OTTOMAN TURKEY, 1475-1485

Lot essay written by Carlo Maria Suriano

This previously unrecorded Ushak Medallion carpet joins the small number of other examples in which the grand decorative pattern features a pseudo-kufic calligraphic frieze along the main border. Now seven in total, these carpets belong to the earliest phase of Ushak Medallion production which, as I shall discuss here, covered a period of some fifty years. While they have generally been given an undifferentiated 15th century date, it is clear that these seven carpets represent not a single production moment, but a rapidly evolving period of design evolution covering perhaps the first half-century in the story of early Ottoman court art.

Of the now seven known Medallion Ushak carpets of this group, two others feature precisely the same pseudo-kufic border as the present example. A fourth complete example shows slight variations in the border, along with stylistic evidence suggesting it was made after the previous three.

Three more complete Ushak Medallion carpets make up this early production group. These show only a partial or fragmented version of the pseudo-kufic frieze, reflecting rapid changes in court taste as luxuriant floral elements increasingly replaced the original calligraphic pattern.

1. The present example; 2. Czartoryski-Altounian-Wher; 3. Castellani-Bruschettini; 4. Maciet-Musée du Louvre, Paris, inv.no. AD 14 428/UCAD 14 428; 5. Chevalier-Musée du Louvre, Paris, inv.no. MAO 959; 6. Grassi-Thyssen-Bornemisza; 7. Stroganoff-Museum für Kunst und Gewerbe, Hamburg, inv.no.1958.43.

The first four Ushak Medallion carpets listed here feature a red ground while the last three have a blue ground.

The Ushak Medallion design is one of the expressions of a new artistic programme first conceived under Court directives during the last years of the reign of Mehmed II Fatih (1432-1481). The discovery of a previously unrecorded example of the early Ushak Medallion production, published here for the first time, is therefore a significant moment, prompting reappraisal of the whole group with a view to establishing a weaving chronology. Stylistic evidence, detailed below, suggests that the newly discovered example stands alongside the Wher and Bruschettini carpets as the three earliest known Ushak Medallion carpets ever produced. They can be ascribed to around 1475-1485, a period that includes the last years of reign of Mehmet II Fatih (1432-1481) and the very early years of reign of his son, Bayazid II (1481-1512).

The Maciet and the Chevalier carpets at the Musée du Louvre would appear to have been made around 1500 to 1510 (the last decade of the reign of Bayazid II), while the last two carpets would appear to have been made some time in the following decade, thus during the short reign of Bayazid’s son, Selim I.

THE CREATION OF USHAK MEDALLION CARPETS

There is evidence that the formation of an Ottoman royal scriptorium or kitabhane under the patronage of Mehmed the Conqueror was responsible for the creation of the first court style (J. Raby, ‘Court Patronage and Design. Birth of a Court Style’, pp. 76-77, in Iznik. The Pottery of Ottoman Turkey. N. Atasoy and J. Raby, London 1989). A leading figure in the royal scriptorium was ‘Baba Nakkaş’ whose celebrated album illustrates decorative designs related to the bindings and illumination of manuscripts dedicated to Mehmed II (known as Mehmed Fatih - known in the west as the Conqueror). This decorative programme, which Julian Raby labelled as the Rumi-Hatayi style, is characterised by a unique combination of arabesque decoration (Rumi) combined with floral (Hatayi) elements of Chinese origin. Raby speaks of an “Ottoman style (that) was denser, more compact and forceful than the somewhat languorous Timurid version”. These major artistic changes occurred during the 1460s and 1470s and they informed all minor arts.

It has long been agreed that the original design of the Ushak Medallion carpets powerfully embodies the creative impulse that inspired the first Ottoman court style (J. Raby, ‘Court and Export: Part 2. The Ushak Carpets’, in R. Pinner & W.B. Denny (eds.), Oriental Carpet & Textile Studies II. Carpets of the Mediterranean Countries 1400-1600. London 1986, pp.177-187). In practice, designs provided by the royal scriptorium would be implemented by specialist workshops producing specific media, although in the case of carpets it would have taken a specialist to transfer the designs to full-scale weaving cartoons. The role of the cartoon master in bringing these carpets to production may provide the answer to a riddle which I will discuss later in relation to the last two blue-ground Ushak Medallion carpets listed above.

Available evidence suggests that the genesis of Ushak Medallion carpet production can be dated to the last years in the reign of Mehmed Fatih (d.1481). In 1473, following his defeat of the Aqqoyunlus, Mehmed had relocated a number of artists from Tabriz to Istanbul. The marked increase in Turcoman influence during the latter part of his reign makes clear that Iranian scribes soon came to dominate the production of royal manuscripts. One scribe in particular is noteworthy here: Ghyyath al-Din al-Isfahani, who signed himself ‘al-Mujallid’ (the bookbinder), appears in several colophons. As discussed in an earlier article (C.M. Suriano, ‘Oak Leaves and Arabesques’, HALI 116, pp.106-115, May-June 2001) a close parallel can be traced between a new type of floral decoration seen in the pendants of one of Ghiyath’s undated manuscript bindings and an almost identical floral programme in the pendants of the earliest extant Ushak Medallion carpets. This evidence suggests a date between 1475 to 1485 for the first Ushak Medallion carpets, a period which corresponds to the last years of Mehmed II Fatih and the earliest regnal years of his son and successor Bayazid II. The decorative pattern devised for these early carpets proved to be so successful that it continued almost undisturbed through to the second half of the 16th century and well into the early 17th century. But the nature of the production changed. Around 1540 the Ushak Medallion pattern ceased to dominate Ottoman courtly taste as new carpet designs appeared, and by the second half of the reign of Suleyman the Magnificent (1520-1566), production of these carpets was directed at a well-established general market.

In discussing the earliest phase of Ushak Medallion carpet production we are therefore looking at those extant early examples conceived and produced exclusively for the Ottoman court of Mehmed II Fatih or for his son Bayazid II. The overall quality of design and detail places these very few carpets as among the finest examples of the type ever produced. As noted, this group is characterised by a shared pseudo-kufic main border not found in later Ushak Medallion production.

Of these I consider the newly discovered carpet, together with the one formerly in the Wher collection in Lugano and the one in the Bruschettini Foundation in Genova (both illustrated above), to be the three earliest (1475-1485). The pseudo-Kufic pattern (see below border details) on these three carpets consists of two alternating elements: an endless knot, formed by a pair of half arabesque polychrome palmettes on a continuous baseline and terminating at the top in two vegetal ‘flag-poles’, and a lotus palmette, terminating at the top in two more ‘flag-poles’. In some ways this design reflects the taste of late 15th century Turkish carpets, notably the Small Pattern Holbein rugs, with their strong geometric pseudo-kufic friezes. As discussed elsewhere, there is nevertheless a substantial difference between the way the frieze is used in these two contrasting contexts: “While the decorative intention of the 15th century Turkish rug border is to create an essentially calligraphic effect, the frieze that characterises this group of Ushak carpets reveals a quite different idea. The artist has taken a single element - the endless knot - from an older tradition, and has then gone on to contextualise it within a much more ambitious and mature ornamental programme” (Suriano 2001).

The pseudo-kufic border in these earliest Ushak Medallion carpets is not experimental, as has been suggested, but part of a new fully realised visual programme created by the leading masters who worked in the kitabhane for Mehmed II Fatih and were responsible for the creation of the grandiose Ushak Medallion design.

The original version of the elegant pseudo-kufic border is found across the three examples we have assigned to the first decade of production (1475-1485), but arguably the design has a clarity, graphic precision and overall balance in the newly discovered carpet that is not quite matched in the other two examples. For instance the spacing between the end of the ‘flag-pole’ elements and the beginning of the inner border in the present example enables the frieze to breathe, while in the Wher and Bruschettini carpets the ‘flag-pole’ elements push against the minor border. I would suggest that these individual variations correspond to the hand of the specialist whose task it was to transfer designs provided by the royal scriptorium onto full-scale weaving cartoons.

Kurt Erdmann was the first to point out that the concept behind the new Ushak Medallion design is that of an infinitely repeating ground of circular and polylobed (so-called secondary) medallions. (K. Erdmann, The Early Turkish Carpet, p.70, London 1977). Effectively it would be the artist in the royal scriptorium who would select a portion of this ‘endless repeat’ and turn it into a fully realised pattern, frozen in time by the borders around it. Here matters of balance and proportion become key to the success or otherwise of the end result. According to these criteria the design solution proposed for the present example emerges as the grandest and best balanced of the group. By including a much wider section of the ‘secondary medallions’ than most examples and thus including the greater part of their pendants, the artist creates substantially more space around the ‘central medallion’, creating a strikingly elegant and spacious layout. The layout of the Wher example is also based on a wider than average section of the design, but due to the unusual width to length ratio it fails to achieve the same monumental thrust. It nevertheless has an essential grandeur as does the third of the earliest examples, the Bruschettini carpet. In time this powerful assertive quality will come to be replaced by the lighter more elongated format seen in the remaining four carpets of our group of seven, in which the central medallion becomes the dominating feature.

Looking for a moment at the central medallion and remaining with the first three carpets in our group, both the present example and the Wher carpet feature a round medallion of great elegance. The central medallion in the Bruschettini, by contrast, shows the ogival form that will go on to become the standard in 16th and 17th century Medallion Ushak production. In layout too the Bruschettini proposes a design solution that will become the preferred variant in later production. Kurt Erdmann suggested misleadingly that the ‘simple’ arrangement (as seen in the Bruschettini) was a useful feature in dating carpets: “from the material I know, all the Ushak carpets with a simple medallion arrangement are a later development of the classical pattern in which the medallions are arranged in an endless repeat” (Erdmann 1977, loc.cit.). There is plenty of evidence starting from our three very early examples and continuing right through 16th and 17th century production, to show that the two variant layouts coexisted. Having come into being at the same time they both remained in the design repertoire over the centuries. (One can think here of the 16th century grand fragmentary examples in Istanbul: one on blue ground in the TIEM, inv. no. 77, and the other on red ground in the Vakiflar Museum, inv. no. A142, as well as the late 16th or early 17th century grand white-ground Ushak Medallion carpet at the Metropolitan Museum of Art, New York, inv. no.1984.69).

Ultimately however it is design quality rather than issues of chronology that determine the aesthetic balance and impact of any individual carpet.

USHAK MEDALLION CARPETS DURING THE REIGN OF BAYAZID II

It was under Mehmed II Fatih’s son, Bayazid II, that the new court style entered its second stylistic phase. This is clearly evidenced in the production of Iznik ceramics. It is not my intention to try and establish a direct stylistic link between the ateliers producing Iznik pottery and Ushak carpets. The scenario is rather, as discussed above, one in which the artists of the imperial scriptorium would create patterns for court approval and these in turn, once selected for use, could enter the repertoire of each individual atelier and appear in its production.

An example of such a shared source can be seen in the Louvre’s Maciet Ushak Medallion carpet with pseudo-Kufic border. It contains a feature not found in the other six carpets discussed here: in place of the oak leaf floral field ornamentation it has a pattern of dots. This is a feature associated solely with a production that makes its appearance in Iznik pottery between 1500-1512, part of the new visual programme introduced under the patronage of Bayazid II. Evidence can be seen in the Iznik footed basin with dotted ground in the Victoria and Albert Museum, London (inv.no.C.1981-1910) or the hemispherical footed basin with dotted ground in the Musée du Louvre (inv.no. 7880-92). With its plethora of polychrome pseudo-floral dots filling the space between the endless knot and the palmette elements along one end of the frieze (see border details), the Maciet carpet evidently came into being in exactly the same period (1500-1512). It is unquestionably a very elegant, successful example with a well designed ogival central medallion whose shape emphasises the overall elongated effect of the design. While the pseudo-Kufic border retains the original elements seen in its three predecessors, the ‘flag-pole’ elements on the lotus palmette now have less of a calligraphic and more of a budlike quality. Compared to the first version of the pseudo-Kufic frieze the border design appears less muscular here, creating overall a softer and more floral effect.

The movement away from the first stylistic phase is even more evident in the Chevalier Ushak Medallion carpet, also in the collections of the Louvre. This carpet, which features a prominent and well controlled ogival central medallion, is the first of our seven examples to feature a blue ground. Compared to the Maciet frieze, the pseudo-kufic border in the Chevalier moves a step further away from the original concept (see border details below). While the floral stylisation of the reduced ‘flag-pole’ elements on the lotus palmette continues, the endless knot has disappeared altogether. In its place the Chevalier Ushak features a lotus palmette while the ‘flag-pole’ elements at the top have now evolved into floral tendrils.

The move towards a more accentuated floral taste in these two carpets, albeit proposing ‘bud-like’ rather than fully blossomed floral elements, is in keeping with the second Ottoman court phase sponsored under Bayazid II. Both the Maciet and the Chevalier carpets are extremely elegant and accomplished, showing clear evidence of newly created designs produced by the artists of the kitabhane. Court artists were now striving for new design solutions as official taste moved away from 15th century aesthetics towards a taste that will eventually reflect 16th century Ottoman floral opulence.

Overall it has become clear by now that while the field design of the large Medallion Ushaks is remarkably stable, it is the borders of these carpets that may serve as indicators of the arrival of new ideas and changes in fashion.

USHAK MEDALLION CARPETS DURING THE REIGN OF SELIM I

In spite of continuous military campaigns, court patronage continued during the reign of Selim I (1512-1520). Julian Raby mentions the commission for hanging lamps made in Iznik ordered by Selim I around 1512-13 for the tomb of his father Sultan Bayazid II (Raby 1989, p.100). The dominant motifs are large, stylised lotus blossoms on the bodies of the lamps. This is the style, as Raby suggests, that defines the 1510s and heralds that of the 1520s. It is this new floral device that leads Raby to speak of “the workshop of the Master of the Lotuses”. We are now in the third stylistic phase of Ottoman court style and I would propose that the two remaining carpets of the pseudo-Kufic border group belong to this stage. Both these carpets have a blue ground, a feature that cannot go unnoticed. These two, together with the blue-ground Chevalier carpet, suggest that it was under the patronage of Bayazid II that the blue variant of this design first appeared, to carry on with considerable success through the period of Selim I and on into the first two decades of the reign of Suleyman the Magnificent (1520-1566).

The first one is the more intriguing of the two. This is the Grassi carpet now in the Thyssen-Bornemisza collection. Here we find, at one end only of the main border, the pseudo-Kufic frieze. What we see along the three remaining sides of the main border is a fully formed repeating design of large stylised lotus blossoms of a type that is in keeping with the contemporary work of the ‘Master of the Lotuses’ seen on Iznik pottery.

The lotus blossom border in this carpet points forward to the border design of Ushak Medallion carpets of the second quarter of the 16th century such as the famous example in the Victoria and Albert Museum, London, inv.no. T.71-1914, or the one in Philadelphia, inv.no. 1955-65-15. But what of this strange, unique combination within the same carpet of the older pseudo-kufic frieze together with the latest decorative fashion? There is no doubt that the Thyssen-Bornemisza carpet is the expression of a transitional stage between old and new, as has long been suggested. However can we be sure that this carpet, like the five other carpets discussed above, was made for the Ottoman court?

Indeed, the use of two different patterns – the Kufic frieze and the lotus blossom - in the main border of the Thyssen carpet is at odds with the coherent layout of the composition as found in the previous five examples. If we then look at the pseudo-Kufic frieze along the top end of the carpet, we again find differences relative to the original version. The vegetal tendrils that spring out at each side of the endless knot lack the clarity that guided them in the earlier version of the design. The top (flagpole) element is missing, although this may be due to the rewoven outer band. By contrast, the lotus blossoms along the other three sides of the main border are sharp and incisive, suggesting that this newly fashionable element was perhaps more central to the designer’s concerns than the Kufic frieze. Interestingly the same disparities we notice in the pseudo-Kufic frieze can be equally observed in the drawing of the ogival central medallion, which places it at odds with a royal commission.

Let’s now look at the Ushak Medallion carpet in the Museum für Kunst und Gewerbe in Hamburg. This carpet, like the Thyssen example, is on a blue ground but, unlike the Thyssen carpet, it features the pseudo-kufic frieze along all four sides of the main border (see border detail above). Interestingly when we compare the Hamburg frieze with the Thyssen pseudo-kufic frieze (see detail below), we find a very similar rendition of the design. The endless knots are again clumsy, lacking the graphic definition of the earlier version. Equally the vegetal tendrils that spring out at each side of the endless knot look tentative and confused. The top part of the original pseudo-kufic design, involving the ‘flag-pole’ element at the top of the endless knot and the palmette, is unfortunately missing as a result of clear loss. Nevertheless there is enough here to enable us to identify a very close analogy in design between the friezes in these two last carpets.

It might be worth adding that the Hamburg carpet shows a very irregular drawing of the Ushak Medallion layout. The ogival central medallion has an invasive presence which stifles the overall composition, while the section of the ‘endless repeat’ defined by the borders looks random: the pendants at the top of the secondary medallions in the upper half of the carpet are almost entirely missing, as opposed to the arrangement in the lower half of the carpet. Consequently the section of the second central medallion at the top is considerably smaller than its equivalent at the bottom. The end result delivers a rather unbalanced composition which again seems to be at odds with what the artistic refinement of the imperial scriptorium of the time would have been able to deliver.

Perhaps the answer to the incongruity noted in relation to these two carpets is not so difficult to find. As Julian Raby writes in relation to the Iznik production: “The decade of the 1510s itself saw continuing ceramic production, the most important aspect of which was the broadening of the market base. Popularization was reflected in several factors. One was wider distribution, both geographically and socially….. A more popular market would have been encouraged, in the first place, by an increase in disposable incomes, and there must have been many Ottomans who benefitted, overnight as it were, from Selim’s conquests. It would have been encouraged, by reduced production costs….There was scope to cut labour costs, by being less fastidious in execution…. and, secondly, by altering the character of the decoration” (Raby 1989, p.98).

Raby’s account of the broadening of the market base for Iznik ceramics can surely be applied to Ushak Medallion carpet production. I would suggest that the Thyssen and Hamburg carpets are representative of this earliest stage of market production in which carpets were still primarily aimed at a local market, as opposed to a subsequent phase of production aimed at the flourishing export market of the second half of the 16th century.

We can now finally turn to the observation we made earlier in discussing the implementation of designs from the royal scriptorium in media-specific workshops. It seems clear that in the case of carpets specialist input was needed to transfer the designs to full-scale weaving cartoons. I would suggest that it was precisely this last figure, the designer of the cartoons, within the chain of production of these carpets, who may account for the lower quality of these last two carpets. The design was no longer provided by the artists of the imperial scriptorium. As commissions were enacted at a lower social level it was the designer of the cartoons, and not the artists of the imperial scriptorium, who became responsible for the Ushak Medallion market production. In the case of the Thyssen and Hamburg carpets the cartoon designer played with both old and new ideas on a purely decorative level. Here there are no intellectual artistic forces at play to conjure up a programmatic visual statement. Rather the opposite: it is decoration for the market.

If what we propose here is valid then a correction must now be made. It remains true that the small group of Ushak Medallion carpets featuring a pseudo-kufic border is representative of the earliest Ushak Medallion court carpet production. However only five of them can unquestionably be linked with court patronage. Three of them (the Christie's carpet, the Wher carpet and the Bruschettini) can be linked with the direct patronage of Mehmed II Fatih and, as such, they therefore reflect the first stage of the Ottoman court style, while two others (the Maciet and the Chevalier carpets at the Louvre), can be linked with the direct patronage of Bayazid II and, as such, they represent the second stage of the Ottoman court style. The last two examples (the Thyssen-Bornemisza and the Hamburg carpets) are representative of arguably the earliest stage of market production with no further existing link to court patronage.