Home Page Discussion
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.
by Pierre Galafassi
Italian artists often included rugs in paintings with religious themes (mostly "Virgin and Child enthroned"), more rarely in portraits of cardinals, popes, rich merchants, upper class bureaucrats of city states and of (minor) local dynasts. The carpets were sometimes laid at the feet of the Virgin, but usually decorated a table, chest or wall, according to the actual usage of the time. Rugs were much too expensive during the Italian Renaissance to be routinely walked upon; indeed, they carried price tags similar to those of bronze statues or paintings from top artists (1).
English painters used rugs nearly exclusively in standing portraits of royalty and high ranking nobility. The immensely wealthy Cardinal-Chancellor Wolsey, arguably the greatest rug collector of the time, was never shown with any of his cherished carpets: he died shortly before Hans Holbein launched the fashion in England. This is most unfortunate; how much would you have enjoyed a painted catalogue of Wolsey’s unique collection?
In most cases, in English painting the rug was laid on the floor as a show of power and wealth. Only a few leaders of the Church of England granted themselves the privilege of a rug, but with fitting modesty had it decorating their armchair or table instead. Again, in daily life rugs were much too expensive to be floor coverings.
The painters of the Dutch Golden Age (seventeenth century) used rugs as studio props much more frequently than earlier painters from Renaissance Italy, Spain, Flanders (fourteenth- to early sixteenth centuries), and England (second third of sixteenth century to first half of seventeenth century). As mentioned by Onno Ydema (2), even if we consider only rugs of the Lotto type, there are at least 190 of them featured in extant Dutch paintings (3)!In seventeenth century Holland, rugs were prevalently represented in portraits of rich merchants and officials of civil- or military associations, in "genre" painting, in paintings of interiors of Dutch burger’s homes, in a few paintings with mythologic themes and in a number of still-lives. Given the canons of the Protestant Church, there were hardly any religious paintings and therefore no rugs under the feet of any "enthroned Virgin".
With the notable exception of a few standing portraits of Oranje-Nassau Stadholders, rugs in Dutch paintings were rarely featured on the floor, but nearly always as table covering; showing off too much wealth was not politically correct at the time (again, the Protestant ethic at work) and even though cheaper rugs became available on the market, especially during the second part of the seventeenth century (judging from the low knot-count and limited palette of many rugs used as studio props), one would still not boast of wasting good money by walking on one’s rugs. In fact, rugs started to be used in that "wasteful" way much later, during the eighteenth century, only after they had lost most of their prestige as luxury objects reserved to the moneyed upper classes. This change was probably due to larger and more frequent shipments of cheaper imports from India and Persia (4) and perhaps to the existence of Dutch weaving workshops as well. By then, even the Dutch painters had long lost interest in them.
During the Golden Age, rugs were probably still too expensive for painters to own (5) and were mostly provided by the sitter. It has been argued that the fact that the same rug was sometime featured in paintings of different contemporaneous painters might indicate that the Saint Luke painter’s guilds may have owned a few rugs and rented them out to guild members. A painting by J.De Bray, featuring the governors of the Saint Luke Guild of Harleem could be used as evidence for its ownership of rugs (FIG 144).
Fig. 144: 1675. J. de Bray. Governors of the Guild of Saint Luke in Harleem. Rijksmuseum. Amsterdam.
quite credible possibility is that by the end of the century the rug export
business from Persia or India became large enough for some weavers to "mass-produce" identical rugs featuring popular motifs.
O. Ydema distinguishes the following main categories of rugs in seventeenth century Dutch paintings:
Ottoman rugs from Anatolia, Egypt, Syria: They were nearly always woven in geometric style, but Ydema attributes a few rugs with floral, curvilinear style to the Ottomans. Cairene workshops did produce such types of motifs during this century. Ottoman sultans apparently thought that the ornate Persian style was more fitting than the geometric style to their exalted glory. Accordingly, they also created some workshops working for the court in Anatolia and Istanbul. They were staffed, in part, by displaced Cairene weavers (6) as well as by Persian artisans displaced after Ottoman victories over the Safavid and were using the asymmetric knot. Ottoman rugs in geometric style were frequent in Dutch painting, especially during the early part of the Golden Age, while Ottoman curvilinear floral style was seldom represented by Dutch painters and hardly ever by other European painters. However, such rugs are featured in Ottoman miniatures (FIG 145).
Fig. 145: 1568. Miniature by Nakkas Osman. Sultan Selim II receives the Safavid ambassador. Topkapi Museum, Istanbul
Persian and Indian rugs with curvilinear floral patterns, of which nearly 300 appear in seventeenth century Dutch painting, some as early as 1615. According to Ydema, about 70-75% of these rugs were woven in Persia, the remainder in Moghul India. Both became increasingly frequent during the second half of the century. Not many seem to achieve the standard of upper class Safavid or Moghul production as we know them from extant museum pieces. Many appear to be of a rather ordinary quality with medium knot count and despite their familiar motifs, they frequently feature less "horror vacui syndrom" and a more limited number of colors than generally expected from this origin, as seen for example in FIG 146.
Fig. 146: 1659-1660. F. van Mieris the Elder. Young women in the morning. Hermitage. St Petersburg
Although import of Persian/Indian rugs via Istanbul by land is also documented (7), their relative importance in Dutch homes of the golden age was probably due to the extensive ship lanes of the VOC (The Dutch East India Company) (8) to the spice islands of Indonesia and beyond to China and Japan, via Persian and Indian stopovers and trading posts. Rugs were not listed as one of the most frequent official ship cargoes (spared for the more lucrative spice, silk, chinaware, precious wood or silver VOC businesses). Nevertheless, some large orders are documented (9). In addition, one can assume that senior ship and military officers as well as traveling VOC functionaries would buy them at the Indian or Persian stopovers for the limited private business which the rules of the company allowed them to do or as souvenirs.
Some rather remarkable rugs, which Ydema calls "Scheunemann rugs": Not particularly densely woven, in curvilinear style but with some touches of geometric motifs, and a palette that is neither typically Persian nor Anatolian, they featured rather spectacular floral central medallions and a variety of borders of which some can be met in extant Anatolian, northwest Persian, or south Caucasian rugs. Their weavers did not suffer from "horror-vacui syndrom", unlike too many Persian and Indian rugs (FIG 147).
Fig. 147: 1669. J. de Bray. The de Bray family. Manchester, New Hampshire. Style and palette related to "Scheunemann rugs"
Except for a couple of rather similar, but not identical and perhaps unrelated extant examples which somehow surfaced in Kyoto, "Scheunemann rugs" do not have extant counterpart and their representations are exclusively found in Dutch painting. These rugs were rare before 1650 and became frequent between 1660 and 1690, a total of about 150 pieces being referenced by Ydema (13). This could imply that the rugs were either a recent creation or that their source only became accessible for European trade during this later period. The fact that these rugs were featured only in Dutch painting could support the hypothesis of their weaving either in Holland or near one of the VOC stopovers on the Asian southern shores. None of the mentioned Asian sources completely satisfies the experts, though. At the risk of seeming obsessed by my quest of the elusive Caucasus rug in old masters paintings, I would suggest that those composite influences and a "je-ne -sais-quoi-de-Caucasien" (10) could also mean that the origin was perhaps somewhere near Armenia, South Caucasus or Tabriz, a hypothesis that Ydema does not even consider worth mentioning (:-). C.G. Ellis, who analyzed the Kyoto pieces, hesitated between Dutch and Japanese production, while not fully excluding Persian. Ydema cautiously agrees that the hypothesis of European production could have some merit (11, 12).
A fourth type, also with unclear geographical attribution, was called "Unbekannte Gattung" ("unknown type") by Ydema and tentatively linked to the "Scheunemann" type because of some similarities of the palette (mainly because of a frequent juxtaposition of large orange and red areas, for example) and despite a more anarchic disposition of the motifs. It was characterized by curvilinear floral patterns, a low knot count and a rough design. Ydema mentions that many motifs were again borrowed from rugs from various Asian origin, but somehow look odd to her. Ellis and Ydema, both give a slight preference to India as possible origin of this "unknown type", of which about 80 were identified by Ydema in Dutch painting(14). To my unexperienced eyes, some of these rugs look like low quality Indian weavings with messy motifs, (cheap, made-for-export and for "tourists"?). They also could have been rough Dutch copies of course, although, to me, their anarchic design does not quite fit with the rational and organized mentality for which the Dutch are known.
of Likely Anatolian Origin:
The most frequent Anatolian rugs with
geometric style in seventeenth century Dutch painting was the so-called Lotto
type . A darling of Italian, Spanish and Portuguese painters and their patrons
since about 1525 (15), Lotto rugs were rare in Netherland paintings until
about 1615, then remained in fashion for at least the next 60 years (3). Ellis
popularized three main subtypes of Lotto field style, called "Anatolian" (FIG
148), "Kilim" (FIG 149), and an intermediary style called "ornate" . Ydema
notes that both Anatolian and Kilim styles were present from 1540 onwards
in Netherland paintings, a point that does not support the common
wisdom (on which dating of extant Lotto rugs is in part based) that the former
style was the oldest.
Fig. 148: Turkish. Ushak. Lotto. «Anatolian» field & «kufesque» border. XVI. St Louis.
Fig. 150: Turkish. Ushak Lotto. «Anatolian field & «cartouche» border. XVI.Tappeti classici d’Oriente del XVI e XVII secolo. M Tabibnia.
These rugs came with a number of borders, the most frequent being grouped in three categories, dubbed "kufesque" (FIG 148), "cartouche" (FIG 150) and "cloudband" (FIG 149). The relative rarity in Dutch painting of the "kufesque" border (which was quasi- mandatory in earlier paintings of the Renaissance) is conspicuous (16). On the other hand "cartouche" borders, which were the most frequent in Dutch representation of Lotto rugs and in extant rugs, never appeared in sixteenth century Italian painting, according to Ydema (17). The frequent "cloudband" border is generally considered a marker for a western Anatolian, especially Ushak origin.
Fig. 151: 1615. A. Bloemaert. The four evangelists. Detail. Princeton Art Museum.
Fig. 153: 1618. J. Brueghel the elder. The sense of Touch. Detail. Prado. Madrid.
Fig. 154: 1622. W. J. van den Valckert. Regents of the Groot-Kramergild. Detail. Berlin.
Fig. 155: 1623. G. van Honthorst. The merry fiddler. Detail. Rijksmuseum. Amsterdam.
Fig. 156: 1625-1630. W. Duyster. Playing tric-trac . Detail. Nat. Gal. London
Fig. 157: 1630-1650. Anonymous (Dutch?). Still-life.
Fig. 158: 1650. G. Coques. Family portrait. Detail. Budapest
Fig. 159: 1660-1690. J. van Geel. A Dutch interior.
Ydema states that a fourth border (FIG 160) was specific for a rare and lovely type of blue field Lotto carpets. One of C. de Vos’ patrons probably owned such a blue Lotto, of which the painter represented frustratingly small portions in at least two portraits, including in one of his many portraits of his own large family (in an ostentatiously luxurious and hardly credible setting, by the way).
Fig. 160: Turkish. Ushak.
Lotto. Blue ground. XVIII.Turkish
handwoven carpets. Vol 5.
Fig. 161: 1620 ca. C. de Vos. The painter and his family. Brussels.
The second most frequent Anatolian rug type in Dutch painting of the time was the so-called "Transylvanian" Ushak (FIG 163) or Manisa carpet (FIG 162). They were not represented by earlier European artists, which supports the hypothesis that these rugs appeared on the European market by the end of the sixteenth century, when most European painters and their patrons (except English and Dutch) had already largely lost interest in carpets.
Fig. 162: Turkish. Manisa Transylvanian prayer rug. XVII. Ottoman Turkish carpets of the MAA Budapest. F. Batari.
Fig. 164: 1620. C. de Vos. Abraham Grapheus. Detail. RMFA Antwerp.
Fig. 165: 1627. T. de Keyser. Portrait of C. Huygens. Detail. Nat.Gal. London.
Fig. 166: 1630-1645. P.J. Codde. Dame à sa toilette. Louvre. Paris.
Fig. 167: 1632. T. de Keyser. Portrait of a lady. Berlin.
Fig. 168: 1664. C. Netscher. Lady with a parrot. Detail. Columbus.
Fig. 169: 1670. C. de Man. Interior of townhouse.
Fig. 170: 1675. J. Verkolje. Mother and baby. Louvre. Paris.
Fig. 171: 1678. E. de Witte. Family in an interior. Detail. Munich.
Anatolian "Medallion Ushak" rugs (FIG 172) were somewhat less frequently represented in paintings. Perhaps they were less popular in Dutch homes because their usual large dimensions weren't adaptable to the Dutch usage as table cover.
Fig. 172: Turkish. Ushak medallion . XVI XVII.
Fig. 173: 1646. G. van Honthorst. Amalia van Solms-Nassau and daughters. Detail. The Hague.
Fig. 174: 1656. J. Vermeer. The Procuress. Detail. Dresden.
Fig. 175: 1667-1668. G. Ter Borch. Woman playing the theorbo. National Gallery London.Browsing through Dutch painting one occasionally meets other "Anatolian" rugs. Most are quite plausible, but some are rather strange. T. Rombouts, for example, committed a couple of carpets, apparently related to each other, that are illustrated below. Not only the musician in FIG 176 but the rug as well looks a little Spanish to me.
Fig. 176: 1620-1625. T. Rombouts. Lute player. Philadelphia.
Fig. 177: 1620-1625. T. Rombouts. Card game. Hermitage.
G. Dou (FIG 178) shows a rather faded "Bellini" rug that could have been already quite old at the time of the portrait and P. de Hooch shows a carpet (FIG 179) featuring what looks like a variation of the large Holbein field motif.
Fig. 178: 1667. G. Dou. Young woman at her toilet. Detail. Boijmans van Beunigen Mus. Rotterdam.
Fig. 179: 1663-1665. P. De Hooch. Merry company. MNAA Lisbon.
floral rugs of Persian or Indian origin
The Moghul rulers staffed their first rug workshops with imported Persian weavers. Distinguishing seventeenth century Persian from Indian carpets is, therefore, a task for real experts. I would rather not have a serious try at it and leave it to the Turkotek specialists. It is only my wildest guess (protected by the first amendment) that the rugs featured in FIG 180 to 190 might be Persian (or, maybe not).
Fig. 180: 1616-1618. P.P. Rubens. Portrait of Nicolas de Respaigne. Kassel.
Fig. 181: 1630. T. de Keyser. Portrait of a silversmith.
Fig. 182: 1631. T. de Keyser. A gentleman. Mauritshuis. The Hague.
Fig. 183: 1642. P.J. Codde. Detail of a familly group. Rijksmuseum. Amsterdam.
Fig. 184: 1654. G. Ter Borch. Messenger. Hermitage. St Petersburg.
Fig. 185: 1655. A. van den Hecken. Portrait of C. J. Meyer. Rijksmuseum. Amsterdam.
Fig. 186: 1660. F. van Mieris the Elder. The cloth shop. Vienna.
Fig. 187: 1663. J. de Bray. Regents of the of the orphanage. Haarlem.
Fig. 188: 1668. P.de Hooch. Couple with parrot. Detail. Köln.
Fig. 189: 1670-1674. P. de Hooch. Card players.
The rugs in FIG 191 to 198
have a rather Indian look to me, but don’t bet the house cat on it.
Fig. 191: 1656. C. Fabritius. Wilhelm van der Helm. Detail. Rijksmuseum. Amsterdam.
Fig. 192: 1666. E. van der Neer. Portrait of man and woman in interior. detail. Boston.
Fig. 193: 1669. C. Netscher. Lady playing the guitar. Detail. Wallace Collection.
Fig. 195: 1670. J. van Rossum. Portrait of J. W. Verbrugge and wife. Detail. Rotterdam.
Fig. 197: 1700-1720 W. van Mieris. Dutchman and wife. Detail. Leamington S.A.M.
As already mentioned, "Scheunemann rugs" were
woven with medium or even low knot density. Most motifs were drawn in
curvilinear, floral style, but some others, especially in the border, were
inspired by the geometric style and are vaguely reminiscent of northwestern Persia
or southern Caucasus. The fields are less crowded than in most Persian or
Indian classics. The field palette shows a particular liking for a medium blue,
a rather strong orange and a red, the latter two often being adjacent. The
type was obviously rather popular in Dutch burger’s homes and did not lack prestige,
since even prominent members of the Oranje-Nassau ruling family wanted to be
painted with such a rug underfoot. Mmm, orange shade / Oranje family: A
subliminal message perhaps? Does it advance in any way the theory of a Dutch
origin of the rugs?
Fig. 199: 1652. B. van der Helst. Mary of Orange. Detail. Rijksmuseum. Amsterdam.
Fig. 200: 1655. G. Ter Borch. Woman washing hands. Dresden.
Fig. 201: 1658. G. Ter Borch. Woman playing the theorbo and cavalier. Detail. MET. New York.
Fig. 202: 1660. G. Ter Borch. Lady at her toilette. Detail. Detroit.
Fig. 203: 1667. P. de Hooch. Lady with a cittern. Detail. Taft museum. Cincinnati.
Fig. 204: 1665-1670. J. Steen. Esther before Ahasuerus. Detail. Hermitage. St Petersburg.
Fig. 205: 1672. F. van Mieris the Elder. François Sylvius and wife. Detail. Dresden.
Fig. 206: 1670-1675. J. Ochtervelt. Young lady trimming her
fingernails. Detail. National Gallery. London
This essay is the last of the series, since the eighteenth and nineteenth centuries were, from a ruggies' point of view, as bare and dry as the Takla Makan. Carpets were rarely featured in painting during these centuries and even when one was shown, it was,more likely than not degraded to an indistinct splash of color, fully bastardized, or worse still, it was an awful wall-to-wall product of a European carpet factory.
There was a notable
exception though: nineteenth century Orientalist painting, to which
Filiberto Boncompagni dedicated an outstanding Salon in
2004-2005, which triggered a huge number of interesting
posts. These are at http://www.turkotek.com/salon_00105/salon.html
and its follow-up http://www.turkotek.com/misc_00024/orientalist.htm I thank my accomplices, Steve Price and Filiberto Boncompagni, for their help in preparing these essays.
(1) Comparative money value of oriental rugs and top painter’s / sculptor’s jobs during the Renaissance are discussed here. http://www.turkotek.com/VB37/showthread.php?t=1371
(2) O. Ydema has written what is surely the best and most exhaustive book on rugs in Dutch painting to date (many thanks again to Patricia Jansma for her tip!). Most information in this essay is borrowed from her outstanding work: Carpets and Their Datings in Netherlandish Paintings. Antique Collector’s Club. I agree with Patricia, it is, indeed, a must in any ruggie’s library.
(3) O. Ydema, page 29
(4) Checking the import documents of Dutch ships active during the 1600’s, Jonathan Johnson found little mention of rug cargo. One can suppose that the imports were initially mainly made on a small scale by individuals.
(5) Even in the inventory of a successful painter like Vermeer, who commanded relatively high prices for his work and was comparatively well-to-do during most of his life, there was neither a rug nor any of the other luxury items like virginals (small harpsicord), tapestry or silver trays which appear so frequently in his paintings. (Sources: "Essential Vermeer" http://www.essentialvermeer.com/inventory.html and
Although a few Dutch painters, especially C. de Vos, represented their families or themselves in a richly furnished environment, with rugs, silverware musical instruments etc.., one can suspect that these objects belonged to their patrons.
(6) O. Ydema, page 20 . After the 1517 defeat of the Mamluk to the Ottomans and the annexation of their empire, the Sultans kept the Cairene weaving workshops active. (Ndr. It is likely that most extant so-called "Mamluk" rugs were woven long after the hanging of the last Mamluk ruler). From the second part of the sixteenth century onwards, part of the Cairene production was inspired by contemporaneous Persian curvilinear floral designs and, according to Mrs Ydema, used the asymmetric knot, perhaps because it was better suited than the symmetric knot for this kind of job. The same thing happened at about the same time in royal workshops in Anatolia and Istanbul. Ydema mentions that around 1580, Sultan Murad III displaced Cairene weavers to Anatolian workshops to help develop the production of these more prestigious carpets, which we can see represented in several Ottoman miniatures.
(7) O. Idema, pages 124-125
(8) The Dutch East India Company (VOC) eclipsed all of its rivals in the Asia trade. Between 1610 and 1798 the VOC sent almost a million Europeans to work in the Asia trade on 4,785 ships (about twice as many as English ships), and netted for their effort more than 2.5 million tons of Asian trade goods. (source, Wikipedia)
(9) O. Ydema, pages 78-79: Around 1625 the VOC started importing some carpets from India. In 1624, for example, they ordered from their trading post in Suratte (Northwestern India) 540 high quality Persian rugs and, conditionally, 300 additional India-made ones "should they be of comparable quality and lower cost than the Persian production".
(10) By the standards of Rugdom Science a "je-ne-sais-quoi" usually qualifies as robust proof (:-)).
(11) O. Ydema, page 104.
(12) At the time, rug-weaving workshops were indeed active in several European countries including northern Netherlands, Flanders, France, England, Italy, Poland and, of course, in Spain, where this activity was important and dating back at least half a millennium. It is also documented that emigrated Flemish weavers started a rug weaving activity in England.
(13) O. Ydema, pages 106, 120
(14) O. Ydema, page 120
(15) Arguably, the first lotto rug appeared in Gregorio Lopes’ "Salome" in 1525. http://www.turkotek.com/old_masters/salon_3.html
(16) Which has been tentatively explained by a ban imposed by Ottoman religious authorities during the sixteenth century, of the export to non-believers of rugs featuring motifs with symbolic religious significance, as is perhaps the case with kufic borders in rugs. Such a ban has been suggested in carpet literature but seems purely speculative.
Idema, pages 33-34