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: Part 2 - Geometric Rugs in Early Renaissance
(14th and 15th Century)  Paintings.

by Pierre Galafassi

Animal rugs were sometimes used as studio props for early Renaissance paintings (14th and 15th century) but were not the main type of rug at the time.   Most carpets featured in paintings of the period have purely geometric field motifs, which are often unfamiliar to us and even puzzle the very few scholars who have studied them.  In many cases there is no extant rug featuring these patterns.  It is generally agreed that these rugs were mostly woven in the Islamic world, in the large crescent from Anatolia and Egypt to Maghreb and Al-Andalus (southern Spain).   However, scholars are rarely more specific than that.

I have made a feeble attempt at dividing these rugs based on their motifs.

A first group are very small rugs featured in a fairly large number of early paintings, mainly of the schools of Florence, Siena, Venice and Ferrara, often depicted as decorations for windowsills and gondolas during festivals (1). John Mills (2) thinks that their “geometric designs look like highly simplified versions of Turkish rugs of the second half of the (15th) century. Whether they were in fact early versions of these or were locally made in imitation of them cannot be certainly decided, but the latter does seem the more likely”.

As a possible location for the workshops, a few southern Italian cities, like Lucera (Puglia region) have been tentatively mentioned by some sources, mainly owing to the fact that part of their population was Muslims expelled from Sicily by Frederik II Hohenstaufen. Mills’ theory of Italian copies is credible of course, but one could just as well suspect Muslim populations of Sicily, Spain (as suggested by Giovanni Curatola (3)) or Anatolia, (Walter. B. Denny (4)), to have woven them.

FIG 28 
A. di Giovanni 1440-1460.

Cassone ( marriage chest) panel with scenes of tournament.
National Gallery London

FIG 29 
A. Baldovinetti.Madonna and Child.
Uffizi. Florence.

FIG 30

A.  da Fabriano. 1474.
Virgin and Child.


FIG 31 
G. del Cossa. 1476-1484.
Allegory of April, detail.
Palazzo Schifanoia. Ferrara.


FIG 32 
Gh. di Giovanni. 1488.
St Gerome in his study.

FIG 33

G. del Biondo. 1370.
Annunciation. Detail.
 Kress Collection.


FIG 34 
Mansueti. 1494.
Miracle of the Holy Cross in campo San Lio.
Accademia. Venice

FIG 35 & FIG 36 
V. Carpaccio. 1495.
Two details from the «Cycle of St Ursula, the departure of the pilgrims»
Accademia. Venice.


FIG 37
Master of Darmstadt Passion. 1455.

FIG 38
B. D’Eyck. 1460.
Livre du coeur d’amour espris.

A second group of rugs features geometrical patterns which, contrary to most of those shown above, do not strike me as being particularly “Turkish”. Their inspiration could just as well be European, Armenian or Byzantine, for example. Onno Ydema (5) states that the undulating “curvilinear stem with trefoil leaves (in the border) is almost certainly derived from Western Gothic ornament”. The rugs are evidently more densely knotted than the previous group, feature a normal border and come in various sizes similar to those in later classical rugs. About ten different rugs of this type can be seen in J. van Eyck’s, G. David’s and Petrus Christus’ paintings.  The strong similarities of the motifs make a common origin very likely.  To my limited knowledge, no rug with such motifs is extant; thus, we have little reason to expect that a structural analysis could give us clues to their real origin. 

FIG 39
J. van Eyck. 1436.
The Lucca Madonna.

FIG 40 
Petrus Christus.1457
Virgin and Child.

FIG 41
J. van Eyck. 1436.
Virgin, Child & Canon Van der  Paele.
Nat. Gal. of Art Washington.
A nearly identical rug is shown in G. David’s «St Anne altarpiece» painted 60 years later.

A third group of early paintings features rugs that can be clearly linked to classical Anatolian patterns which were successfully copied and expanded over the next few centuries.  Especially so-called “small pattern Holbein”, “large pattern Holbein”, “Para-Mamluk”, “Bellini” and “Ghirlandaio” rugs. Many were used as studio props in paintings, and even a few 15th century rugs are extant.

 Four of the oldest representations of “small pattern Holbein rugs” are found in paintings by A. Mantegna, B. Baró, S. Botticelli and T. del Trombetto.

FIG 42

A. Mantegna. 1457-1459.
Virgin and Child.
San Zeno altarpiece. Verona (Detail below)

FIG 43
A. Mantegna. 1457-1459.
Virgin and Child. Detail.
San Zeno altarpiece. Verona

FIG 44
B.  Baró 1450-1470
Virgen de la leche.
Ademuz. Spain.

FIG 45
S. Botticelli. 1460.
F.  de Montefeltre and Landino.

FIG 46 and FIG 47
Anatolian rugs with «small Holbein» pattern. 15th century.                                        
In «Anatolian Carpets», W. B Denny. (FIG 46)
&  in «Vakiflar Museum. Teppiche», B. Balpinar & U. Hirsch (FIG 47)  


FIG 48 shows, arguably, the first representation of a Ghirlandaio pattern:

FIG 48
D. Ghirlandaio. 1486.
Virgin and Child.
Uffizi. Florence

The first “para-mamluk” rugs were probably painted by F. Foppa and by Giovanni da Udine.


FIG 49
V. Foppa. 1485.
Virgin and Child. Detail.
Brera. Milan


FIG 50
Unknown painter. 15th century

The eponymous ancestor of the famous “Bellini” rug of the MIK in Berlin (FIG 52) appears under the throne of the Virgin in a 1470 painting (FIG 51).
There are few extant “sober Bellini rugs”, none with this border (6), but the single- and double “keyhole” (or “re-entrant”) pattern remained a favorite of Turkish weavers for another several centuries (for example, FIG 53). 

FIG 51
Gentile Bellini 1470.
Virgin and Child.
National Gallery, London

FIG 52
Ushak prayer rug with «Bellini» pattern.
16th century
Museum für Islamische Kunst. Berlin.

FIG 53
Anatolian prayer rug.
15th-16th century

The so-called “large Holbein” pattern might have been first featured in a well known Crivelli painting, FIG 54. According to Mills, the origin of the pattern might be in northwestern Anatolia (7) 

FIG 54
C. Crivelli. 1456
 Annunciation with St Emidius.
National Gallery. London.

The rug in Crivelli’s painting looks like a smaller, simplified version of the roughly contemporaneous extant rug in FIG 55, given as 14th or 15th century. There is also a clear pattern analogy with the 15-16th century southern Spanish rug in FIG 56 and several other extant Spanish carpets.

FIG 55
Anatolia. 14th-15th century.
Vakiflar Museum, Teppiche, B. Balkinar & U.Hirsch.

FIG 56
Murcia. 15th century
Large Holbein motif.
MET. New York.

This leads us to a fourth group of 14th and 15th century rugs, featured in Spanish paintings and quite possibly woven in southern Spain.  It is well documented that rugs have been produced since the Muslim conquest in this region, mainly in Alcaraz, Letur, Chinchilla, Cuenca, Alpujara and Murcia, and that production continued long after the Christian Reconquista was completed in 1494, and even after a large part of the Muslim population was forced to emigrate (8).

Several Moslem writers praised the outstanding quality of Al-Andalus rugs, for example Ibn Hawkal (10th century) and Al-Idrissi (12th century). Al-Saqundi (12th century) boasts that these rugs were “exported to all countries of Orient and Occident”, Cairo being a particularly good customer. When Leonor de Castilla married the future King Edward I of England in 1254, the young bride had a main street of London decorated with her many Spanish rugs (8).
Nearly three centuries later, Henry VIII’s chancellor, Cardinal Wolsey, still appreciated and collected Spanish rugs (we will say more about this fanatical ruggie in our next essay.)
Spanish weavers used a specific knot, shared only with some Egyptian (coptic?) weavers.  Thus, the attributions of extant Spanish rugs are reasonably certain despite the strong analogy of many motifs with those of Anatolia. 

FIG 57
J.  Huguet. 1460.
St Vincent ordained by St Valerius
MNAC. Barcelona.

FIG 58
J.Huguet. 1466-1475.
Consecration of St Augustin.
MNAC. Barcelona.


FIG 59
V. Macip. 1480-1500.
Christ resuscitated.
Valencia Cathedral

FIG 60
Master of the Sisla. ca. 1500.
Prado. Madrid.

Note the curvilinear pattern inscribed in hexagons (Renaissance influence?) and the narrow (mudejar?) kufic border of FIG 60.

FIG 61
B. Martorell. 1437-1442.
St. Peter’s altar.
Girona. Spain.

Many rugs shown in fourteenth- and fifteenth century paintings feature unique patterns which fully defeat my occidental urge for classifying all and everything. Unless I err, most of these orphan carpets do not even have any extant descendants, poor lonely chaps.
One cannot exclude that a few rugs were the fruit of the painters imagination, but most look quite credible, at least to my uneducated eyes. The scholar, R.A. Mack states that “The Italian elite who commissioned paintings showing their costly carpets undoubtedly wanted them represented in details” (9).

FIG 62
Giotto di Bondone.
St Francis & Honorius III
Assisi Basilica.

FIG 63
A.    Lorenzetti. 1340.
The «Small Maesta». Detail.

Quiz for our many Turkotek rug experts:  Is the rug in FIG 63 a kilim? A rug made sewing together bands of flat weaves like tent bands?
The field of the “Arnolfini rug” (FIG 64) appears to be woven in parallel bands featuring the same four motifs on a surmey background. Mills does not exclude the possibility that it could be a local needlework (10).  The “Sedano” rug (FIG 66) seems to be a variation of the “large Holbein” pattern.


FIG 64
J. van Eyck. 1434
Portrait of G. Arnolfini and his wife. Detail                  
National Gallery. London.


FIG 65
The Osma Master. 1450.
St Ann, Virgin & Child. Detail
MET. New York.


FIG 66
G.  David 1490-1495
The Sedano Virgin. Detail.
Louvre. Paris

FIG 67
Qu. Massys 1495
Virgin & Child. Detail.
National Gallery. London

Massy’s rug (FIG 67) particularly puzzles John Mills (11): “...the elaborate border of kufic plus endless knot resembles known Turkish borders of the period, but the exactly contrived solution to the problem of turning the corners is most unusual, the field design is not a recognizable one and the curvilinear border accords ill with the otherwise starkly geometrical border...”

I find the rug in the French miniature (FIG 68) quite interesting too, with its sober two-tone indigo field much reminiscent of some earlier Seldjuk rugs except for the fact that the small repeating motif here is a cross.

FIG. 68
 Limburg brothers.1409.
 Les Riches Heures du Duc de Berry.
 Chantilly. France

My puzzle-o-meter hits its highest marks with FIG 69, a painting by A. del Castagno, and with FIG 70, an Il-Khanid (12) miniature from early 14th century with its field of repeating endless knots and sophisticated borders, including one kufic.

FIG 69   
A.  del Castagno. 1445.Virgin & Child.
The Pazzi Madonna
Uffizi. Florence

FIG 70
Folio of one of the earliest and most celebrated illustrated manuscripts of Firdawsi’s Shahnama.
About 1330-1340 representing Alexander on his bejeweled coffin surrounded by mourners.
This picture was already shown in an earlier discussion in which I dumbly misquoted the date.
Freer Gallery. Smithsonian.

You may have noticed the scandalous absence of Memling in this essay. Indeed this rug-obsessed artist was also very active during the last quarter of the fifteenth century, but abusing our authority as authors of this essay, we opted to tell you soon «a tale of two ruggies. 1475-1540» starring Memling and Holbein.
Sources and notes.

(1) Rosamond E. Mack, Bazaar to Piazza, page 78
(2) John Mills, Carpets in Paintings, 1983, pages 11-14.
(3) Giovanni Curatola, in Venise et l’Orient, Tissus et tapis à Venise, page 210. Gallimard. 2006.
(4) Walter. B. Denny, in «Venise et l’Orient», Textiles et tapis d’orient à Venise, page 178. Gallimard. 2006.
(5) Onno Ydema, Carpets in Netherlandish paintings, page 9. (*)
(6) John Mills, carpets in Paintings, 1983, page 20.
(7) John Mills, carpets in Paintings, 1983, page 21.
(8) Alberto Bartolomé Arraiza, Alfombras Españolas, MNAD, page 21-22.
(9) Rosamond E. Mack, Bazaar to Piazza, page 78
(10) John Mills, carpets in Paintings, 1983, pages 11-14.
(11) John Mills, carpets in Paintings, 1983, pages 15-18.
(12) Il-Khanid: Mongol dynasty ruling greater Persia between 1256 and 1335, reporting to the Great Khan.
(*) Many thanks to Patricia Jansma who suggested  buying Onno Ydema’s  outstanding «Carpets and their dating in Netherlandish paintings»

Discussion Home Page

Introductory Essay

Part 1: Animal Rugs in Renaissance Paintings

Part 3: A Tale of Three Renaissance Ruggies

Part 4: Rugs in Sixteenth- and Seventeenth Century English Paintings

Part 5 - Rugs in Paintings of the Dutch Golden Age