small-scale, rather rapid and sometimes sketchily executed designs painted directly in luster on an opaque white ground, and often featuring multiple figures; this style includes the key diagnostics of checkerboard trees (commonly approximating a cypress shape) and plants formed of lines sandwiched between rows of dots. In both painting and vessel forms (such as the lobed bowl) the lusterware decorated in this manner has much in common with enameled wares of approximately the same period (see below), leading to the suggestion that the “miniature style” was first developed for enameled wares on which the detailing would have been more legible (Watson, 1985, p. 70). At the same time, several authors have noted the very close relationship between the “miniature style,” with its fixed ground planes and small, dynamic figures, and book illustration (particularly the earliest surviving illustrated Persian manuscript (ca. 648/1250) of the versified romance, Varqa o Golšāh). These parallels between book arts and ceramic painting reveal a high degree of cross-pollination between different artistic media in medieval Iran .
The final act proposed in Watson’s tripartite model of pre-Mongol luster painting, the so-called “Kashan” style, was underway by 595/1199 .The characteristics of this phase of production were first proposed by Richard Ettinghausen in his article of 1936, forming part of his powerful argument for top-tier medieval lusterware production at Kashan rather than the previously favored site of Ray. The “Kashan” style of painting exploits the attractive qualities of luster through the use of large figures and registers of repeating patterns painted in reserve on an opaque white ground, with luster coverage of much of the surface relieved by small loops and curlicues scratched through the paint surface . Fine drawing and dense compositions dominate, with fat but elegant birds and frilled, twining half palmettes or other vegetation in reserve often surrounding moon-faced figures who sometimes sit, stand, or ride before a fishpond. Typical forms of this phase include the conical bowl.
Inscriptional bands in rapidly executed cursive script, painted directly in luster or created in reserve by scratching through panels of luster paint, were popularized in this last phase and remained a defining feature of Iranian luster tiles and vessels for many decades. The dense inscriptions found on luster ceramics of both the pre-Mongol and the Il-khanid periods of production encompass all manner of poetic sources as well as Qurʾanic and other holy texts, signatures, and other information. Relationships between the iconography of some of these pieces and the textual content of their copious inscriptions have raised many questions and prompted some complex analysis: Guest and Ettinghausen’s article of 1961, on the subject matter of a luster plate in the Freer Gallery, was a landmark in this respect, while increasingly detailed and nuanced readings of the textual components of individual pieces have added much to study of the medium in recent years (Qučāni, 1992; Pancaroğlu, 2007; idem, 2012; Blair, 2008, pp. 164-69).
The class of polychrome enameled ceramics commonly known as mināʾi ware was apparently created only in pre-Mongol Iran. Abu’l-Qāsem Kāšāni’s ceramic treatise regards “seven-color vessels,” presumed by most scholars to refer to enameled wares, as a long-dead practice by the beginning of the fourteenth century (Abu’l-Qāsem Kāšāni, p. 347, Allan tr., p. 115). Dated wares in this technique, as recorded by Porter and Watson, cover the years 573/1177 to 616/1219 . Thermoluminescence testing of a small selection of enameled fragments supports a date no later than the thirteenth century (Koss et al., p. 33). The small number of dated examples of enameled vessels includes works by the famous Abu Zayd, and there are a few pieces on which both luster and enameled decoration appear, such as a fragmentary star tile in the Museum of Fine Arts, Boston .While it is reasonable to assume that highly skilled potters working in luster and associated with Kashan were part of a luxury ceramic industry that included other very specialized techniques, it should be noted that there is at present little or no evidence to link most enameled wares directly with the site.
At its best the pre-Mongol enameled technique created finely detailed polychromatic designs: blue, turquoise, and purple seem to have been painted over the base glaze after the first application and fixed in the first firing (“in-glaze” painting), while certain other pigments—notably red and black, but also green, pink, brown, yellow, and white (on black)—were fixed over the glaze through a second firing. Gilding also sometimes appears, and the use of gilding on ceramics is recorded in Abu’l-Qāsem’s treatise (e.g., p. 347, Allan tr., pp. 114-15), but it should be noted that this is also one means by which modern agents have been known to “enhance” pieces of enameled ware (for example, see a medieval enameled bowl with recently applied gilding as analyzed in Watson, 2004, pp. 71-72). The technical demands of the enameling process must have been taxing, to say the least. Kerith Koss et al. (p. 46) have noted in their study of enameled sherds in the Freer Gallery that variations in glaze mixtures, order of application, and firing process were observed across the group, indicating that there was not one single fixed technology for this method, a circumstance that is in turn suggestive of multiple ateliers. The colors were painted onto an opaque white, or, more rarely, turquoise glaze that had been fired over a fine stonepaste body, and fixed in second firing .The forms of the vessels can be extremely delicate, with thin walls, and draftsmanship is often crisp. The precise polychromy of enameled wares has a jewel-like appeal that delighted early European and American collectors, while at the same time the overglaze technique has lent itself particularly well to over-zealous restoration and outright forgery. For these reasons many enameled pieces, particularly whole vessels, should be treated with caution. A truly in-depth survey of this type of ware is yet to be undertaken and would ideally include detailed technical analysis of all pieces.
While tiles in the enameled technique exist, the technique is now best known for its use on vessels. A wide variety of forms are known, from pitchers to plates, and many pieces decorated in the enameled technique make use of the medium’s capacity for fine lines and precisely placed contrasting colors to create illustrative compositions. Common scenes found on enameled plates and bowls include richly-dressed figures before a pond or mounted horsemen flanking a central tree, but a far more complex example of narrative enameled decoration is to be found in the three registers of discrete rectilinear illustrative cells that decorate the famous Freer Gallery cup (F1928.2) with the romance of Bižan and Maniža . The largest vessel decorated in this medium, and also one of the best known, is a plate also in the Freer Gallery (F1943.3; Figure 9) depicting a pitched battle and the siege of a castle. Research recently published by Renata Holod proposes a reading for this remarkable piece that reflects Shiʿite Imami sympathies in the depiction of a victory over the Ismaʿilis, sentiments that would not be surprising in medieval Kashan (Holod, pp. 208-11). The schema of the plate is unique amongst known enameled wares in its depiction of what appears to be a contemporary event rather than an idealized scene or spectacle drawn from poetic narrative. Based on historical events, Holod suggests a possible date range of 1210–27 (607-24),which could extend the period of enameled production at Kashan considerably.
Luster tilework in the Il-khanid period. It is tilework, a product that had apparently begun to develop and refine in tandem with the “Kashan” style around 596/1200, whichapparently dominated the output of Iranian luster potters from the second half of the thirteenth century. The chaos of the Mongol conquests seems to have brought production more or less to a halt by the mid-1220s; only a handful of dated tiles survive from the years between 625/1227-28 and 660/1261-22. Production appears to have resumed from the660s/1260s; some dated luster vessels are known from this period, but tiles appear to have been made on a much larger scale. The use of cobalt and sometimes turquoise in combination with luster painting, a technique already developed in pre-Mongol vessels and tilework by potters thought to have been working at Kashan (e.g., the 602/1206 panel at Qom signed by Abu Zayd), remained popular in the Il-khanid period.
Although the painted decoration of the tile assemblages produced in the Il-khanid period can certainly be traced back to the so-called “Kashan” style of luster painting that directly preceded the Mongol invasions, the experimental qualities of the pre-Mongol material were replaced by a model of production that favored rather simpler painting. At the same time, the accelerated impact of Chinese art felt throughout Iran in the Mongol period, particularly from the 1270s, can be seen in vessel shapes and in chinoiserie motifs on vessels and tiles alike. The latter includedragons, phoenixes, cloud-scrolls, and lotus flowers, thought to have their sources in imported textiles and ceramics.
The luster tilework of pre-Mongol and Mongol Iran was apparently intended only for interior use, where the glittering effect of luster in combination with raised inscriptions and molded decoration must have been exceptionally striking by flickering lamplight. Most arresting are the large mehrābs (prayer niche) composed of complex assemblages of luster tiles: several major examples survive whole or in large sections from which the original site, date, and potter are known. One example, now in Berlin, was taken from the Maydān Mosque (Masjed-e Maydān) in Kashan; this work bears the signature of Ḥasan b. ʿArabšāh and is dated 623/1226 .Smaller luster panels and single-tile creations centered on the image of the arched niche, such as the famous “Salting mihrab” in the Victoria and Albert Museum (C.1977-1910), also exist in significant numbers; some of the larger of these were made to cover cenotaphs (Blair, 1986, pp. 393-94). Large quantities of frieze tiles, often with raised inscriptions in cursive script, were created for use above dados, on cenotaphs and around mehrābs, while panels of star-and-cross tiles were used to cover sections of wall. Major tile projects of the second half of the thirteenth century included the tomb chambers of the Emāmzāda Yaḥyā at Varāmin and the Emāmzāda Jaʿfar at Dāmḡān (660s/1260s), and the Taḵt-e Solaymān palace complex (670s/1270s) near Takāb, while the tomb of Shaikh ʿAbd-al-Ṣamad at Naṭanz provided a major commission in 707/1307-8. Some or all of the tiles at these and other sites were ripped from the walls and sold on the international market in the nineteenth and early twentieth centuries, with the result that individual tiles are now held in public and private collections around the world.
Kashan.Fritware with Underglaze Cobalt and Overglaze
Enamel 1218 CE
It is notable that many of the locations in which luster tiles have been found in situ are sites of Shiʿite veneration . When taken in conjunction with the Shiʿite identities of some of the named potters, this raises the interesting possibility of luster tilework as a material manifestation of sectarian identity in medieval Iran, although Blair has suggested that this interpretation should be treated with caution for several reasons .
With the latest dated luster tiles of the Il-khanid type stating that they were made at Kashan in 739/1339, luster production in the town seems to have come to an end around 1340, although Watson notes that a few fifteenth-century and later luster tiles and tombstones of considerably reduced quality have been found in buildings in or near Kashan (Watson, 1975a, pp. 71-73). The reasons for this collapse are not known: luster tiles simply may have ceased to be a fashionable form of architectural decoration as the fourteenth century progressed, and their fall from favor signaled the end of the kilns at Kashan.