Home Page  Discussion

Salon du Tapis d'Orient

The Salon du Tapis d'Orient is a moderated discussion group in the manner of the 19th century salon devoted to oriental rugs and textiles and all aspects of their appreciation. Please include your full name and e-mail address in your posting.

Rugs and Old Masters: Addendum to part 5 - Quest for the elusive Caucasian rugs in Renaissance paintings.

by Pierre Galafassi 

In articles and books about Rugs in Renaissance Paintings, at least in those known to me, experts nearly always attribute the carpets with geometric motifs to Turkey, Syria or Egypt.
There is hardly any mention of other, perfecly credible places of origin like Spain (1), Caucasus and Armenia, Timurid Persia (3), or Turkmen Qara-Koyunlu’s Empire, including at the time the Tabriz area (4).

As Yohann Gissinger repeatedly argued in Turkotek posts, leaving Spain out of the picture is particularly illogical, since there is documented evidence enough that the Kingdom kept producing rugs, silk textiles, artifacts and architecture in Oriental style (Mudejar art), during the whole Reconquista and even long after its end and exported them. There are also enough reasonable clues that painters did introduce Spanish rugs in their works. Besides, as Filiberto Boncompagni mentioned, Italy, but also Flanders and Germany, had strong political- and economical links with Spain at the time, which makes it quite obvious that their elites (the painter’s patrons) knew Spanish rugs and were most likely to own some of them too (5), next to Ottoman- or Cairene ones.
 I would like to concentrate in this thread on another, probably unfairly neglected potential source of «studio props» for Renaissance painters and enjoyment for their patrons: Caucasus, including Armenia. Not only to lift Filiberto’s spirits, assuming there would be any need for that.

One of the most numerous populations of the Caucasus area, the Armenians, were already praised by medieval Arab visitors for their talent as dyers and rug weavers. Their great competence as international merchants is beyond doubt too. They are documented as playing a relevant role in the import of rugs to Venice, (at the time the main rug emporium in Europe), by sea and by land (via Ottoman Transylvania). Why would these active merchants and rug connoisseurs have refrained from shipping rugs made by their own weavers, or by their close Eastern- (Including Tabriz, Akh Koyunlu Turkmen etc..) and Northern- (Circassians, Avars, Georgians  etc..) neighbors ?

Besides, not far from the Caucasus (FIG 1), Kaffa and  Tana (northern shore of the Black Sea) and Trebizond (southern shore) were antique and important ports of call for Venetian-, Pisan- and Genoan ships, until the Ottoman restricted the direct access for European business, after their conquest of Constantinople (1453), thus certainly reducing the flow of Caucasian or Persian goods, but not necessarily canceling it completely, nor permanently.


FIG 1. Venetian commercial route to the Black Sea.

 While Venice and the Ottoman Empire were indeed frequently at war, during their long periods of truce they still engaged in eager commercial exchange of luxury artifacts. Venetian artists, including Gentile Bellini (1479-1481) visited and worked for the Sultan.

FIG 2.  Gentile Bellini. Portrait of Sultan Mehmet II. National Gallery, London.


The voyages, via Tana, of Venice’s ambassadors Zeno (1472) Contarini (1473) and Barbaro (1487), to meet the Aq Koyunlu ruler in Tabriz and the Safavid ruler in Sultanye, also show that even ambassadors notoriously trying to create military alliances hostile to the Ottomans were able to make the trip quite safely. 

Florence was mostly at peace with the Ottomans, like for example during Lorenzo de’ Medici’s rule (who is documented as ordering rugs from Istanbul and Cairo) and had probably no major difficulty at sourcing also the odd Caucasian- or Persian product in Ottoman ports.

In fact, Caucasus rugs (called for example «Circassians») are quite frequently mentioned (6) in inventories of rich Renaissance collectors. In at least one documented case they were even the dominant «brand» in the collection ( the collector was surely an ancestor of yours, Filiberto):

According to R.E. Mack the 1584 inventory of a member of the Correr family of Venice, lists «..table carpets from Cairo, Turkey, Persia, and Circassia or the Caucasus and more Circassian carpets for benches and chests...» (8) and Marin Sanudo, the gossipy secretary to the Venetian Senate «..confirms that government buildings around the (San Marco) square were lavishly decorated with carpets... For example, for the feast of the Madonna, September 8, 1515, Cairene, Circassian (Caucasian) and other carpets ... were hung across the facade of the Doge’s Palace....» (9).

Thus, I believe that there is a good probability that Caucasian rugs were present (incognito) in Italian Renaissance paintings too.

However, anyone looking for credible Caucasian rugs in old Master’s Painting has a limited numbers of possible clues to look for, as we have no possibility of structural analysis and must rely only on motifs analogies (10)

Rug patterns and motifs have often survived many centuries, therefore one could make the hypothesis that if a design found in a rug painted during the Renaissance is still very frequent in, say, extant seventeenth to nineteenth century Caucasian rugs, but much less frequent or never met in extant rugs from other geographical origins, this could constitute a reasonable clue for a Caucasian origin of the painted rug.  
In previous essays and discussions, we met three cases, which might fall under this rule of thumb:

a) The leaf and wineglass border, which was discussed here:

This «leaf and wineglass motif is met in extant rugs attributed to Anatolia too, from the fifteenth century onward (one fragment found in Konya on Alaeddin’s tomb), but was much less frequent there than in extant southern Caucasian Carpets, including the oldest ones (seventeenth and eighteenth century). One finds this border especially in rugs attributed to so-called Kazak, Gendje, Kuba (Karadashli)- and Karabagh weavers (11).

FIG 3 Gendje. XVIII. 165X112. U. Schurmann, Caucasian rug


FIG 4 Karabagh. XVII. 298X167. Sarre


FIG 5. Kazak, blossom motif. XVII. 194X151. F. Nagel cat. # 286

The hypothesis that rugs represented in Renaissance painting, and featuring both geometric field motifs and «leaf and wineglass» borders, were woven in the Caucasus area, does not seem to me  more harebrained than usual for Rugdom standards. Some rugs even featured a third potential Caucasian marker, the «dragon» motif, as secondary border. (discussed below, point c).

Thus I move that the following rugs, (FIG 6,7,8) painted between 1519 and 1545, carried a «Circassian» ID (12) and not an Anatolian one.



1519. J. C. van Oostanen. Mary Magdalene. St Louis
(Note the secondary «dragon» border too)

FIG 7 1537-1557. Unknown painter. After Holbein. Henry VIII Detail. Petworth House


FIG 8 1545 ca. Anonymous. Portrait of a nobleman. LACMA. Los Angeles
(Note the secondary «dragon» border too)

b) Field motifs with large «sixteen points stars», discussed here


As discussed in that thread, there is a clear analogy between the main medallion motif featured in so-called Crivelli rugs, illustrated in a few Renaissance paintings (FIG 9,10) and those of extant eighteenth and nineteenth century southeastern Caucasian soumaks and rugs, often tentatively attributed to Shemaka, in the southern Shirvan area (FIG 11, 12). A level of analogy which imho is matched by no other origin, except a couple of extant fifteenth century rugs generally attributed to Anatolia (FIG 13,14) but wether this Anatolian attribution is based on solid ground or not I don’t know and decide to ignore it (:-)). Please, on FIG 13, note the «dragon border» too.

FIG 9. 1486. C. Crivelli. Detail of Annunciation with St Emidius. National Gallery. London.

FIG 10 1482. C. Crivelli. Annunciation. Detail. Frankfurt

FIG 11 Southeastern Caucasus (Shemaka?) 6 fragment Crivelli pattern, soumak techn. XVIII XIX

FIG 12 Southeastern Caucasus (Shemaka?) 11 Crivelli pattern, soumak techn. XIX. 186X132. E. Concaro &A. Levi. Sovereign Carpets

FIG 13 Anatolia (?)  Crivelli pattern. XV
(Note the «dragon» main border too!)

FIG 14 Anatolia (?) Computer reconstitution of a fifteenth century Crivelli motif. W. Kirchheim, Orient Stars.

c) «Dragon-» or «s»- border motif (12).

 Another, perhaps less convincing, but still possible marker for a Caucasian or Tabriz-area origin could be the very ancient «dragon» border motif, already met in rugs featured in fourteenth century Il-khanid miniatures painted by the Tabriz kitabkhana (FIG 15 ) and frequent in extant eighteenth- and nineteenth century southern Caucasian rugs (FIG 16, 17, 18 ).

 The «dragon» border motif is also quite often found in the border of painted Renaissance rugs (FIG 19, 20, 21,22), and in at least two cases (FIG 6,8), next to a «leaf-and-wineglass border».

 In fairness, one should add that the motif was used as well in some other weaving regions, including Anatolia and even Spain. One can only argue, that it is notably less frequent in the latter areas than in the Caucasus.

FIG 15 1330-1340. Il-khanid period. The bier of Iskander, detail . Tabriz school. Freer-Sackler

FIG 16 Caucasian. Gendje, soumak techn. XIX. 5'10"X4'5". D. Dodds & M.L. Eiland. Oriental Rugs from Atlantic Collections.

FIG 17. Caucasian. Gendje (?), star motif. XIX. 260X126. W. Kirchheim, Orient Stars

FIG 18 Caucasian. Kazak. XVIII.  D. Eder, Orientteppiche, Band 1, Kaukasische Teppiche.

FIG 19 1460 ca. Filippo Lippi. Funeral of St Stephen. Detail. Duomo. Prato.

FIG 20 1475. H. Memling. Donne Triptych. Detail. National Gallery, London

FIG 21 1480. H. Memling. Virgin and Child. Detail. Berlin

FIG 22 1486. C. Crivelli. Annunciation with St Emidius, detail. National Gallery.

d) So-called «Scheunemann» rugs

In the essay #5 on rugs of the Dutch Golden Age (here 
http://www.turkotek.com/old_masters/salon_5.html  and here 

http://www.turkotek.com/VB37/showthread.php?t=2228  ) we discussed the possibility that so-called «Scheunemann» rugs (FIG 23), which have been illustrated exclusively in Dutch painting and of which there is no extant example, could have been woven in the Caucasus area. Although this theory has it’s drawbacks and is not mentioned by any serious expert yet, there are interesting design analogies, mainly the bold palette, the large field surfaces with no trace of horror-vacui, the self-assured motifs and some border motifs very frequent in Caucasus rugs (Mainly the «spade» and the «dragon»).

FIG 23 1669. J. de Bray. The de Bray family. Manchester New Hampshire.



I am painfully aware of the fragility of my hypothesis based exclusively on motif analogies between rugs separated by at least two or three centuries. I also plead guilty for superbly ignoring the fact that much has happened during this period in a region confronted with major changes in the ethnic distribution, including massive forced emigration, immigration and bloody ethnic confrontation, which must have had a large negative influence on rug weaving as well.


(1)    It would seem logical that especially Portuguese- and Spanish painters (or rather their rich patrons) would have owned at least some samples of the local (Mudejar) rug production. There are strong clues that this indeed happened. For example, one can  suspect that some geometric rugs with strong dominance of a saturated yellow field (2) and rugs with «scorpion» border-motifs, identified as frequent in extant Spanish rugs, were more likely to be made by Mudejar- than by Anatolian weavers. Several yellow- field rugs painted in Bruges by H. Memling and G. David, might be concerned, see here:


(2)    Cristina Partearroyo Lacaba, in "Alfombras Espańolas" mentions (page 13) that Cuenca’s Mudejar weavers copied many Anatolian motifs, but in their own specific palette, in which (saturated) yellow and blue dominated, while red dominated (with blue and green as secondary shades), in contemporaneous Anatolian rugs. Browsing through representations of rugs in Spanish paintings, one gets indeed the feeling that the dominance of a saturated yellow might be a marker for some Spanish Renaissance rugs.

(3)   The probable origin of the «small Holbein» motif in Persia and its great popularity during the Timurid period was discussed here:


(4)   The «para-mamluk» rugs have been attributed to various Middle East places (Damas, Cairo, Anatolia, northwestern Persia,..) or ethnic groups (in particular the Akh-Koyunlu and Qara-Koyunlu), with, as usual, scant evidences (if any). IMHO, the most attractive theory, backed by some interesting facts, proposes an attribution to the Qara Koyunlu, an Oghuz Turkmen conglomerate, whose (alas nondescript-) rugs were highly praised by the Venetian G. Barbaro (ambassador to the Tabriz court of the Qara-Koyunlu ruler ), as being «superior to Ottoman and Syrian rugs». Since Barbaro noticed this «superiority», the Qara Koyunlu rugs might have shown a particularly striking pattern and / or a much different palette from Anatolian ones. Given the Oghuz origin of the Qara Koyunlu (shared with «our» modern Turkmen and by the Seljuk-, Akh Koyunlu- and Ottoman dynasties), one could even phantasize that these rugs featured some characteristics of Turkmen- or Seljuk rugs, like the passion of the former for saturated red shades and spectacular «guls» or the strict field geometry and repeating motifs typical of both origins.

(5)   The possibility that rugs with the «Large Holbein» motif were mainly an expression of Mudejar art and not merely copied from Anatolian prototypes (perhaps even the other way around) was discussed here:


(6)   While «Circassian» is a rather precise designation of origin, many names found in old rug inventories can be misleading: one can never be sure whether the name relates to the emporium where the rug was bought, to the style of the rug or to the place where it was woven. A «Damascene rug» for example does not necessarily imply that it was woven around Damas, it might have been bought in this important Bazaar, woven elsewhere, feature any kind of style or could be meant to describe a particular rug pattern. Forty rugs from the inventory of Henry VIII went under the name «carpets of Venice» (7). As there is no evidence of any rug weaving in Venice, it is much more likely that the rugs came from the shops of Venetian merchants, but were woven somewhere in Orient or Spain. Alternatively, some might have come, from small and temporary Italian weaving workshops, such as those created by the Courts of Ferrara or Mantova.

(7)     R.E. Mack. Bazaar to Piazza, page 202, citing Donald King. «The inventories of the carpets of King Henry VIII», Hali 5, # 3, 1982

(8)   R. E. Mack. Bazaar to Piazza, page 200, citing Thornton page 64-65.

(9)   R. E. Mack. Bazaar to Piazza, page 78, citing the painfully precise Venetian civil servant Marin Sanudo, in Rogers 1986, page 18.

(10)       Marla Malett has rightly warned Turkotekers looking for rugs’ origins, that design and motifs, can be easily copied and travel quickly, thus, they are less reliable indicators of provenance than a set of structural parameters.

(11)         The nomenclature of Caucasian rugs, based on U. Schurmann’s superficial work is widely criticized, but, alas, is still used by just about everybody. For a more credible nomenclature one can rather take inspiration from Kerimov. His work is largely cited in «Azebaidjanisch-kaukasische Teppiche» by S. U. Azadi, L. Kerimov & W. Zollinger. 

(12)       «Circassian» is not to be taken literally of course (see also note 6). The Caucasus, including in the centuries considered, has always been a vast mosaic of ethnics, remains of vanquished people and of their settled victors. Many people had an antique tradition of rug weaving, Azeris ( and other Turkik tribes), Armenians, Avars, Kurds, Lezgins,Turkmen etc..

FIG 24. Caucasus: Ethnolinguistic groups (map reference lost)

(13)   «Dragon or S Border" : The info about these usual name of the motif is courtesy of Filiberto Boncompagni, quoting Peter F. Stone's "Tribal & Village Rugs - The Definitive Guide to Design Pattern and Motif"  "(page 172, C-102)

Discussion  Home Page

Part 1: Animals Rugs in Renaissance Paintings

Part 2: Geometric Rugs in Early Renaissance Paintings

Part 3: A Tale of Three Renaissance Ruggies

Part 4:  Rugs in 16th and 17th Century English Paintings